diff --git a/README.md b/README.md
index b7484a1..4948d67 100644
--- a/README.md
+++ b/README.md
@@ -1,5 +1,7 @@
 # sskj — 多平台大模型推理性能基准测试项目
 
+**更新（2026-08-27 13:53:26 CST）**：完成 Kimi-K3 八节点标准 PD 第一阶段。P 组 601-604 使用 PP8×TP4×EP4、FlashInfer MXFP4、Chunk 8K，D 组 605-608 使用 PP1×TP32×EP32、Marlin，通过 Mooncake 0.3.12.post1 和 4 Rail RDMA 传输；统一 P/D `page_size=64` 后，16K→1 与 16K→512 的 C1/C8 共 91/91 请求成功。代表结果：16K→1 C8 Input TPS 6364.31、TTFT P50/P95 20.555/21.345 秒；16K→512 C8 TPOT P50/P95 63.20/66.55 ms。详见 `experiments/pro6000/kimi3_pro6000_pd_pp8_standard/README.md`。
+
 **更新（2026-08-20 10:13:01 CST）**：完成 Kimi-K3 Prefill TP Reduce Scatter 可行性审计并停止该方向。K3 的 MLA 输出门控仍依赖完整 7168 维 hidden，且 69/93 层为 KDA；保留或恢复 gate hidden 后，原 hidden All-Reduce 无法消除并新增 2112 维 latent All-Gather，估算通信量反增约 14.7%。研究原型仅保留为否决证据，不进入四节点实验或上游 PR；后续转向 MoE A2A 与 Pipeline Parallelism。详见 `experiments/pro6000/kimi3_pro6000_sglang_tp_reduce_scatter_prefill/README.md`。
 
 **更新（2026-08-19 16:51:43 CST）**：完成 Kimi-K3 / RTX 6000D / SGLang Prefill MoE backend 交付文档收敛与 Chunk=4K 补测。四节点 TP32×EP4、16K→1、C=8/16 下 12/12 重复、480/480 请求成功；FlashInfer 相对 Marlin 提升约 9.8% Input TPS、降低约 9.0% TTFT。结合 4K/8K/16K 全部结果，最终推荐保持 `flashinfer_mxfp4 + chunked_prefill_size=8192`。
@@ -73,6 +75,7 @@
 | `experiments/pro6000/kimi3_pro6000_sglang_tp32ep32/` | RTX 6000D + SGLang，Kimi-K3（TP32×EP32，部署手册见 docs/KIMI_K3_DEPLOY.md） |
 | `experiments/pro6000/kimi3_pro6000_sglang_tp32ep32_moe_backend_prefill/` | RTX 6000D + SGLang，Kimi-K3 真实 Prefill 的 Marlin/FlashInfer MXFP4 与 Chunk 8K/16K 对照 |
 | `experiments/pro6000/kimi3_pro6000_sglang_tp_reduce_scatter_prefill/` | Kimi-K3 Prefill TP Reduce Scatter 可行性审计；因 MLA gate 与 KDA 依赖完整 hidden，结论为不推荐继续 |
+| `experiments/pro6000/kimi3_pro6000_pd_pp8_standard/` | Kimi-K3 八节点标准 PD：P=PP8/TP4/EP4、D=PP1/TP32/EP32、Mooncake 4 Rail RDMA，含 16K→1/512 C1/C8 原始证据 |
 | `experiments/pro6000/kimi3_pro6000_pd_rdma/` | RTX 6000D + SGLang，Kimi-K3 PD 分离（MoonCake RDMA，8 节点，见 deploy_pd.sh + docs/KIMI_K3_DEPLOY.md 附录 B） |
 
 每个目录内：`run_bench.sh` 跑固定并发矩阵；`run_adaptive_concurrency.sh` 从 C=1 指数倍增搜饱和点；`run_adaptive_concurrency_add16.sh` 从 C=16 线性 +16 步进、带 TTFT SLO 停止与回退（当前主力用法，见 `experiments/ADAPTIVE_CONCURRENCY_USAGE.md`）。
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/Dockerfile.pd b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/Dockerfile.pd
new file mode 100644
index 0000000..d520879
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/Dockerfile.pd
@@ -0,0 +1,18 @@
+ARG BASE_IMAGE=local/sglang:kimi-k3-sm120-flashinfer-mxfp4-phase5
+FROM ${BASE_IMAGE}
+
+COPY patches/31869.patch /tmp/pd-patches/31869.patch
+COPY patches/32797.patch /tmp/pd-patches/32797.patch
+COPY mooncake.whl /tmp/mooncake_transfer_engine_cuda13-0.3.12.post1-cp312-cp312-manylinux_2_28_x86_64.whl
+
+RUN cd /sgl-workspace/sglang \
+    && patch -p1 --forward < /tmp/pd-patches/31869.patch \
+    && patch -p1 --forward < /tmp/pd-patches/32797.patch \
+    && python3 -m pip install --no-deps --force-reinstall \
+       /tmp/mooncake_transfer_engine_cuda13-0.3.12.post1-cp312-cp312-manylinux_2_28_x86_64.whl \
+    && python3 -m compileall -q \
+       python/sglang/srt/disaggregation \
+       python/sglang/srt/managers/scheduler.py \
+       python/sglang/srt/managers/scheduler_pp_mixin.py \
+    && rm -rf /tmp/pd-patches \
+       /tmp/mooncake_transfer_engine_cuda13-0.3.12.post1-cp312-cp312-manylinux_2_28_x86_64.whl
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/README.md b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/README.md
new file mode 100644
index 0000000..eefa482
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/README.md
@@ -0,0 +1,78 @@
+# Kimi-K3 Standard PD on 8x RTX 6000D Nodes
+
+This experiment validates the minimum standard PD path before enabling any
+speculative decoding:
+
+- Prefill: nodes 601-604, PP8/TP4/EP4, FlashInfer MXFP4, chunk size 8K.
+- Decode: nodes 605-608, PP1/TP32/EP32, Marlin, speculative decoding disabled.
+- Transfer: Mooncake 0.3.12.post1 over mlx5_0..3 RDMA.
+- Router: SMG MiniLB on node 605, port 31000.
+
+The PD image adds only upstream SGLang PRs #31869 and #32797 plus Mooncake
+0.3.12.post1 on top of the validated phase5 image.
+
+## Files
+
+- `config.env`: topology and fixed P/D parameters.
+- `Dockerfile.pd`: reproducible PD image delta.
+- `deploy_pd.sh`: the only service lifecycle entrypoint.
+- `run_pd_validation.sh`: standard four-case suite plus 16K-to-512 C1/C8 modes.
+- `patches/`: unmodified upstream SGLang patches.
+
+P and D must use the same KV page size. This experiment fixes both sides to
+`page_size=64`; a mismatch is rejected by the decode server before transfer.
+
+## Run
+
+```bash
+cd /data/hzy/sskj/experiments/pro6000/kimi3_pro6000_pd_pp8_standard
+
+bash deploy_pd.sh build-image
+bash deploy_pd.sh preflight
+
+RUN_ID=kimi3-pd-standard-$(date +%Y%m%d-%H%M%S) \
+  bash deploy_pd.sh start
+
+RUN_ID=kimi3-pd-validation-$(date +%Y%m%d-%H%M%S) \
+  bash run_pd_validation.sh
+
+RUN_ID=kimi3-pd-decode512-c1-$(date +%Y%m%d-%H%M%S) \
+  bash run_pd_validation.sh decode512
+
+RUN_ID=kimi3-pd-decode512-c8-$(date +%Y%m%d-%H%M%S) \
+  bash run_pd_validation.sh decode512-c8
+
+bash deploy_pd.sh status
+bash deploy_pd.sh stop
+```
+
+The sudo password is read from `/data/hzy/.sudo_password` by default and is
+never stored in Git.
+
+## Validated Result
+
+Validated on 2026-08-27 with 601-604 as P and 605-608 as D. All 91 requests
+completed with the requested token counts, and the final service-log scans
+contained no fatal pattern.
+
+| Case | Requests | Input TPS | TTFT P50 / P95 | TPOT P50 / P95 |
+|---|---:|---:|---:|---:|
+| 64 -> 8, C1 | 1 | 42.18 | 1.217 / 1.217 s | 33.27 / 33.27 ms |
+| 1K -> 16, C1, non-streaming smoke | 1 | 508.09 | 1.956 / 1.956 s | N/A |
+| 16K -> 1, C1 | 8 | 2161.23 | 7.565 / 7.614 s | N/A |
+| 16K -> 1, C8 | 40 | 6364.31 | 20.555 / 21.345 s | N/A |
+| 16K -> 512, C1 | 1 | 595.27 | 7.702 / 7.702 s | 38.66 / 38.66 ms |
+| 16K -> 512, C8 | 40 | 2544.27 | 18.755 / 23.182 s | 63.20 / 66.55 ms |
+
+Raw results:
+
+- `results/kimi3-pd-validation-final2-20260827-1310/`
+- `results/kimi3-pd-decode512-20260827-1325/`
+- `results/kimi3-pd-decode512-c8-20260827-1340/`
+
+The first request after service startup spent 4.775 seconds in TTFT while
+Mooncake established RDMA endpoints. The next two 64-token smoke runs were
+stable at 1.193 and 1.217 seconds. P logs record TP4-to-TP32 Mamba state-slice
+transfer, and D logs record RDMA-ready acknowledgements, confirming the real
+P-to-D path. PP8 completed without the fixed-delay/deadlock symptom, so the
+optional PP1 control was not required.
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/config.env b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/config.env
new file mode 100644
index 0000000..301891a
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/config.env
@@ -0,0 +1,43 @@
+EXPERIMENT="kimi3_pro6000_pd_pp8_standard"
+
+BASE_IMAGE="local/sglang:kimi-k3-sm120-flashinfer-mxfp4-phase5"
+PD_IMAGE="local/sglang:kimi-k3-sm120-flashinfer-mxfp4-pd1"
+MODEL_PATH="/data/hf_models/Kimi-K3"
+SERVED_MODEL_NAME="kimi-k3"
+MOONCAKE_WHEEL="/data/flashkda_deploy/wheels/mooncake_transfer_engine_cuda13-0.3.12.post1-cp312-cp312-manylinux_2_28_x86_64.whl"
+
+SSH_USER="user"
+SSH_OPTIONS=(-o BatchMode=yes -o StrictHostKeyChecking=no -o ConnectTimeout=10)
+P_NODES=(174.1.60.1 174.1.60.2 174.1.60.3 174.1.60.4)
+D_NODES=(174.1.60.5 174.1.60.6 174.1.60.7 174.1.60.8)
+ALL_NODES=("${P_NODES[@]}" "${D_NODES[@]}")
+
+P_HEAD="174.1.60.1"
+D_HEAD="174.1.60.5"
+MC_MASTER="174.1.60.1:50051"
+P_DIST_PORT=20000
+D_DIST_PORT=20000
+P_PORT=30000
+D_PORT=30000
+P_BOOTSTRAP_PORT=28800
+ROUTER_HOST="174.1.60.5"
+ROUTER_PORT=31000
+
+P_TP=4
+P_PP=8
+P_EP=4
+P_CHUNKED_PREFILL_SIZE=8192
+P_PAGE_SIZE=64
+P_MEM_FRACTION_STATIC=0.88
+P_MAMBA_FULL_MEMORY_RATIO=0.36
+
+D_TP=32
+D_PP=1
+D_EP=32
+D_PAGE_SIZE=64
+D_MEM_FRACTION_STATIC=0.88
+D_MAMBA_FULL_MEMORY_RATIO=0.21
+
+HEALTH_WAIT_S=2400
+DISAGG_IB_DEVICES="mlx5_0,mlx5_1,mlx5_2,mlx5_3"
+NCCL_IB_HCAS="mlx5_0,mlx5_1,mlx5_2,mlx5_3"
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/deploy_pd.sh b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/deploy_pd.sh
new file mode 100755
index 0000000..fc7835c
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/deploy_pd.sh
@@ -0,0 +1,389 @@
+#!/usr/bin/env bash
+# Deploy Kimi-K3 standard PD on 601-608. Speculative decoding stays disabled.
+set -Eeuo pipefail
+
+SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
+# shellcheck source=/dev/null
+source "${SCRIPT_DIR}/config.env"
+
+ACTION="${1:-status}"
+RUN_ID="${RUN_ID:-kimi3-pd-standard-$(date '+%Y%m%d-%H%M%S')}"
+RESULT_ROOT="${RESULT_ROOT:-${SCRIPT_DIR}/results/${RUN_ID}}"
+SUDO_PASSWORD_FILE="${SUDO_PASSWORD_FILE:-/data/hzy/.sudo_password}"
+
+P_CONTAINER_PREFIX="${EXPERIMENT}_prefill"
+D_CONTAINER_PREFIX="${EXPERIMENT}_decode"
+MC_CONTAINER="${EXPERIMENT}_mc_master"
+ROUTER_CONTAINER="${EXPERIMENT}_router"
+
+mkdir -p "${RESULT_ROOT}"/{commands,logs,metadata,bench}
+
+log() {
+  printf '[%s] %s\n' "$(date '+%F %T')" "$*" | tee -a "${RESULT_ROOT}/orchestrator.log"
+}
+
+require_password() {
+  if [[ -z "${SUDO_PASSWORD:-}" ]]; then
+    [[ -r "${SUDO_PASSWORD_FILE}" ]] || {
+      echo "ERROR: set SUDO_PASSWORD or create ${SUDO_PASSWORD_FILE}" >&2
+      exit 2
+    }
+    IFS= read -r SUDO_PASSWORD <"${SUDO_PASSWORD_FILE}"
+  fi
+}
+
+is_local() {
+  [[ "$1" == "${P_HEAD}" ]]
+}
+
+remote() {
+  local host="$1"
+  shift
+  if is_local "$host"; then
+    "$@"
+    return
+  fi
+  local command
+  printf -v command '%q ' "$@"
+  ssh "${SSH_OPTIONS[@]}" "${SSH_USER}@${host}" "${command}"
+}
+
+sudo_host() {
+  local host="$1"
+  shift
+  require_password
+  if is_local "$host"; then
+    printf '%s\n' "${SUDO_PASSWORD}" | sudo -S -p '' -- "$@"
+    return
+  fi
+  local command
+  printf -v command '%q ' "$@"
+  printf '%s\n' "${SUDO_PASSWORD}" | ssh "${SSH_OPTIONS[@]}" \
+    "${SSH_USER}@${host}" "sudo -S -p '' -- ${command}"
+}
+
+node_suffix() {
+  printf '%s' "${1##*.}"
+}
+
+container_name() {
+  local group="$1" rank="$2"
+  if [[ "$group" == "p" ]]; then
+    printf '%s_node%s' "${P_CONTAINER_PREFIX}" "$rank"
+  else
+    printf '%s_node%s' "${D_CONTAINER_PREFIX}" "$rank"
+  fi
+}
+
+build_image_on_host() {
+  local host="$1" suffix build_dir
+  suffix="$(node_suffix "$host")"
+  build_dir="/tmp/${EXPERIMENT}_build"
+  log "staging PD image build context on node ${suffix}"
+
+  if is_local "$host"; then
+    rm -rf "$build_dir"
+    mkdir -p "$build_dir/patches"
+    cp "${SCRIPT_DIR}/Dockerfile.pd" "$build_dir/Dockerfile"
+    cp "${SCRIPT_DIR}/patches/31869.patch" "$build_dir/patches/31869.patch"
+    cp "${SCRIPT_DIR}/patches/32797.patch" "$build_dir/patches/32797.patch"
+    cp "${MOONCAKE_WHEEL}" "$build_dir/mooncake.whl"
+  else
+    ssh "${SSH_OPTIONS[@]}" "${SSH_USER}@${host}" \
+      "rm -rf ${build_dir} && mkdir -p ${build_dir}/patches"
+    scp "${SSH_OPTIONS[@]}" "${SCRIPT_DIR}/Dockerfile.pd" \
+      "${SSH_USER}@${host}:${build_dir}/Dockerfile"
+    scp "${SSH_OPTIONS[@]}" "${SCRIPT_DIR}/patches/31869.patch" \
+      "${SSH_USER}@${host}:${build_dir}/patches/31869.patch"
+    scp "${SSH_OPTIONS[@]}" "${SCRIPT_DIR}/patches/32797.patch" \
+      "${SSH_USER}@${host}:${build_dir}/patches/32797.patch"
+    scp "${SSH_OPTIONS[@]}" "${MOONCAKE_WHEEL}" \
+      "${SSH_USER}@${host}:${build_dir}/mooncake.whl"
+  fi
+
+  sudo_host "$host" docker build \
+    --build-arg "BASE_IMAGE=${BASE_IMAGE}" \
+    -t "${PD_IMAGE}" "$build_dir" \
+    >"${RESULT_ROOT}/logs/build_node${suffix}.log" 2>&1
+  log "PD image built on node ${suffix}"
+}
+
+build_image_all() {
+  local host
+  local -a build_nodes
+  read -r -a build_nodes <<<"${BUILD_NODES_OVERRIDE:-${ALL_NODES[*]}}"
+  for host in "${build_nodes[@]}"; do
+    build_image_on_host "$host"
+  done
+}
+
+preflight() {
+  require_password
+  local host suffix
+  : >"${RESULT_ROOT}/metadata/preflight.tsv"
+  printf 'node\timage\tmodel\trdma\tgpu_processes\tports\n' \
+    >>"${RESULT_ROOT}/metadata/preflight.tsv"
+  for host in "${ALL_NODES[@]}"; do
+    suffix="$(node_suffix "$host")"
+    log "preflight node ${suffix}"
+    sudo_host "$host" docker image inspect "$PD_IMAGE" >/dev/null
+    remote "$host" test -d "$MODEL_PATH"
+    remote "$host" test -e /dev/infiniband/uverbs0
+    local gpu_pids ports
+    gpu_pids="$(remote "$host" nvidia-smi --query-compute-apps=pid --format=csv,noheader | xargs || true)"
+    ports="$(remote "$host" sh -c "ss -lnt | grep -E ':(20000|28800|30000|31000|50051)[[:space:]]' || true")"
+    if [[ -n "$gpu_pids" ]]; then
+      echo "ERROR: node ${suffix} still has GPU processes: ${gpu_pids}" >&2
+      return 1
+    fi
+    if [[ -n "$ports" ]]; then
+      echo "ERROR: node ${suffix} has occupied PD ports: ${ports}" >&2
+      return 1
+    fi
+    printf '%s\tok\tok\tok\t0\tfree\n' "$suffix" \
+      >>"${RESULT_ROOT}/metadata/preflight.tsv"
+  done
+
+  local hash_cmd
+  hash_cmd="sha256sum /sgl-workspace/sglang/python/sglang/srt/disaggregation/prefill.py /sgl-workspace/sglang/python/sglang/srt/disaggregation/decode.py /sgl-workspace/sglang/python/sglang/srt/disaggregation/utils.py /sgl-workspace/sglang/python/sglang/srt/managers/scheduler_pp_mixin.py /sgl-workspace/sglang/python/sglang/srt/managers/scheduler.py; python3 -m pip list | grep -E 'sglang|flashinfer|mooncake-transfer-engine'"
+  sudo_host "$P_HEAD" docker run --rm --entrypoint bash "$PD_IMAGE" -lc "$hash_cmd" \
+    >"${RESULT_ROOT}/metadata/pd_image_versions.txt"
+  grep -q 'mooncake-transfer-engine-cuda13.*0.3.12.post1' \
+    "${RESULT_ROOT}/metadata/pd_image_versions.txt"
+  log "preflight passed on all 8 nodes"
+}
+
+common_docker_args() {
+  local host="$1"
+  local suffix
+  suffix="$(node_suffix "$host")"
+  COMMON_DOCKER_ARGS=(
+    --gpus all --network host --ipc=host --ulimit memlock=-1
+    --device /dev/infiniband --shm-size 32g --entrypoint bash
+    -v "${MODEL_PATH}:${MODEL_PATH}:ro"
+    -e CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7
+    -e "SGLANG_HOST_IP=174.1.60.${suffix}"
+    -e GLOO_SOCKET_IFNAME=bond0
+    -e NCCL_SOCKET_IFNAME=bond1
+    -e "NCCL_IB_HCA=${NCCL_IB_HCAS}"
+    -e NCCL_IB_GID_INDEX=3
+    -e NCCL_IB_TIMEOUT=22
+    -e NCCL_IB_RETRY_CNT=7
+    -e NCCL_CUMEM_ENABLE=1
+    -e "MOONCAKE_MASTER=${MC_MASTER}"
+    -e MOONCAKE_PROTOCOL=rdma
+    -e SGLANG_ENABLE_TP_MEMORY_INBALANCE_CHECK=0
+    -e SGLANG_MOE_FUSED_GATE_RADIX=1
+    -e FLASHINFER_DISABLE_JIT=1
+    -e FLASHINFER_DISABLE_VERSION_CHECK=1
+  )
+}
+
+start_node() {
+  local group="$1" rank="$2" host name bootstrap
+  local -a nodes launch_args cmd
+  if [[ "$group" == "p" ]]; then
+    nodes=("${P_NODES[@]}")
+    host="${nodes[$rank]}"
+    name="$(container_name p "$rank")"
+    launch_args=(
+      --model-path "$MODEL_PATH" --served-model-name "$SERVED_MODEL_NAME"
+      --tp-size "$P_TP" --pp-size "$P_PP" --ep-size "$P_EP"
+      --nnodes 4 --node-rank "$rank" --dist-init-addr "${P_HEAD}:${P_DIST_PORT}"
+      --trust-remote-code --moe-runner-backend flashinfer_mxfp4
+      --chunked-prefill-size "$P_CHUNKED_PREFILL_SIZE"
+      --page-size "$P_PAGE_SIZE"
+      --mem-fraction-static "$P_MEM_FRACTION_STATIC"
+      --cuda-graph-max-bs-decode 16
+      --mamba-radix-cache-strategy extra_buffer_lazy
+      --disable-radix-cache --dist-timeout 3600
+      --mamba-full-memory-ratio "$P_MAMBA_FULL_MEMORY_RATIO"
+      --disaggregation-transfer-backend mooncake
+      --disaggregation-bootstrap-port "$P_BOOTSTRAP_PORT"
+      --disaggregation-ib-device "$DISAGG_IB_DEVICES"
+      --disaggregation-mode prefill
+      --host 0.0.0.0 --port "$P_PORT"
+    )
+  else
+    nodes=("${D_NODES[@]}")
+    host="${nodes[$rank]}"
+    name="$(container_name d "$rank")"
+    launch_args=(
+      --model-path "$MODEL_PATH" --served-model-name "$SERVED_MODEL_NAME"
+      --tp-size "$D_TP" --pp-size "$D_PP" --ep-size "$D_EP"
+      --nnodes 4 --node-rank "$rank" --dist-init-addr "${D_HEAD}:${D_DIST_PORT}"
+      --trust-remote-code --moe-runner-backend marlin
+      --mem-fraction-static "$D_MEM_FRACTION_STATIC"
+      --cuda-graph-max-bs-decode 16
+      --mamba-radix-cache-strategy extra_buffer_lazy
+      --disable-radix-cache --dist-timeout 3600
+      --mamba-full-memory-ratio "$D_MAMBA_FULL_MEMORY_RATIO"
+      --page-size "$D_PAGE_SIZE"
+      --disaggregation-transfer-backend mooncake
+      --disaggregation-bootstrap-port "$P_BOOTSTRAP_PORT"
+      --disaggregation-ib-device "$DISAGG_IB_DEVICES"
+      --disaggregation-mode decode
+      --host 0.0.0.0 --port "$D_PORT"
+    )
+  fi
+
+  printf -v bootstrap '%q ' python3 -m sglang.launch_server "${launch_args[@]}"
+  bootstrap="exec ${bootstrap}"
+  common_docker_args "$host"
+  cmd=(docker run -d --name "$name" "${COMMON_DOCKER_ARGS[@]}" "$PD_IMAGE" -lc "$bootstrap")
+  printf '%q ' "${cmd[@]}" >"${RESULT_ROOT}/commands/${group}_node${rank}.cmd.txt"
+  printf '\n' >>"${RESULT_ROOT}/commands/${group}_node${rank}.cmd.txt"
+  sudo_host "$host" docker rm -f "$name" >/dev/null 2>&1 || true
+  sudo_host "$host" "${cmd[@]}" >/dev/null
+  log "started ${group} rank ${rank} on node $(node_suffix "$host")"
+}
+
+wait_health() {
+  local label="$1" host="$2" port="$3" elapsed
+  for ((elapsed = 1; elapsed <= HEALTH_WAIT_S; elapsed++)); do
+    if curl --fail --silent --max-time 5 "http://${host}:${port}/health" >/dev/null 2>&1; then
+      log "${label} healthy after ${elapsed}s"
+      return 0
+    fi
+    if (( elapsed % 30 == 0 )); then
+      log "waiting for ${label}, elapsed=${elapsed}s"
+      collect_group_logs "${label}_starting"
+      if grep -Eiq 'Traceback|CUDA out of memory|NCCL.*(error|failed)|AssertionError|RuntimeError' \
+          "${RESULT_ROOT}/logs/${label}_starting_"*.log 2>/dev/null; then
+        log "ERROR: fatal pattern found while starting ${label}"
+        return 1
+      fi
+    fi
+    sleep 1
+  done
+  return 1
+}
+
+start_group() {
+  local group="$1" rank
+  for rank in 1 2 3; do
+    start_node "$group" "$rank"
+  done
+  sleep 5
+  start_node "$group" 0
+  if [[ "$group" == "p" ]]; then
+    wait_health p "$P_HEAD" "$P_PORT"
+  else
+    wait_health d "$D_HEAD" "$D_PORT"
+  fi
+  collect_group_logs "${group}_healthy"
+}
+
+start_mc_master() {
+  sudo_host "$P_HEAD" docker rm -f "$MC_CONTAINER" >/dev/null 2>&1 || true
+  sudo_host "$P_HEAD" docker run -d --name "$MC_CONTAINER" --network host --gpus all \
+    --entrypoint /usr/local/bin/mooncake_master "$PD_IMAGE" >/dev/null
+  for _ in $(seq 1 30); do
+    remote "$P_HEAD" sh -c "ss -lnt | grep -q ':50051 '" && {
+      log "Mooncake master listening on 50051"
+      return 0
+    }
+    sleep 1
+  done
+  return 1
+}
+
+start_router() {
+  local -a cmd=(
+    docker run -d --name "$ROUTER_CONTAINER" --network host
+    --entrypoint python3 "$PD_IMAGE" -m sglang_router.launch_router
+    --pd-disaggregation --mini-lb
+    --prefill "http://${P_HEAD}:${P_PORT}" "$P_BOOTSTRAP_PORT"
+    --decode "http://${D_HEAD}:${D_PORT}"
+    --host 0.0.0.0 --port "$ROUTER_PORT"
+  )
+  printf '%q ' "${cmd[@]}" >"${RESULT_ROOT}/commands/router.cmd.txt"
+  printf '\n' >>"${RESULT_ROOT}/commands/router.cmd.txt"
+  sudo_host "$ROUTER_HOST" docker rm -f "$ROUTER_CONTAINER" >/dev/null 2>&1 || true
+  sudo_host "$ROUTER_HOST" "${cmd[@]}" >/dev/null
+  wait_health router "$ROUTER_HOST" "$ROUTER_PORT"
+}
+
+collect_group_logs() {
+  local label="$1" group rank host name
+  for group in p d; do
+    for rank in 0 1 2 3; do
+      if [[ "$group" == "p" ]]; then
+        host="${P_NODES[$rank]}"
+      else
+        host="${D_NODES[$rank]}"
+      fi
+      name="$(container_name "$group" "$rank")"
+      sudo_host "$host" docker logs "$name" \
+        >"${RESULT_ROOT}/logs/${label}_${group}_node${rank}.log" 2>&1 || true
+    done
+  done
+  sudo_host "$P_HEAD" docker logs "$MC_CONTAINER" \
+    >"${RESULT_ROOT}/logs/${label}_mc_master.log" 2>&1 || true
+  sudo_host "$ROUTER_HOST" docker logs "$ROUTER_CONTAINER" \
+    >"${RESULT_ROOT}/logs/${label}_router.log" 2>&1 || true
+}
+
+status_all() {
+  local host
+  for host in "${ALL_NODES[@]}"; do
+    echo "===== node $(node_suffix "$host") ====="
+    sudo_host "$host" docker ps --format '{{.Names}}|{{.Image}}|{{.Status}}' \
+      --filter "name=${EXPERIMENT}"
+    remote "$host" nvidia-smi --query-gpu=index,memory.used,utilization.gpu \
+      --format=csv,noheader
+  done
+  for endpoint in \
+    "p=http://${P_HEAD}:${P_PORT}/health" \
+    "d=http://${D_HEAD}:${D_PORT}/health" \
+    "router=http://${ROUTER_HOST}:${ROUTER_PORT}/health"; do
+    local name="${endpoint%%=*}" url="${endpoint#*=}"
+    if curl --fail --silent --max-time 5 "$url" >/dev/null 2>&1; then
+      echo "${name}=healthy"
+    else
+      echo "${name}=down"
+    fi
+  done
+}
+
+stop_all() {
+  local rank host name
+  sudo_host "$ROUTER_HOST" docker rm -f "$ROUTER_CONTAINER" >/dev/null 2>&1 || true
+  for rank in 0 1 2 3; do
+    host="${D_NODES[$rank]}"
+    name="$(container_name d "$rank")"
+    sudo_host "$host" docker rm -f "$name" >/dev/null 2>&1 || true
+  done
+  for rank in 0 1 2 3; do
+    host="${P_NODES[$rank]}"
+    name="$(container_name p "$rank")"
+    sudo_host "$host" docker rm -f "$name" >/dev/null 2>&1 || true
+  done
+  sudo_host "$P_HEAD" docker rm -f "$MC_CONTAINER" >/dev/null 2>&1 || true
+  log "standard PD service stopped"
+}
+
+start_all() {
+  preflight
+  start_mc_master
+  start_group p
+  start_group d
+  start_router
+  collect_group_logs ready
+  status_all | tee "${RESULT_ROOT}/metadata/status_ready.txt"
+  log "standard PD service ready at http://${ROUTER_HOST}:${ROUTER_PORT}"
+}
+
+case "$ACTION" in
+  build-image) build_image_all ;;
+  preflight) preflight ;;
+  start) start_all ;;
+  stop) stop_all ;;
+  restart) stop_all; start_all ;;
+  status) status_all ;;
+  logs) collect_group_logs manual ;;
+  *)
+    echo "Usage: $0 {build-image|preflight|start|stop|restart|status|logs}" >&2
+    exit 2
+    ;;
+esac
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/patches/31869.patch b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/patches/31869.patch
new file mode 100644
index 0000000..6deb6be
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/patches/31869.patch
@@ -0,0 +1,303 @@
+From 5d8dd1f7f99f4c74bc302dd9ef6d32c00c41f7a8 Mon Sep 17 00:00:00 2001
+From: ziang663 <1902123738@qq.com>
+Date: Fri, 24 Jul 2026 03:15:10 +0800
+Subject: [PATCH] fix(disagg): honor PP consensus for bootstrap and prealloc
+
+---
+ python/sglang/srt/disaggregation/decode.py    | 50 +++++++++++++++----
+ python/sglang/srt/disaggregation/prefill.py   | 46 +++++++++++------
+ python/sglang/srt/disaggregation/utils.py     | 29 ++++++++++-
+ .../sglang/srt/managers/scheduler_pp_mixin.py |  8 +--
+ .../test_decode_queue_cleanup.py              |  2 +
+ ...test_priority_scheduling_disaggregation.py |  1 +
+ .../mem_cache/test_decode_radix_lock_ref.py   |  1 +
+ 7 files changed, 106 insertions(+), 31 deletions(-)
+
+diff --git a/python/sglang/srt/disaggregation/decode.py b/python/sglang/srt/disaggregation/decode.py
+index 5224920c4481..a664fbe85119 100644
+--- a/python/sglang/srt/disaggregation/decode.py
++++ b/python/sglang/srt/disaggregation/decode.py
+@@ -55,6 +55,7 @@
+     is_dsv4_c128_online_enabled,
+     is_mla_backend,
+     poll_and_all_reduce,
++    poll_and_all_reduce_pp,
+     poll_and_all_reduce_with_staging,
+     prepare_abort,
+     setup_state_kv_args,
+@@ -325,6 +326,7 @@ def __init__(
+         self.bootstrap_port = bootstrap_port
+         self.max_total_num_tokens = max_total_num_tokens
+         self.pp_rank = pp_rank
++        self.pp_size = scheduler.ps.pp_size
+         self.num_reserved_decode_tokens = num_reserved_decode_tokens
+         self.transfer_backend = transfer_backend
+         # Queue for requests pending pre-allocation
+@@ -719,23 +721,40 @@ def resume_retracted_reqs(
+         return resumed_reqs
+ 
+     def _update_handshake_waiters(
+-        self, rids_to_check: Optional[List[str]] = None
++        self,
++        rids_to_check: Optional[List[str]] = None,
++        pp_good_rids: Optional[List[str]] = None,
++        pp_bad_rids: Optional[List[str]] = None,
+     ) -> None:
+         if not self.queue:
+             return
+ 
+         # Still poll if any receiver was aborted, otherwise it stays stuck.
+-        if all(decode_req.waiting_for_input for decode_req in self.queue) and not any(
+-            decode_req.kv_receiver.conclude_state == KVPoll.Failed
+-            for decode_req in self.queue
++        if (
++            self.pp_size <= 1
++            and all(decode_req.waiting_for_input for decode_req in self.queue)
++            and not any(
++                decode_req.kv_receiver.conclude_state == KVPoll.Failed
++                for decode_req in self.queue
++            )
+         ):
+             return
+ 
+-        polls = poll_and_all_reduce(
+-            [decode_req.kv_receiver for decode_req in self.queue], self.gloo_group
+-        )
++        if self.pp_size > 1:
++            polls = poll_and_all_reduce_pp(
++                (decode_req.req.rid for decode_req in self.queue),
++                KVPoll.WaitingForInput,
++                pp_good_rids,
++                pp_bad_rids,
++            )
++        else:
++            polls = poll_and_all_reduce(
++                [decode_req.kv_receiver for decode_req in self.queue], self.gloo_group
++            )
+ 
+-        for i, (decode_req, poll) in enumerate(zip(self.queue, polls)):
++        for decode_req, poll in zip(self.queue, polls):
++            if poll is None:
++                continue
+             if rids_to_check is not None and decode_req.req.rid not in rids_to_check:
+                 continue
+ 
+@@ -856,11 +875,22 @@ def _resolve_pending_reqs(self) -> None:
+             decode_req.kv_receiver.init(prefill_dp_rank)
+ 
+     def pop_preallocated(
+-        self, rids_to_check: Optional[List[str]] = None
++        self,
++        rids_to_check: Optional[List[str]] = None,
++        pp_good_rids: Optional[List[str]] = None,
++        pp_bad_rids: Optional[List[str]] = None,
+     ) -> Tuple[List[DecodeRequest], List[DecodeRequest]]:
+         """Pop the preallocated requests from the pending queue (FIFO)."""
++        is_pp_mode = self.pp_size > 1
++        if is_pp_mode and (pp_good_rids is None or pp_bad_rids is None):
++            raise ValueError("PP consensus is required when pp_size > 1")
++        if is_pp_mode and rids_to_check is not None:
++            raise ValueError("rids_to_check cannot be used in PP mode")
++
+         self._resolve_pending_reqs()
+-        self._update_handshake_waiters(rids_to_check)
++        self._update_handshake_waiters(rids_to_check, pp_good_rids, pp_bad_rids)
++        if is_pp_mode:
++            rids_to_check = pp_good_rids + pp_bad_rids
+ 
+         failed_reqs = []
+         preallocated_reqs = []
+diff --git a/python/sglang/srt/disaggregation/prefill.py b/python/sglang/srt/disaggregation/prefill.py
+index 1f4c947fa3fa..51d6e2dba7f2 100644
+--- a/python/sglang/srt/disaggregation/prefill.py
++++ b/python/sglang/srt/disaggregation/prefill.py
+@@ -45,6 +45,7 @@
+     is_dsv4_c128_online_enabled,
+     is_mla_backend,
+     poll_and_all_reduce_attn_cp_tp_group,
++    poll_and_all_reduce_pp,
+     prepare_abort,
+     setup_state_kv_args,
+ )
+@@ -333,13 +334,15 @@ def _process_req(self, req: Req) -> None:
+     def pop_bootstrapped(
+         self,
+         return_failed_reqs: bool = False,
+-        rids_to_check: Optional[List[str]] = None,
+-    ) -> List[Req]:
++        pp_good_rids: Optional[List[str]] = None,
++        pp_bad_rids: Optional[List[str]] = None,
++    ) -> List[Req] | tuple[List[Req], List[Req]]:
+         """
+         pop the reqs which has finished bootstrapping
+ 
+         return_failed_reqs: For PP, on rank 0, also return the failed reqs to notify the next rank
+-        rids_to_check: For PP, on rank > 0, check the rids from the previous rank has consensus with the current rank.
++        pp_good_rids: RIDs that PP consensus determined as WaitingForInput.
++        pp_bad_rids: RIDs that PP consensus determined as Failed.
+         """
+ 
+         bootstrapped_reqs = []
+@@ -352,21 +355,32 @@ def pop_bootstrapped(
+             else:
+                 return [], []
+ 
+-        polls = poll_and_all_reduce_attn_cp_tp_group(
+-            [req.disagg_kv_sender for req in self.queue],
+-            self.scheduler.attn_cp_cpu_group,
+-            self.scheduler.attn_tp_cpu_group,
+-        )
++        if self.pp_size > 1:
++            polls = poll_and_all_reduce_pp(
++                (req.rid for req in self.queue),
++                KVPoll.WaitingForInput,
++                pp_good_rids,
++                pp_bad_rids,
++            )
++            uncovered = [i for i, poll in enumerate(polls) if poll is None]
++            if uncovered:
++                local_polls = poll_and_all_reduce_attn_cp_tp_group(
++                    [self.queue[i].disagg_kv_sender for i in uncovered],
++                    self.scheduler.attn_cp_cpu_group,
++                    self.scheduler.attn_tp_cpu_group,
++                )
++                for i, local_poll in zip(uncovered, local_polls):
++                    if local_poll == KVPoll.Failed:
++                        polls[i] = KVPoll.Failed
++        else:
++            polls = poll_and_all_reduce_attn_cp_tp_group(
++                [req.disagg_kv_sender for req in self.queue],
++                self.scheduler.attn_cp_cpu_group,
++                self.scheduler.attn_tp_cpu_group,
++            )
+ 
+         for i, (req, poll) in enumerate(zip(self.queue, polls)):
+-            if (
+-                rids_to_check is not None
+-                and req.rid not in rids_to_check
+-                and poll != KVPoll.Failed
+-            ):
+-                # In PP mode, successful bootstrap still requires cross-rank
+-                # consensus. Local failures are terminal and must be drained
+-                # even if an earlier PP rank has already removed the request.
++            if poll is None:
+                 continue
+ 
+             if poll == KVPoll.Failed:
+diff --git a/python/sglang/srt/disaggregation/utils.py b/python/sglang/srt/disaggregation/utils.py
+index 8d741804d875..7893d588248d 100644
+--- a/python/sglang/srt/disaggregation/utils.py
++++ b/python/sglang/srt/disaggregation/utils.py
+@@ -5,7 +5,16 @@
+ from collections import deque
+ from contextlib import nullcontext
+ from enum import Enum
+-from typing import TYPE_CHECKING, List, Literal, Optional, Tuple, Type, overload
++from typing import (
++    TYPE_CHECKING,
++    Iterable,
++    List,
++    Literal,
++    Optional,
++    Tuple,
++    Type,
++    overload,
++)
+ 
+ import numpy as np
+ import torch
+@@ -34,6 +43,24 @@
+ _IS_HIP = is_hip()
+ 
+ 
++def poll_and_all_reduce_pp(
++    rids: Iterable[str],
++    ready_poll: int,
++    pp_good_rids: Optional[List[str]] = None,
++    pp_bad_rids: Optional[List[str]] = None,
++) -> List[Optional[int]]:
++    """Map authoritative PP consensus to poll states without polling again."""
++    if pp_good_rids is None or pp_bad_rids is None:
++        raise ValueError("PP consensus is required")
++
++    good_rids = set(pp_good_rids)
++    bad_rids = set(pp_bad_rids)
++    return [
++        KVPoll.Failed if rid in bad_rids else ready_poll if rid in good_rids else None
++        for rid in rids
++    ]
++
++
+ def get_dsa_seed_metadata_dim(hf_config) -> int:
+     """Return the model-defined PD seed width, independent of local spec mode."""
+     if not getattr(hf_config, "index_share_for_mtp_iteration", False):
+diff --git a/python/sglang/srt/managers/scheduler_pp_mixin.py b/python/sglang/srt/managers/scheduler_pp_mixin.py
+index 0a55601d4128..f3e1f76f4ad1 100644
+--- a/python/sglang/srt/managers/scheduler_pp_mixin.py
++++ b/python/sglang/srt/managers/scheduler_pp_mixin.py
+@@ -811,8 +811,8 @@ def process_bootstrapped_queue(
+             good_reqs, failed_reqs = (
+                 self.disagg_prefill_bootstrap_queue.pop_bootstrapped(
+                     return_failed_reqs=True,
+-                    rids_to_check=good_consensus_bootstrapped_rids
+-                    + bad_consensus_bootstrapped_rids,
++                    pp_good_rids=good_consensus_bootstrapped_rids,
++                    pp_bad_rids=bad_consensus_bootstrapped_rids,
+                 )
+             )
+             self.waiting_queue.extend(good_reqs)
+@@ -1417,8 +1417,8 @@ def process_prealloc_queue(self: Scheduler, prealloc_rids: Optional[List[str]]):
+                 bad_consensus_prealloc_rids,
+             ) = prealloc_rids
+             good_reqs, failed_reqs = self.disagg_decode_prealloc_queue.pop_preallocated(
+-                rids_to_check=good_consensus_prealloc_rids
+-                + bad_consensus_prealloc_rids,
++                pp_good_rids=good_consensus_prealloc_rids,
++                pp_bad_rids=bad_consensus_prealloc_rids,
+             )
+             self.disagg_decode_transfer_queue.extend(good_reqs)
+             return [
+diff --git a/test/registered/unit/disaggregation/test_decode_queue_cleanup.py b/test/registered/unit/disaggregation/test_decode_queue_cleanup.py
+index e136befc4d54..0a904792b5af 100644
+--- a/test/registered/unit/disaggregation/test_decode_queue_cleanup.py
++++ b/test/registered/unit/disaggregation/test_decode_queue_cleanup.py
+@@ -40,6 +40,7 @@ def test_prealloc_abort_clears_receiver_before_removing_request(self):
+         decode_req = SimpleNamespace(req=req, kv_receiver=receiver)
+ 
+         queue = DecodePreallocQueue.__new__(DecodePreallocQueue)
++        queue.pp_size = 1
+         queue.queue = [decode_req]
+         queue.pending_reqs = []
+         queue.retracted_queue = []
+@@ -86,6 +87,7 @@ def __eq__(self, other):
+         decode_req = SimpleNamespace(req=req, kv_receiver=receiver)
+ 
+         queue = DecodePreallocQueue.__new__(DecodePreallocQueue)
++        queue.pp_size = 1
+         queue.queue = [decode_req]
+         queue.pending_reqs = [decode_req]  # same object, dual ownership
+         queue.retracted_queue = []
+diff --git a/test/registered/unit/managers/test_priority_scheduling_disaggregation.py b/test/registered/unit/managers/test_priority_scheduling_disaggregation.py
+index a5f052c7311e..762cb5489ccb 100644
+--- a/test/registered/unit/managers/test_priority_scheduling_disaggregation.py
++++ b/test/registered/unit/managers/test_priority_scheduling_disaggregation.py
+@@ -119,6 +119,7 @@ def _new_decode_req(self, rid: str, priority: int, *, failed: bool = False):
+ 
+     def _new_queue(self, decode_reqs, *, low_priority_values_first: bool = False):
+         queue = DecodePreallocQueue.__new__(DecodePreallocQueue)
++        queue.pp_size = 1
+         queue.queue = list(decode_reqs)
+         queue.pending_reqs = []
+         queue.retracted_queue = []
+diff --git a/test/registered/unit/mem_cache/test_decode_radix_lock_ref.py b/test/registered/unit/mem_cache/test_decode_radix_lock_ref.py
+index 7cba0af202b7..4abdfdc1b933 100644
+--- a/test/registered/unit/mem_cache/test_decode_radix_lock_ref.py
++++ b/test/registered/unit/mem_cache/test_decode_radix_lock_ref.py
+@@ -294,6 +294,7 @@ def test_full_transfer_failure(self):
+ 
+     def test_pop_preallocated_rechecks_budget_after_lock(self):
+         queue = DecodePreallocQueue.__new__(DecodePreallocQueue)
++        queue.pp_size = 1
+ 
+         req = MagicMock()
+         req.rid = "req-1"
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/patches/32797.patch b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/patches/32797.patch
new file mode 100644
index 0000000..6e3ae7b
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/patches/32797.patch
@@ -0,0 +1,99 @@
+diff --git a/python/sglang/srt/disaggregation/decode.py b/python/sglang/srt/disaggregation/decode.py
+index ec0579ec1e24..524f0bdb6fea 100644
+--- a/python/sglang/srt/disaggregation/decode.py
++++ b/python/sglang/srt/disaggregation/decode.py
+@@ -898,7 +898,7 @@ def pop_preallocated(
+         self._resolve_pending_reqs()
+         self._update_handshake_waiters(rids_to_check, pp_good_rids, pp_bad_rids)
+         if is_pp_mode:
+-            rids_to_check = pp_good_rids + pp_bad_rids
++            rids_to_check = set(pp_good_rids) | set(pp_bad_rids)
+ 
+         failed_reqs = []
+         preallocated_reqs = []
+diff --git a/python/sglang/srt/managers/scheduler.py b/python/sglang/srt/managers/scheduler.py
+index 862ba08f5ad3..861c13fb4e49 100644
+--- a/python/sglang/srt/managers/scheduler.py
++++ b/python/sglang/srt/managers/scheduler.py
+@@ -4224,6 +4224,8 @@ def abort_request(self, recv_req: AbortReq):
+ 
+                     if hasattr(req.disagg_kv_sender, "abort"):
+                         req.disagg_kv_sender.abort()
++                    if self.ps.pp_size > 1:
++                        prepare_abort(req, "Aborted by AbortReq.")
+ 
+             # Abort in-flight requests
+             for req in self.disagg_prefill_inflight_queue:
+@@ -4238,6 +4240,8 @@ def abort_request(self, recv_req: AbortReq):
+                 if recv_req.abort_all or decode_req.req.rid.startswith(recv_req.rid):
+                     logger.debug(f"Abort prealloc queue request. {decode_req.req.rid=}")
+                     decode_req.kv_receiver.abort()
++                    if self.ps.pp_size > 1:
++                        prepare_abort(decode_req.req, "Aborted by AbortReq.")
+ 
+             # Abort requests waiting for kvcache to release tree cache
+             for decode_req in self.disagg_decode_transfer_queue.queue:
+diff --git a/python/sglang/srt/managers/scheduler_pp_mixin.py b/python/sglang/srt/managers/scheduler_pp_mixin.py
+index f3e1f76f4ad1..4e475f8a76a5 100644
+--- a/python/sglang/srt/managers/scheduler_pp_mixin.py
++++ b/python/sglang/srt/managers/scheduler_pp_mixin.py
+@@ -24,7 +24,7 @@
+     set_is_extend_in_batch,
+ )
+ from sglang.srt.managers.overlap_utils import RelayPayload
+-from sglang.srt.managers.schedule_batch import Req, ScheduleBatch
++from sglang.srt.managers.schedule_batch import FINISH_ABORT, Req, ScheduleBatch
+ from sglang.srt.managers.utils import (
+     GenerationBatchResult,
+     get_logprob_dict_from_result,
+@@ -847,6 +847,18 @@ def _pp_pd_get_bootstrapped_ids(self: Scheduler):
+             bad_bootstrapped_rids = list(
+                 set(prev_bad_bootstrapped_rids) | set(curr_bad_bootstrapped_rids)
+             )
++        # Route locally-aborted reqs through the bad-union consensus so every PP
++        # rank flushes them in the same consensus round, regardless of when the
++        # AbortReq reaches each rank and regardless of whether
++        # disagg_kv_sender.abort() drives the poll to Failed (it is optional).
++        aborted_rids = {
++            req.rid
++            for req in self.disagg_prefill_bootstrap_queue.queue
++            if isinstance(req.finished_reason, FINISH_ABORT)
++        }
++        good_bootstrapped_rids, bad_bootstrapped_rids = self._route_aborts_to_bad(
++            good_bootstrapped_rids, bad_bootstrapped_rids, aborted_rids
++        )
+         return [good_bootstrapped_rids, bad_bootstrapped_rids]
+ 
+     def _pp_pd_get_prefill_transferred_ids(self: Scheduler):
+@@ -1369,8 +1381,31 @@ def _pp_pd_get_prealloc_ids(self: Scheduler):
+             bad_prealloc_rids = list(
+                 set(prev_bad_prealloc_rids) | set(curr_bad_prealloc_rids)
+             )
++        # Same abort routing as the prefill bootstrap consensus above.
++        aborted_rids = {
++            decode_req.req.rid
++            for decode_req in self.disagg_decode_prealloc_queue.queue
++            if isinstance(decode_req.req.finished_reason, FINISH_ABORT)
++        }
++        good_prealloc_rids, bad_prealloc_rids = self._route_aborts_to_bad(
++            good_prealloc_rids, bad_prealloc_rids, aborted_rids
++        )
+         return [good_prealloc_rids, bad_prealloc_rids]
+ 
++    @staticmethod
++    def _route_aborts_to_bad(good_rids, bad_rids, aborted_rids):
++        """Move aborted rids out of the good (intersection) set and into the
++        bad (union) set, so PP consensus fails them uniformly on every rank.
++
++        This also flushes aborted reqs that never reached good/bad consensus
++        (e.g. stuck in Bootstrapping with a sender that has no working abort()).
++        """
++        if not aborted_rids:
++            return good_rids, bad_rids
++        good_rids = [rid for rid in good_rids if rid not in aborted_rids]
++        bad_rids = list(set(bad_rids) | set(aborted_rids))
++        return good_rids, bad_rids
++
+     def _pp_pd_get_decode_transferred_ids(self: Scheduler):
+         # get the current stage transfer success
+         if self.pp_group.is_first_rank:
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-20260827-1325/bench/decode_16k_512_c1.cmd.txt b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-20260827-1325/bench/decode_16k_512_c1.cmd.txt
new file mode 100644
index 0000000..44f26dc
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-20260827-1325/bench/decode_16k_512_c1.cmd.txt
@@ -0,0 +1 @@
+docker run --rm --network host -v /data/hf_models/Kimi-K3:/data/hf_models/Kimi-K3:ro -v /data/hzy/sskj/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-20260827-1325:/data/hzy/sskj/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-20260827-1325 --entrypoint python3 local/sglang:kimi-k3-sm120-flashinfer-mxfp4-pd1 -m sglang.benchmark.serving --backend sglang --host 174.1.60.5 --port 31000 --tokenizer /data/hf_models/Kimi-K3 --dataset-name random-ids --random-input-len 16384 --random-output-len 512 --random-range-ratio 1.0 --num-prompts 1 --max-concurrency 1 --request-rate 10000 --warmup-requests 0 --output-file /data/hzy/sskj/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-20260827-1325/bench/decode_16k_512_c1.jsonl --output-details --disable-tqdm 
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-20260827-1325/bench/decode_16k_512_c1.jsonl b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-20260827-1325/bench/decode_16k_512_c1.jsonl
new file mode 100644
index 0000000..597db3e
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-20260827-1325/bench/decode_16k_512_c1.jsonl
@@ -0,0 +1 @@
+{"tag": null, "backend": "sglang", "dataset_name": "random-ids", "request_rate": 10000.0, "max_concurrency": 1, "sharegpt_output_len": null, "random_input_len": 16384, "random_output_len": 512, "random_range_ratio": 1.0, "server_info": {"internal_states": [{"model_path": "/data/hf_models/Kimi-K3", "tokenizer_path": "/data/hf_models/Kimi-K3", "tokenizer_mode": "auto", "tokenizer_backend": "huggingface", "tokenizer_worker_num": 1, "detokenizer_worker_num": 1, "skip_tokenizer_init": false, "load_format": "auto", "model_loader_extra_config": "{}", "trust_remote_code": true, "context_length": null, "is_embedding": false, "enable_multimodal": null, "revision": null, "model_impl": "auto", "model_config_parser": "auto", "json_model_override_args": "{}", "dtype": "auto", "quantization": null, "quantization_param_path": null, "kv_cache_dtype": "auto", "enable_fp32_lm_head": false, "modelopt_quant": null, "modelopt_checkpoint_restore_path": null, "modelopt_checkpoint_save_path": null, "modelopt_export_path": null, "quantize_and_serve": false, "rl_quant_profile": null, "enable_tf32_matmul": false, "mem_fraction_static": 0.88, "max_running_requests": null, "max_queued_requests": null, "max_total_tokens": null, "chunked_prefill_size": 8192, "enable_dynamic_chunking": false, "max_prefill_tokens": 16384, "prefill_max_requests": null, "schedule_policy": "fcfs", "enable_priority_scheduling": false, "disable_priority_preemption": false, "default_priority_value": null, "abort_on_priority_when_disabled": false, "schedule_low_priority_values_first": false, "priority_scheduling_preemption_threshold": 10, "retraction_policy": "length", "schedule_conservativeness": 1.0, "page_size": 64, "swa_full_tokens_ratio": 0.8, "disable_hybrid_swa_memory": false, "radix_eviction_policy": "lru", "prefill_only_disable_kv_cache": false, "disable_radix_cache": true, "enable_page_major_kv_layout": false, "enable_unified_memory": false, "disable_chunked_prefix_cache": true, "disable_overlap_schedule": false, "num_continuous_decode_steps": 1, "scheduler_recv_interval": 1, "enable_mixed_chunk": false, "nccl_port": null, "dist_timeout": 3600, "dist_init_addr": "174.1.60.5:20000", "nnodes": 4, "node_rank": 0, "tp_size": 32, "dcp_size": 1, "pp_size": 1, "pp_max_micro_batch_size": 181, "pp_async_batch_depth": 0, "dp_size": 1, "load_balance_method": "round_robin", "attn_cp_size": 1, "moe_dp_size": 1, "dcp_comm_backend": "ag_rs", "dcp_replicate_q_proj": null, "dwdp_size": 1, "enable_prefill_cp": false, "cp_strategy": null, "enable_dsa_cache_layer_split": false, "enable_dsa_prefill_context_parallel": false, "dsa_prefill_cp_mode": "round-robin-split", "enable_prefill_context_parallel": false, "prefill_cp_mode": "in-seq-split", "enable_cp_decode_attn_tp": false, "enable_dp_attention": false, "enable_dp_attention_local_control_broadcast": false, "enable_dp_lm_head": false, "enable_attn_tp_input_scattered": false, "disable_attn_tp_gather": false, "enable_p2p_check": false, "device": "cuda", "base_gpu_id": 0, "gpu_id_step": 1, "random_seed": 369235288, "watchdog_timeout": 300, "soft_watchdog_timeout": null, "sleep_on_idle": false, "use_ray": false, "numa_node": null, "gc_threshold": null, "host": "0.0.0.0", "port": 30000, "fastapi_root_path": "", "smg_grpc_mode": false, "grpc_mode": false, "grpc_port": null, "sidecar": null, "sidecar_args": null, "skip_server_warmup": false, "warmups": null, "enable_http2": false, "ssl_keyfile": null, "ssl_certfile": null, "ssl_ca_certs": null, "ssl_keyfile_password": null, "enable_ssl_refresh": false, "api_key": null, "admin_api_key": null, "served_model_name": "kimi-k3", "weight_version": "default", "chat_template": null, "hf_chat_template_name": null, "completion_template": null, "file_storage_path": "sglang_storage", "enable_cache_report": false, "reasoning_parser": null, "default_chat_template_kwargs": null, "strip_thinking_cache": false, "enable_strict_thinking": false, "tool_call_parser": null, "tool_server": null, "sampling_defaults": "model", "asr_max_buffer_seconds": 60, "asr_max_concurrent_sessions": 32, "preferred_sampling_params": null, "allow_auto_truncate": false, "stream_interval": 1, "batch_notify_size": 16, "stream_response_default_include_usage": false, "incremental_streaming_output": false, "enable_streaming_session": false, "enable_session_radix_cache": false, "log_level": "info", "log_level_http": null, "log_requests": false, "log_requests_level": 2, "log_requests_format": "text", "log_requests_target": null, "uvicorn_access_log_exclude_prefixes": [], "crash_dump_folder": null, "show_time_cost": false, "enable_metrics": false, "smg_http_sidecar_port": null, "enable_mfu_metrics": false, "enable_metrics_for_all_schedulers": false, "load_snapshot_publish_interval": 15, "tokenizer_metrics_custom_labels_header": "x-custom-labels", "tokenizer_metrics_allowed_custom_labels": null, "extra_metric_labels": null, "bucket_time_to_first_token": null, "bucket_inter_token_latency": null, "bucket_e2e_request_latency": null, "prompt_tokens_buckets": null, "generation_tokens_buckets": null, "gc_warning_threshold_secs": 0.0, "decode_log_interval": 40, "enable_request_time_stats_logging": false, "kv_events_config": null, "enable_forward_pass_metrics": false, "forward_pass_metrics_worker_id": "", "forward_pass_metrics_ipc_name": null, "enable_trace": false, "trace_modules": "request", "otlp_traces_endpoint": "localhost:4317", "export_metrics_to_file": false, "export_metrics_to_file_dir": null, "stat_loggers": null, "constrained_json_whitespace_pattern": null, "constrained_json_disable_any_whitespace": false, "attention_backend": "triton", "decode_attention_backend": null, "prefill_attention_backend": null, "sampling_backend": "flashinfer", "grammar_backend": "xgrammar", "radix_cache_backend": null, "mm_attention_backend": null, "fp8_gemm_runner_backend": "auto", "fp4_gemm_runner_backend": "auto", "bf16_gemm_backend": "auto", "dsa_prefill_backend": null, "dsa_decode_backend": null, "dsa_paged_mqa_logits_backend": "auto", "dsa_topk_backend": "sgl-kernel", "disable_flashinfer_autotune": false, "flashinfer_autotune_skip_ops": null, "mamba_backend": "triton", "cuda_graph_config": {"decode": {"backend": "full", "max_bs": 16, "bs": [1, 2, 4, 8, 12, 16], "tc_compiler": "eager", "full_prefill_max_req": null}, "prefill": {"backend": "disabled", "max_bs": 2048, "bs": [4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], "tc_compiler": "eager", "full_prefill_max_req": null}}, "cuda_graph_backend_decode": null, "cuda_graph_backend_prefill": null, "cuda_graph_max_bs_decode": 16, "cuda_graph_max_bs_prefill": null, "cuda_graph_bs_decode": null, "cuda_graph_bs_prefill": null, "cuda_graph_tc_compiler": null, "disable_prefill_cuda_graph": false, "disable_decode_cuda_graph": false, "disable_cuda_graph": false, "disable_cuda_graph_padding": false, "enable_profile_cuda_graph": false, "enable_cudagraph_gc": false, "debug_cuda_graph": false, "enable_layerwise_nvtx_marker": false, "enable_nccl_nvls": false, "enable_symm_mem": false, "triton_attention_reduce_in_fp32": false, "triton_attention_num_kv_splits": 8, "triton_attention_split_tile_size": null, "flashinfer_mla_disable_ragged": false, "enable_fused_qk_norm_rope": false, "enable_precise_embedding_interpolation": false, "enable_fused_moe_sum_all_reduce": false, "enable_deepseek_v4_fp4_indexer": false, "disable_custom_all_reduce": false, "enable_mscclpp": false, "enable_torch_symm_mem": false, "enable_scattered_sconv": false, "pre_warm_nccl": false, "enable_quant_communications": false, "enable_flashinfer_allreduce_fusion": false, "enforce_disable_flashinfer_allreduce_fusion": false, "flashinfer_allreduce_fusion_backend": null, "enable_aiter_allreduce_fusion": false, "enable_torch_compile": false, "enable_torch_compile_debug_mode": false, "torch_compile_max_bs": 32, "torchao_config": "", "speculative_algorithm": null, "speculative_draft_model_path": null, "speculative_draft_model_revision": null, "speculative_draft_load_format": null, "speculative_num_steps": null, "speculative_eagle_topk": null, "speculative_num_draft_tokens": null, "speculative_dflash_block_size": null, "speculative_dspark_block_size": null, "speculative_dspark_sps_table_path": null, "speculative_dspark_confidence_sts_path": null, "speculative_dspark_align_verify_tokens_to_graph_tier": false, "speculative_accept_threshold_single": 1.0, "speculative_accept_threshold_acc": 1.0, "speculative_use_rejection_sampling": false, "speculative_token_map": null, "speculative_attention_mode": "prefill", "speculative_draft_attention_backend": null, "speculative_draft_window_size": null, "speculative_moe_runner_backend": "marlin", "speculative_moe_a2a_backend": null, "speculative_draft_model_quantization": null, "speculative_skip_dp_mlp_sync": false, "enable_multi_layer_eagle": false, "speculative_adaptive": false, "speculative_adaptive_config": null, "decoupled_spec_bind_endpoint": null, "decoupled_spec_connect_endpoints": null, "decoupled_spec_rank": null, "decoupled_spec_role": "null", "spec_trace_dir": null, "speculative_ngram_min_bfs_breadth": 1, "speculative_ngram_max_bfs_breadth": 10, "speculative_ngram_match_type": "BFS", "speculative_ngram_max_trie_depth": 18, "speculative_ngram_capacity": 10000000, "speculative_ngram_external_corpus_path": null, "speculative_ngram_external_sam_budget": 0, "speculative_ngram_external_corpus_max_tokens": 10000000, "ep_size": 32, "moe_a2a_backend": "none", "moe_runner_backend": "marlin", "flashinfer_mxfp4_moe_precision": "default", "deepep_mode": "auto", "fuseep_mode": 2, "deepep_dispatcher_output_dtype": "auto", "ep_num_redundant_experts": 0, "ep_dispatch_algorithm": null, "init_expert_location": "trivial", "enable_eplb": false, "eplb_algorithm": "auto", "eplb_rebalance_num_iterations": 1000, "eplb_rebalance_layers_per_chunk": null, "eplb_min_rebalancing_utilization_threshold": 1.0, "expert_distribution_recorder_mode": null, "expert_distribution_recorder_buffer_size": 1000, "enable_expert_distribution_metrics": false, "deepep_config": null, "moe_dense_tp_size": null, "elastic_ep_backend": null, "enable_elastic_expert_backup": false, "mooncake_ib_device": null, "enable_waterfill": false, "ep_join_mode": null, "ep_join_rank_offset": 0, "elastic_ep_initial_size": null, "max_ep_size": null, "elastic_ep_scale_timeout": 600, "elastic_ep_rejoin": false, "disable_flashinfer_cutlass_moe_fp4_allgather": false, "disable_shared_experts_fusion": false, "enforce_shared_experts_fusion": false, "max_mamba_cache_size": 181, "mamba_ssm_dtype": null, "mamba_max_states_per_path": -1, "enable_mamba_cache_stochastic_rounding": false, "mamba_cache_philox_rounds": 0, "mamba_full_memory_ratio": 0.21, "mamba_radix_cache_strategy": "extra_buffer_lazy", "uses_mamba_radix_cache": false, "mamba_track_interval": 256, "enable_int8_mamba_checkpoint": false, "int8_mamba_ckpt_size": null, "linear_attn_backend": "triton", "linear_attn_decode_backend": null, "linear_attn_prefill_backend": null, "linear_attn_verify_backend": null, "enable_linear_replayssm": false, "linear_replayssm_cache_len": 16, "enable_linear_replayssm_spec": false, "enable_hierarchical_cache": false, "hicache_ratio": 2.0, "hicache_size": 0, "hicache_write_policy": "write_through", "hicache_io_backend": "kernel", "hicache_mem_layout": "page_first", "hicache_storage_backend": null, "hicache_storage_prefetch_policy": "timeout", "hicache_storage_backend_extra_config": null, "enable_hisparse": false, "hisparse_config": null, "enable_broadcast_mm_inputs_process": false, "enable_prefix_mm_cache": false, "mm_enable_dp_encoder": false, "mm_process_config": {}, "mm_processor_worker_num": 0, "mm_io_worker_num": 0, "limit_mm_data_per_request": null, "enable_mm_global_cache": false, "disable_fast_image_processor": false, "mm_feature_transport": "cpu", "keep_mm_feature_on_device": false, "enable_lora": null, "enable_lora_overlap_loading": null, "max_lora_rank": null, "lora_target_modules": null, "lora_paths": null, "max_loaded_loras": null, "max_loras_per_batch": 8, "lora_eviction_policy": "lru", "lora_backend": "csgmv", "max_lora_chunk_size": 16, "experts_shared_outer_loras": null, "lora_use_virtual_experts": false, "lora_strict_loading": false, "lora_drain_wait_threshold": 0.0, "enable_two_batch_overlap": false, "enable_single_batch_overlap": false, "tbo_token_distribution_threshold": 0.48, "cpu_offload_gb": 0, "offload_group_size": -1, "offload_num_in_group": 1, "offload_prefetch_step": 1, "offload_mode": "cpu", "enable_lmcache": false, "lmcache_config_file": null, "enable_flexkv": false, "flexkv_config_file": null, "kt_weight_path": null, "kt_method": "AMXINT4", "kt_cpuinfer": null, "kt_threadpool_count": 2, "kt_num_gpu_experts": null, "kt_max_deferred_experts_per_token": null, "dllm_algorithm": null, "dllm_algorithm_config": null, "dllm_fdfo": true, "disaggregation_mode": "decode", "disaggregation_transfer_backend": "mooncake", "disaggregation_bootstrap_port": 28800, "disaggregation_ib_device": "mlx5_0,mlx5_1,mlx5_2,mlx5_3", "disaggregation_decode_enable_radix_cache": false, "disaggregation_decode_enable_offload_kvcache": false, "num_reserved_decode_tokens": 512, "disaggregation_decode_extra_slots": 0, "disaggregation_decode_polling_interval": 1, "optimistic_prefill_attempts": 0, "encoder_only": false, "language_only": false, "encoder_transfer_backend": "zmq_to_tokenizer", "encoder_urls": [], "encoder_bootstrap_port": 8997, "encoder_register_urls": [], "enable_adaptive_dispatch_to_encoder": false, "enable_pdmux": false, "pdmux_config_path": null, "sm_group_num": 8, "custom_weight_loader": [], "weight_loader_disable_mmap": false, "weight_loader_prefetch_checkpoints": false, "weight_loader_prefetch_num_threads": 4, "weight_loader_drop_cache_after_load": false, "remote_instance_weight_loader_seed_instance_ip": null, "remote_instance_weight_loader_seed_instance_service_port": null, "remote_instance_weight_loader_send_weights_group_ports": null, "remote_instance_weight_loader_backend": "nccl", "remote_instance_weight_loader_start_seed_via_transfer_engine": false, "engine_info_bootstrap_port": 6789, "modelexpress_config": null, "download_dir": null, "model_checksum": null, "delete_ckpt_after_loading": false, "decrypted_config_file": null, "decrypted_draft_config_file": null, "checkpoint_engine_wait_weights_before_ready": false, "enable_prefill_delayer": false, "prefill_delayer_max_delay_passes": 30, "prefill_delayer_token_usage_low_watermark": null, "prefill_delayer_forward_passes_buckets": null, "prefill_delayer_wait_seconds_buckets": null, "prefill_delayer_queue_min_ratio": null, "prefill_delayer_max_delay_ms": null, "min_free_slots_delay": null, "enable_deterministic_inference": false, "rl_on_policy_target": null, "kv_canary": "none", "kv_canary_real_data": "none", "kv_canary_sweep_interval": 0, "enable_dynamic_batch_tokenizer": false, "dynamic_batch_tokenizer_batch_size": 32, "dynamic_batch_tokenizer_batch_timeout": 0.002, "enable_tokenizer_batch_encode": false, "disable_tokenizer_batch_decode": false, "debug_tensor_dump_output_folder": null, "debug_tensor_dump_layers": null, "debug_tensor_dump_input_file": null, "enable_memory_saver": false, "enable_weights_cpu_backup": false, "enable_draft_weights_cpu_backup": false, "enable_custom_logit_processor": false, "enable_return_hidden_states": false, "enable_return_routed_experts": false, "enable_return_indexer_topk": false, "disable_outlines_disk_cache": false, "enable_mis": false, "weight_cache_mode": "off", "weight_cache_socket": null, "weight_cache_timeout": 1800, "forward_hooks": null, "msprobe_dump_config": null, "_resolved_overrides": [["_sampling_backend_default", {"sampling_backend": "flashinfer"}], ["_attention_backend_default", {"attention_backend": "triton"}], ["_data_parallelism_defaults", {"enable_dp_attention": false, "enable_dp_lm_head": false}], ["_speculative_moe_runner_default", {"speculative_moe_runner_backend": "marlin"}]], "grpc_worker_threads": 4, "_quantization_explicitly_unset": false, "_cuda_graph_config_locked": [["decode", "max_bs"]], "_declarations_materialized": true, "_runtime_mutations": [["model_runner.chunked_prefix_cache_gate", {"disable_chunked_prefix_cache": true}], ["mamba_pool.memory_budget", {"max_mamba_cache_size": 181}]], "_in_override": false, "_mx_config_cache": {}, "max_speculative_num_draft_tokens": null, "last_gen_throughput": 26.412623015816244, "memory_usage": {"weight": 58.57, "kvcache": 11.34, "token_capacity": 440256, "graph": 0.6}, "effective_max_running_requests_per_dp": 181}], "prefill": [{"model_path": "/data/hf_models/Kimi-K3", "tokenizer_path": "/data/hf_models/Kimi-K3", "tokenizer_mode": "auto", "tokenizer_backend": "huggingface", "tokenizer_worker_num": 1, "detokenizer_worker_num": 1, "skip_tokenizer_init": false, "load_format": "auto", "model_loader_extra_config": "{}", "trust_remote_code": true, "context_length": null, "is_embedding": false, "enable_multimodal": null, "revision": null, "model_impl": "auto", "model_config_parser": "auto", "json_model_override_args": "{}", "dtype": "auto", "quantization": null, "quantization_param_path": null, "kv_cache_dtype": "auto", "enable_fp32_lm_head": false, "modelopt_quant": null, "modelopt_checkpoint_restore_path": null, "modelopt_checkpoint_save_path": null, "modelopt_export_path": null, "quantize_and_serve": false, "rl_quant_profile": null, "enable_tf32_matmul": false, "mem_fraction_static": 0.88, "max_running_requests": null, "max_queued_requests": null, "max_total_tokens": null, "chunked_prefill_size": 8192, "enable_dynamic_chunking": false, "max_prefill_tokens": 16384, "prefill_max_requests": null, "schedule_policy": "fcfs", "enable_priority_scheduling": false, "disable_priority_preemption": false, "default_priority_value": null, "abort_on_priority_when_disabled": false, "schedule_low_priority_values_first": false, "priority_scheduling_preemption_threshold": 10, "retraction_policy": "length", "schedule_conservativeness": 1.0, "page_size": 64, "swa_full_tokens_ratio": 0.8, "disable_hybrid_swa_memory": false, "radix_eviction_policy": "lru", "prefill_only_disable_kv_cache": false, "disable_radix_cache": true, "enable_page_major_kv_layout": false, "enable_unified_memory": false, "disable_chunked_prefix_cache": false, "disable_overlap_schedule": true, "num_continuous_decode_steps": 1, "scheduler_recv_interval": 1, "enable_mixed_chunk": false, "nccl_port": null, "dist_timeout": 3600, "dist_init_addr": "174.1.60.1:20000", "nnodes": 4, "node_rank": 0, "tp_size": 4, "dcp_size": 1, "pp_size": 8, "pp_max_micro_batch_size": null, "pp_async_batch_depth": 0, "dp_size": 1, "load_balance_method": "follow_bootstrap_room", "attn_cp_size": 1, "moe_dp_size": 1, "dcp_comm_backend": "ag_rs", "dcp_replicate_q_proj": null, "dwdp_size": 1, "enable_prefill_cp": false, "cp_strategy": null, "enable_dsa_cache_layer_split": false, "enable_dsa_prefill_context_parallel": false, "dsa_prefill_cp_mode": "round-robin-split", "enable_prefill_context_parallel": false, "prefill_cp_mode": "in-seq-split", "enable_cp_decode_attn_tp": false, "enable_dp_attention": false, "enable_dp_attention_local_control_broadcast": false, "enable_dp_lm_head": false, "enable_attn_tp_input_scattered": false, "disable_attn_tp_gather": false, "enable_p2p_check": false, "device": "cuda", "base_gpu_id": 0, "gpu_id_step": 1, "random_seed": 984607766, "watchdog_timeout": 300, "soft_watchdog_timeout": null, "sleep_on_idle": false, "use_ray": false, "custom_sigquit_handler": null, "numa_node": null, "gc_threshold": null, "host": "0.0.0.0", "port": 30000, "fastapi_root_path": "", "smg_grpc_mode": false, "grpc_mode": false, "grpc_port": null, "sidecar": null, "sidecar_args": null, "skip_server_warmup": false, "warmups": null, "enable_http2": false, "ssl_keyfile": null, "ssl_certfile": null, "ssl_ca_certs": null, "ssl_keyfile_password": null, "enable_ssl_refresh": false, "api_key": null, "admin_api_key": null, "served_model_name": "kimi-k3", "weight_version": "default", "chat_template": null, "hf_chat_template_name": null, "completion_template": null, "file_storage_path": "sglang_storage", "enable_cache_report": false, "reasoning_parser": null, "default_chat_template_kwargs": null, "strip_thinking_cache": false, "enable_strict_thinking": false, "tool_call_parser": null, "tool_server": null, "sampling_defaults": "model", "asr_max_buffer_seconds": 60, "asr_max_concurrent_sessions": 32, "preferred_sampling_params": null, "allow_auto_truncate": false, "stream_interval": 1, "batch_notify_size": 16, "stream_response_default_include_usage": false, "incremental_streaming_output": false, "enable_streaming_session": false, "enable_session_radix_cache": false, "log_level": "info", "log_level_http": null, "log_requests": false, "log_requests_level": 2, "log_requests_format": "text", "log_requests_target": null, "uvicorn_access_log_exclude_prefixes": [], "crash_dump_folder": null, "show_time_cost": false, "enable_metrics": false, "smg_http_sidecar_port": null, "enable_mfu_metrics": false, "enable_metrics_for_all_schedulers": false, "load_snapshot_publish_interval": 15, "tokenizer_metrics_custom_labels_header": "x-custom-labels", "tokenizer_metrics_allowed_custom_labels": null, "extra_metric_labels": null, "bucket_time_to_first_token": null, "bucket_inter_token_latency": null, "bucket_e2e_request_latency": null, "prompt_tokens_buckets": null, "generation_tokens_buckets": null, "gc_warning_threshold_secs": 0.0, "decode_log_interval": 40, "enable_request_time_stats_logging": false, "kv_events_config": null, "enable_forward_pass_metrics": false, "forward_pass_metrics_worker_id": "", "forward_pass_metrics_ipc_name": null, "enable_trace": false, "trace_modules": "request", "otlp_traces_endpoint": "localhost:4317", "export_metrics_to_file": false, "export_metrics_to_file_dir": null, "stat_loggers": null, "constrained_json_whitespace_pattern": null, "constrained_json_disable_any_whitespace": false, "attention_backend": "triton", "decode_attention_backend": null, "prefill_attention_backend": null, "sampling_backend": "flashinfer", "grammar_backend": "xgrammar", "radix_cache_backend": null, "mm_attention_backend": null, "fp8_gemm_runner_backend": "auto", "fp4_gemm_runner_backend": "auto", "bf16_gemm_backend": "auto", "dsa_prefill_backend": null, "dsa_decode_backend": null, "dsa_paged_mqa_logits_backend": "auto", "dsa_topk_backend": "sgl-kernel", "disable_flashinfer_autotune": false, "flashinfer_autotune_skip_ops": null, "mamba_backend": "triton", "cuda_graph_config": {"decode": {"backend": "disabled", "max_bs": 16, "bs": [1, 2, 4, 8, 12, 16], "tc_compiler": "eager", "full_prefill_max_req": null}, "prefill": {"backend": "disabled", "max_bs": 2048, "bs": [4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], "tc_compiler": "eager", "full_prefill_max_req": null}}, "cuda_graph_backend_decode": null, "cuda_graph_backend_prefill": null, "cuda_graph_max_bs_decode": 16, "cuda_graph_max_bs_prefill": null, "cuda_graph_bs_decode": null, "cuda_graph_bs_prefill": null, "cuda_graph_tc_compiler": null, "disable_prefill_cuda_graph": false, "disable_decode_cuda_graph": false, "disable_cuda_graph": false, "disable_cuda_graph_padding": false, "enable_profile_cuda_graph": false, "enable_cudagraph_gc": false, "debug_cuda_graph": false, "enable_layerwise_nvtx_marker": false, "enable_nccl_nvls": false, "enable_symm_mem": false, "triton_attention_reduce_in_fp32": false, "triton_attention_num_kv_splits": 8, "triton_attention_split_tile_size": null, "flashinfer_mla_disable_ragged": false, "enable_fused_qk_norm_rope": false, "enable_precise_embedding_interpolation": false, "enable_fused_moe_sum_all_reduce": false, "enable_deepseek_v4_fp4_indexer": false, "disable_custom_all_reduce": false, "enable_mscclpp": false, "enable_torch_symm_mem": false, "enable_scattered_sconv": false, "pre_warm_nccl": false, "enable_quant_communications": false, "enable_flashinfer_allreduce_fusion": false, "enforce_disable_flashinfer_allreduce_fusion": false, "flashinfer_allreduce_fusion_backend": null, "enable_aiter_allreduce_fusion": false, "enable_torch_compile": false, "enable_torch_compile_debug_mode": false, "torch_compile_max_bs": 32, "torchao_config": "", "speculative_algorithm": null, "speculative_draft_model_path": null, "speculative_draft_model_revision": null, "speculative_draft_load_format": null, "speculative_num_steps": null, "speculative_eagle_topk": null, "speculative_num_draft_tokens": null, "speculative_dflash_block_size": null, "speculative_dspark_block_size": null, "speculative_dspark_sps_table_path": null, "speculative_dspark_confidence_sts_path": null, "speculative_dspark_align_verify_tokens_to_graph_tier": false, "speculative_accept_threshold_single": 1.0, "speculative_accept_threshold_acc": 1.0, "speculative_use_rejection_sampling": false, "speculative_token_map": null, "speculative_attention_mode": "prefill", "speculative_draft_attention_backend": null, "speculative_draft_window_size": null, "speculative_moe_runner_backend": "flashinfer_mxfp4", "speculative_moe_a2a_backend": null, "speculative_draft_model_quantization": null, "speculative_skip_dp_mlp_sync": false, "enable_multi_layer_eagle": false, "speculative_adaptive": false, "speculative_adaptive_config": null, "decoupled_spec_bind_endpoint": null, "decoupled_spec_connect_endpoints": null, "decoupled_spec_rank": null, "decoupled_spec_role": "null", "spec_trace_dir": null, "speculative_ngram_min_bfs_breadth": 1, "speculative_ngram_max_bfs_breadth": 10, "speculative_ngram_match_type": "BFS", "speculative_ngram_max_trie_depth": 18, "speculative_ngram_capacity": 10000000, "speculative_ngram_external_corpus_path": null, "speculative_ngram_external_sam_budget": 0, "speculative_ngram_external_corpus_max_tokens": 10000000, "ep_size": 4, "moe_a2a_backend": "none", "moe_runner_backend": "flashinfer_mxfp4", "flashinfer_mxfp4_moe_precision": "default", "deepep_mode": "auto", "fuseep_mode": 2, "deepep_dispatcher_output_dtype": "auto", "ep_num_redundant_experts": 0, "ep_dispatch_algorithm": null, "init_expert_location": "trivial", "enable_eplb": false, "eplb_algorithm": "auto", "eplb_rebalance_num_iterations": 1000, "eplb_rebalance_layers_per_chunk": null, "eplb_min_rebalancing_utilization_threshold": 1.0, "expert_distribution_recorder_mode": null, "expert_distribution_recorder_buffer_size": 1000, "enable_expert_distribution_metrics": false, "deepep_config": null, "moe_dense_tp_size": null, "elastic_ep_backend": null, "enable_elastic_expert_backup": false, "mooncake_ib_device": null, "enable_waterfill": false, "ep_join_mode": null, "ep_join_rank_offset": 0, "elastic_ep_initial_size": null, "max_ep_size": null, "elastic_ep_scale_timeout": 600, "elastic_ep_rejoin": false, "disable_flashinfer_cutlass_moe_fp4_allgather": false, "disable_shared_experts_fusion": false, "enforce_shared_experts_fusion": false, "max_mamba_cache_size": null, "mamba_ssm_dtype": null, "mamba_max_states_per_path": -1, "enable_mamba_cache_stochastic_rounding": false, "mamba_cache_philox_rounds": 0, "mamba_full_memory_ratio": 0.36, "mamba_radix_cache_strategy": "extra_buffer_lazy", "uses_mamba_radix_cache": false, "mamba_track_interval": 256, "enable_int8_mamba_checkpoint": false, "int8_mamba_ckpt_size": null, "linear_attn_backend": "triton", "linear_attn_decode_backend": null, "linear_attn_prefill_backend": null, "linear_attn_verify_backend": null, "enable_linear_replayssm": false, "linear_replayssm_cache_len": 16, "enable_linear_replayssm_spec": false, "enable_hierarchical_cache": false, "hicache_ratio": 2.0, "hicache_size": 0, "hicache_write_policy": "write_through", "hicache_io_backend": "kernel", "hicache_mem_layout": "page_first", "hicache_storage_backend": null, "hicache_storage_prefetch_policy": "timeout", "hicache_storage_backend_extra_config": null, "enable_hisparse": false, "hisparse_config": null, "enable_broadcast_mm_inputs_process": false, "enable_prefix_mm_cache": false, "mm_enable_dp_encoder": false, "mm_process_config": {}, "mm_processor_worker_num": 0, "mm_io_worker_num": 0, "limit_mm_data_per_request": null, "enable_mm_global_cache": false, "disable_fast_image_processor": false, "mm_feature_transport": "cpu", "keep_mm_feature_on_device": false, "enable_lora": null, "enable_lora_overlap_loading": null, "max_lora_rank": null, "lora_target_modules": null, "lora_paths": null, "max_loaded_loras": null, "max_loras_per_batch": 8, "lora_eviction_policy": "lru", "lora_backend": "csgmv", "max_lora_chunk_size": 16, "experts_shared_outer_loras": null, "lora_use_virtual_experts": false, "lora_strict_loading": false, "lora_drain_wait_threshold": 0.0, "enable_two_batch_overlap": false, "enable_single_batch_overlap": false, "tbo_token_distribution_threshold": 0.48, "cpu_offload_gb": 0, "offload_group_size": -1, "offload_num_in_group": 1, "offload_prefetch_step": 1, "offload_mode": "cpu", "enable_lmcache": false, "lmcache_config_file": null, "enable_flexkv": false, "flexkv_config_file": null, "kt_weight_path": null, "kt_method": "AMXINT4", "kt_cpuinfer": null, "kt_threadpool_count": 2, "kt_num_gpu_experts": null, "kt_max_deferred_experts_per_token": null, "dllm_algorithm": null, "dllm_algorithm_config": null, "dllm_fdfo": true, "disaggregation_mode": "prefill", "disaggregation_transfer_backend": "mooncake", "disaggregation_bootstrap_port": 28800, "disaggregation_ib_device": "mlx5_0,mlx5_1,mlx5_2,mlx5_3", "disaggregation_decode_enable_radix_cache": false, "disaggregation_decode_enable_offload_kvcache": false, "num_reserved_decode_tokens": 512, "disaggregation_decode_extra_slots": null, "disaggregation_decode_polling_interval": 1, "optimistic_prefill_attempts": 0, "encoder_only": false, "language_only": false, "encoder_transfer_backend": "zmq_to_tokenizer", "encoder_urls": [], "encoder_bootstrap_port": 8997, "encoder_register_urls": [], "enable_adaptive_dispatch_to_encoder": false, "enable_pdmux": false, "pdmux_config_path": null, "sm_group_num": 8, "custom_weight_loader": [], "weight_loader_disable_mmap": false, "weight_loader_prefetch_checkpoints": false, "weight_loader_prefetch_num_threads": 4, "weight_loader_drop_cache_after_load": false, "remote_instance_weight_loader_seed_instance_ip": null, "remote_instance_weight_loader_seed_instance_service_port": null, "remote_instance_weight_loader_send_weights_group_ports": null, "remote_instance_weight_loader_backend": "nccl", "remote_instance_weight_loader_start_seed_via_transfer_engine": false, "engine_info_bootstrap_port": 6789, "modelexpress_config": null, "download_dir": null, "model_checksum": null, "delete_ckpt_after_loading": false, "decrypted_config_file": null, "decrypted_draft_config_file": null, "checkpoint_engine_wait_weights_before_ready": false, "enable_prefill_delayer": false, "prefill_delayer_max_delay_passes": 30, "prefill_delayer_token_usage_low_watermark": null, "prefill_delayer_forward_passes_buckets": null, "prefill_delayer_wait_seconds_buckets": null, "prefill_delayer_queue_min_ratio": null, "prefill_delayer_max_delay_ms": null, "min_free_slots_delay": null, "enable_deterministic_inference": false, "rl_on_policy_target": null, "kv_canary": "none", "kv_canary_real_data": "none", "kv_canary_sweep_interval": 0, "enable_dynamic_batch_tokenizer": false, "dynamic_batch_tokenizer_batch_size": 32, "dynamic_batch_tokenizer_batch_timeout": 0.002, "enable_tokenizer_batch_encode": false, "disable_tokenizer_batch_decode": false, "debug_tensor_dump_output_folder": null, "debug_tensor_dump_layers": null, "debug_tensor_dump_input_file": null, "enable_memory_saver": false, "enable_weights_cpu_backup": false, "enable_draft_weights_cpu_backup": false, "enable_custom_logit_processor": false, "enable_return_hidden_states": false, "enable_return_routed_experts": false, "enable_return_indexer_topk": false, "disable_outlines_disk_cache": false, "enable_mis": false, "weight_cache_mode": "off", "weight_cache_socket": null, "weight_cache_timeout": 1800, "forward_hooks": null, "msprobe_dump_config": null, "status": "ready", "max_total_num_tokens": 3801408, "max_req_input_len": 1048570, "internal_states": [{"model_path": "/data/hf_models/Kimi-K3", "tokenizer_path": "/data/hf_models/Kimi-K3", "tokenizer_mode": "auto", "tokenizer_backend": "huggingface", "tokenizer_worker_num": 1, "detokenizer_worker_num": 1, "skip_tokenizer_init": false, "load_format": "auto", "model_loader_extra_config": "{}", "trust_remote_code": true, "context_length": null, "is_embedding": false, "enable_multimodal": null, "revision": null, "model_impl": "auto", "model_config_parser": "auto", "json_model_override_args": "{}", "dtype": "auto", "quantization": null, "quantization_param_path": null, "kv_cache_dtype": "auto", "enable_fp32_lm_head": false, "modelopt_quant": null, "modelopt_checkpoint_restore_path": null, "modelopt_checkpoint_save_path": null, "modelopt_export_path": null, "quantize_and_serve": false, "rl_quant_profile": null, "enable_tf32_matmul": false, "mem_fraction_static": 0.88, "max_running_requests": null, "max_queued_requests": null, "max_total_tokens": null, "chunked_prefill_size": 8192, "enable_dynamic_chunking": false, "max_prefill_tokens": 16384, "prefill_max_requests": null, "schedule_policy": "fcfs", "enable_priority_scheduling": false, "disable_priority_preemption": false, "default_priority_value": null, "abort_on_priority_when_disabled": false, "schedule_low_priority_values_first": false, "priority_scheduling_preemption_threshold": 10, "retraction_policy": "length", "schedule_conservativeness": 1.0, "page_size": 64, "swa_full_tokens_ratio": 0.8, "disable_hybrid_swa_memory": false, "radix_eviction_policy": "lru", "prefill_only_disable_kv_cache": false, "disable_radix_cache": true, "enable_page_major_kv_layout": false, "enable_unified_memory": false, "disable_chunked_prefix_cache": true, "disable_overlap_schedule": true, "num_continuous_decode_steps": 1, "scheduler_recv_interval": 1, "enable_mixed_chunk": false, "nccl_port": null, "dist_timeout": 3600, "dist_init_addr": "174.1.60.1:20000", "nnodes": 4, "node_rank": 0, "tp_size": 4, "dcp_size": 1, "pp_size": 8, "pp_max_micro_batch_size": 6, "pp_async_batch_depth": 0, "dp_size": 1, "load_balance_method": "follow_bootstrap_room", "attn_cp_size": 1, "moe_dp_size": 1, "dcp_comm_backend": "ag_rs", "dcp_replicate_q_proj": null, "dwdp_size": 1, "enable_prefill_cp": false, "cp_strategy": null, "enable_dsa_cache_layer_split": false, "enable_dsa_prefill_context_parallel": false, "dsa_prefill_cp_mode": "round-robin-split", "enable_prefill_context_parallel": false, "prefill_cp_mode": "in-seq-split", "enable_cp_decode_attn_tp": false, "enable_dp_attention": false, "enable_dp_attention_local_control_broadcast": false, "enable_dp_lm_head": false, "enable_attn_tp_input_scattered": false, "disable_attn_tp_gather": false, "enable_p2p_check": false, "device": "cuda", "base_gpu_id": 0, "gpu_id_step": 1, "random_seed": 984607766, "watchdog_timeout": 300, "soft_watchdog_timeout": null, "sleep_on_idle": false, "use_ray": false, "numa_node": null, "gc_threshold": null, "host": "0.0.0.0", "port": 30000, "fastapi_root_path": "", "smg_grpc_mode": false, "grpc_mode": false, "grpc_port": null, "sidecar": null, "sidecar_args": null, "skip_server_warmup": false, "warmups": null, "enable_http2": false, "ssl_keyfile": null, "ssl_certfile": null, "ssl_ca_certs": null, "ssl_keyfile_password": null, "enable_ssl_refresh": false, "api_key": null, "admin_api_key": null, "served_model_name": "kimi-k3", "weight_version": "default", "chat_template": null, "hf_chat_template_name": null, "completion_template": null, "file_storage_path": "sglang_storage", "enable_cache_report": false, "reasoning_parser": null, "default_chat_template_kwargs": null, "strip_thinking_cache": false, "enable_strict_thinking": false, "tool_call_parser": null, "tool_server": null, "sampling_defaults": "model", "asr_max_buffer_seconds": 60, "asr_max_concurrent_sessions": 32, "preferred_sampling_params": null, "allow_auto_truncate": false, "stream_interval": 1, "batch_notify_size": 16, "stream_response_default_include_usage": false, "incremental_streaming_output": false, "enable_streaming_session": false, "enable_session_radix_cache": false, "log_level": "info", "log_level_http": null, "log_requests": false, "log_requests_level": 2, "log_requests_format": "text", "log_requests_target": null, "uvicorn_access_log_exclude_prefixes": [], "crash_dump_folder": null, "show_time_cost": false, "enable_metrics": false, "smg_http_sidecar_port": null, "enable_mfu_metrics": false, "enable_metrics_for_all_schedulers": false, "load_snapshot_publish_interval": 15, "tokenizer_metrics_custom_labels_header": "x-custom-labels", "tokenizer_metrics_allowed_custom_labels": null, "extra_metric_labels": null, "bucket_time_to_first_token": null, "bucket_inter_token_latency": null, "bucket_e2e_request_latency": null, "prompt_tokens_buckets": null, "generation_tokens_buckets": null, "gc_warning_threshold_secs": 0.0, "decode_log_interval": 40, "enable_request_time_stats_logging": false, "kv_events_config": null, "enable_forward_pass_metrics": false, "forward_pass_metrics_worker_id": "", "forward_pass_metrics_ipc_name": null, "enable_trace": false, "trace_modules": "request", "otlp_traces_endpoint": "localhost:4317", "export_metrics_to_file": false, "export_metrics_to_file_dir": null, "stat_loggers": null, "constrained_json_whitespace_pattern": null, "constrained_json_disable_any_whitespace": false, "attention_backend": "triton", "decode_attention_backend": null, "prefill_attention_backend": null, "sampling_backend": "flashinfer", "grammar_backend": "xgrammar", "radix_cache_backend": null, "mm_attention_backend": null, "fp8_gemm_runner_backend": "auto", "fp4_gemm_runner_backend": "auto", "bf16_gemm_backend": "auto", "dsa_prefill_backend": null, "dsa_decode_backend": null, "dsa_paged_mqa_logits_backend": "auto", "dsa_topk_backend": "sgl-kernel", "disable_flashinfer_autotune": false, "flashinfer_autotune_skip_ops": null, "mamba_backend": "triton", "cuda_graph_config": {"decode": {"backend": "disabled", "max_bs": 16, "bs": [1, 2, 4, 8, 12, 16], "tc_compiler": "eager", "full_prefill_max_req": null}, "prefill": {"backend": "disabled", "max_bs": 2048, "bs": [4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], "tc_compiler": "eager", "full_prefill_max_req": null}}, "cuda_graph_backend_decode": null, "cuda_graph_backend_prefill": null, "cuda_graph_max_bs_decode": 16, "cuda_graph_max_bs_prefill": null, "cuda_graph_bs_decode": null, "cuda_graph_bs_prefill": null, "cuda_graph_tc_compiler": null, "disable_prefill_cuda_graph": false, "disable_decode_cuda_graph": false, "disable_cuda_graph": false, "disable_cuda_graph_padding": false, "enable_profile_cuda_graph": false, "enable_cudagraph_gc": false, "debug_cuda_graph": false, "enable_layerwise_nvtx_marker": false, "enable_nccl_nvls": false, "enable_symm_mem": false, "triton_attention_reduce_in_fp32": false, "triton_attention_num_kv_splits": 8, "triton_attention_split_tile_size": null, "flashinfer_mla_disable_ragged": false, "enable_fused_qk_norm_rope": false, "enable_precise_embedding_interpolation": false, "enable_fused_moe_sum_all_reduce": false, "enable_deepseek_v4_fp4_indexer": false, "disable_custom_all_reduce": false, "enable_mscclpp": false, "enable_torch_symm_mem": false, "enable_scattered_sconv": false, "pre_warm_nccl": false, "enable_quant_communications": false, "enable_flashinfer_allreduce_fusion": false, "enforce_disable_flashinfer_allreduce_fusion": false, "flashinfer_allreduce_fusion_backend": null, "enable_aiter_allreduce_fusion": false, "enable_torch_compile": false, "enable_torch_compile_debug_mode": false, "torch_compile_max_bs": 32, "torchao_config": "", "speculative_algorithm": null, "speculative_draft_model_path": null, "speculative_draft_model_revision": null, "speculative_draft_load_format": null, "speculative_num_steps": null, "speculative_eagle_topk": null, "speculative_num_draft_tokens": null, "speculative_dflash_block_size": null, "speculative_dspark_block_size": null, "speculative_dspark_sps_table_path": null, "speculative_dspark_confidence_sts_path": null, "speculative_dspark_align_verify_tokens_to_graph_tier": false, "speculative_accept_threshold_single": 1.0, "speculative_accept_threshold_acc": 1.0, "speculative_use_rejection_sampling": false, "speculative_token_map": null, "speculative_attention_mode": "prefill", "speculative_draft_attention_backend": null, "speculative_draft_window_size": null, "speculative_moe_runner_backend": "flashinfer_mxfp4", "speculative_moe_a2a_backend": null, "speculative_draft_model_quantization": null, "speculative_skip_dp_mlp_sync": false, "enable_multi_layer_eagle": false, "speculative_adaptive": false, "speculative_adaptive_config": null, "decoupled_spec_bind_endpoint": null, "decoupled_spec_connect_endpoints": null, "decoupled_spec_rank": null, "decoupled_spec_role": "null", "spec_trace_dir": null, "speculative_ngram_min_bfs_breadth": 1, "speculative_ngram_max_bfs_breadth": 10, "speculative_ngram_match_type": "BFS", "speculative_ngram_max_trie_depth": 18, "speculative_ngram_capacity": 10000000, "speculative_ngram_external_corpus_path": null, "speculative_ngram_external_sam_budget": 0, "speculative_ngram_external_corpus_max_tokens": 10000000, "ep_size": 4, "moe_a2a_backend": "none", "moe_runner_backend": "flashinfer_mxfp4", "flashinfer_mxfp4_moe_precision": "default", "deepep_mode": "auto", "fuseep_mode": 2, "deepep_dispatcher_output_dtype": "auto", "ep_num_redundant_experts": 0, "ep_dispatch_algorithm": null, "init_expert_location": "trivial", "enable_eplb": false, "eplb_algorithm": "auto", "eplb_rebalance_num_iterations": 1000, "eplb_rebalance_layers_per_chunk": null, "eplb_min_rebalancing_utilization_threshold": 1.0, "expert_distribution_recorder_mode": null, "expert_distribution_recorder_buffer_size": 1000, "enable_expert_distribution_metrics": false, "deepep_config": null, "moe_dense_tp_size": null, "elastic_ep_backend": null, "enable_elastic_expert_backup": false, "mooncake_ib_device": null, "enable_waterfill": false, "ep_join_mode": null, "ep_join_rank_offset": 0, "elastic_ep_initial_size": null, "max_ep_size": null, "elastic_ep_scale_timeout": 600, "elastic_ep_rejoin": false, "disable_flashinfer_cutlass_moe_fp4_allgather": false, "disable_shared_experts_fusion": false, "enforce_shared_experts_fusion": false, "max_mamba_cache_size": 54, "mamba_ssm_dtype": null, "mamba_max_states_per_path": -1, "enable_mamba_cache_stochastic_rounding": false, "mamba_cache_philox_rounds": 0, "mamba_full_memory_ratio": 0.36, "mamba_radix_cache_strategy": "extra_buffer_lazy", "uses_mamba_radix_cache": false, "mamba_track_interval": 256, "enable_int8_mamba_checkpoint": false, "int8_mamba_ckpt_size": null, "linear_attn_backend": "triton", "linear_attn_decode_backend": null, "linear_attn_prefill_backend": null, "linear_attn_verify_backend": null, "enable_linear_replayssm": false, "linear_replayssm_cache_len": 16, "enable_linear_replayssm_spec": false, "enable_hierarchical_cache": false, "hicache_ratio": 2.0, "hicache_size": 0, "hicache_write_policy": "write_through", "hicache_io_backend": "kernel", "hicache_mem_layout": "page_first", "hicache_storage_backend": null, "hicache_storage_prefetch_policy": "timeout", "hicache_storage_backend_extra_config": null, "enable_hisparse": false, "hisparse_config": null, "enable_broadcast_mm_inputs_process": false, "enable_prefix_mm_cache": false, "mm_enable_dp_encoder": false, "mm_process_config": {}, "mm_processor_worker_num": 0, "mm_io_worker_num": 0, "limit_mm_data_per_request": null, "enable_mm_global_cache": false, "disable_fast_image_processor": false, "mm_feature_transport": "cpu", "keep_mm_feature_on_device": false, "enable_lora": null, "enable_lora_overlap_loading": null, "max_lora_rank": null, "lora_target_modules": null, "lora_paths": null, "max_loaded_loras": null, "max_loras_per_batch": 8, "lora_eviction_policy": "lru", "lora_backend": "csgmv", "max_lora_chunk_size": 16, "experts_shared_outer_loras": null, "lora_use_virtual_experts": false, "lora_strict_loading": false, "lora_drain_wait_threshold": 0.0, "enable_two_batch_overlap": false, "enable_single_batch_overlap": false, "tbo_token_distribution_threshold": 0.48, "cpu_offload_gb": 0, "offload_group_size": -1, "offload_num_in_group": 1, "offload_prefetch_step": 1, "offload_mode": "cpu", "enable_lmcache": false, "lmcache_config_file": null, "enable_flexkv": false, "flexkv_config_file": null, "kt_weight_path": null, "kt_method": "AMXINT4", "kt_cpuinfer": null, "kt_threadpool_count": 2, "kt_num_gpu_experts": null, "kt_max_deferred_experts_per_token": null, "dllm_algorithm": null, "dllm_algorithm_config": null, "dllm_fdfo": true, "disaggregation_mode": "prefill", "disaggregation_transfer_backend": "mooncake", "disaggregation_bootstrap_port": 28800, "disaggregation_ib_device": "mlx5_0,mlx5_1,mlx5_2,mlx5_3", "disaggregation_decode_enable_radix_cache": false, "disaggregation_decode_enable_offload_kvcache": false, "num_reserved_decode_tokens": 512, "disaggregation_decode_extra_slots": null, "disaggregation_decode_polling_interval": 1, "optimistic_prefill_attempts": 0, "encoder_only": false, "language_only": false, "encoder_transfer_backend": "zmq_to_tokenizer", "encoder_urls": [], "encoder_bootstrap_port": 8997, "encoder_register_urls": [], "enable_adaptive_dispatch_to_encoder": false, "enable_pdmux": false, "pdmux_config_path": null, "sm_group_num": 8, "custom_weight_loader": [], "weight_loader_disable_mmap": false, "weight_loader_prefetch_checkpoints": false, "weight_loader_prefetch_num_threads": 4, "weight_loader_drop_cache_after_load": false, "remote_instance_weight_loader_seed_instance_ip": null, "remote_instance_weight_loader_seed_instance_service_port": null, "remote_instance_weight_loader_send_weights_group_ports": null, "remote_instance_weight_loader_backend": "nccl", "remote_instance_weight_loader_start_seed_via_transfer_engine": false, "engine_info_bootstrap_port": 6789, "modelexpress_config": null, "download_dir": null, "model_checksum": null, "delete_ckpt_after_loading": false, "decrypted_config_file": null, "decrypted_draft_config_file": null, "checkpoint_engine_wait_weights_before_ready": false, "enable_prefill_delayer": false, "prefill_delayer_max_delay_passes": 30, "prefill_delayer_token_usage_low_watermark": null, "prefill_delayer_forward_passes_buckets": null, "prefill_delayer_wait_seconds_buckets": null, "prefill_delayer_queue_min_ratio": null, "prefill_delayer_max_delay_ms": null, "min_free_slots_delay": null, "enable_deterministic_inference": false, "rl_on_policy_target": null, "kv_canary": "none", "kv_canary_real_data": "none", "kv_canary_sweep_interval": 0, "enable_dynamic_batch_tokenizer": false, "dynamic_batch_tokenizer_batch_size": 32, "dynamic_batch_tokenizer_batch_timeout": 0.002, "enable_tokenizer_batch_encode": false, "disable_tokenizer_batch_decode": false, "debug_tensor_dump_output_folder": null, "debug_tensor_dump_layers": null, "debug_tensor_dump_input_file": null, "enable_memory_saver": false, "enable_weights_cpu_backup": false, "enable_draft_weights_cpu_backup": false, "enable_custom_logit_processor": false, "enable_return_hidden_states": false, "enable_return_routed_experts": false, "enable_return_indexer_topk": false, "disable_outlines_disk_cache": false, "enable_mis": false, "weight_cache_mode": "off", "weight_cache_socket": null, "weight_cache_timeout": 1800, "forward_hooks": null, "msprobe_dump_config": null, "_resolved_overrides": [["_sampling_backend_default", {"sampling_backend": "flashinfer"}], ["_attention_backend_default", {"attention_backend": "triton"}], ["_data_parallelism_defaults", {"enable_dp_attention": false, "enable_dp_lm_head": false}], ["_pipeline_parallel_overlap_disable", {"disable_overlap_schedule": true}], ["_speculative_moe_runner_default", {"speculative_moe_runner_backend": "flashinfer_mxfp4"}]], "grpc_worker_threads": 4, "_quantization_explicitly_unset": false, "_cuda_graph_config_locked": [["decode", "max_bs"]], "_declarations_materialized": true, "_runtime_mutations": [["model_runner.chunked_prefix_cache_gate", {"disable_chunked_prefix_cache": true}], ["mamba_pool.memory_budget", {"max_mamba_cache_size": 54}]], "_in_override": false, "_mx_config_cache": {}, "max_speculative_num_draft_tokens": null, "last_gen_throughput": 0.0, "memory_usage": {"weight": 43.09, "kvcache": 8.16, "token_capacity": 3801408, "graph": 0}, "effective_max_running_requests_per_dp": 54}], "version": "0.5.16", "kv_events": null}], "decode": [{"model_path": "/data/hf_models/Kimi-K3", "tokenizer_path": "/data/hf_models/Kimi-K3", "tokenizer_mode": "auto", "tokenizer_backend": "huggingface", "tokenizer_worker_num": 1, "detokenizer_worker_num": 1, "skip_tokenizer_init": false, "load_format": "auto", "model_loader_extra_config": "{}", "trust_remote_code": true, "context_length": null, "is_embedding": false, "enable_multimodal": null, "revision": null, "model_impl": "auto", "model_config_parser": "auto", "json_model_override_args": "{}", "dtype": "auto", "quantization": null, "quantization_param_path": null, "kv_cache_dtype": "auto", "enable_fp32_lm_head": false, "modelopt_quant": null, "modelopt_checkpoint_restore_path": null, "modelopt_checkpoint_save_path": null, "modelopt_export_path": null, "quantize_and_serve": false, "rl_quant_profile": null, "enable_tf32_matmul": false, "mem_fraction_static": 0.88, "max_running_requests": null, "max_queued_requests": null, "max_total_tokens": null, "chunked_prefill_size": 8192, "enable_dynamic_chunking": false, "max_prefill_tokens": 16384, "prefill_max_requests": null, "schedule_policy": "fcfs", "enable_priority_scheduling": false, "disable_priority_preemption": false, "default_priority_value": null, "abort_on_priority_when_disabled": false, "schedule_low_priority_values_first": false, "priority_scheduling_preemption_threshold": 10, "retraction_policy": "length", "schedule_conservativeness": 1.0, "page_size": 64, "swa_full_tokens_ratio": 0.8, "disable_hybrid_swa_memory": false, "radix_eviction_policy": "lru", "prefill_only_disable_kv_cache": false, "disable_radix_cache": true, "enable_page_major_kv_layout": false, "enable_unified_memory": false, "disable_chunked_prefix_cache": false, "disable_overlap_schedule": false, "num_continuous_decode_steps": 1, "scheduler_recv_interval": 1, "enable_mixed_chunk": false, "nccl_port": null, "dist_timeout": 3600, "dist_init_addr": "174.1.60.5:20000", "nnodes": 4, "node_rank": 0, "tp_size": 32, "dcp_size": 1, "pp_size": 1, "pp_max_micro_batch_size": null, "pp_async_batch_depth": 0, "dp_size": 1, "load_balance_method": "round_robin", "attn_cp_size": 1, "moe_dp_size": 1, "dcp_comm_backend": "ag_rs", "dcp_replicate_q_proj": null, "dwdp_size": 1, "enable_prefill_cp": false, "cp_strategy": null, "enable_dsa_cache_layer_split": false, "enable_dsa_prefill_context_parallel": false, "dsa_prefill_cp_mode": "round-robin-split", "enable_prefill_context_parallel": false, "prefill_cp_mode": "in-seq-split", "enable_cp_decode_attn_tp": false, "enable_dp_attention": false, "enable_dp_attention_local_control_broadcast": false, "enable_dp_lm_head": false, "enable_attn_tp_input_scattered": false, "disable_attn_tp_gather": false, "enable_p2p_check": false, "device": "cuda", "base_gpu_id": 0, "gpu_id_step": 1, "random_seed": 369235288, "watchdog_timeout": 300, "soft_watchdog_timeout": null, "sleep_on_idle": false, "use_ray": false, "custom_sigquit_handler": null, "numa_node": null, "gc_threshold": null, "host": "0.0.0.0", "port": 30000, "fastapi_root_path": "", "smg_grpc_mode": false, "grpc_mode": false, "grpc_port": null, "sidecar": null, "sidecar_args": null, "skip_server_warmup": false, "warmups": null, "enable_http2": false, "ssl_keyfile": null, "ssl_certfile": null, "ssl_ca_certs": null, "ssl_keyfile_password": null, "enable_ssl_refresh": false, "api_key": null, "admin_api_key": null, "served_model_name": "kimi-k3", "weight_version": "default", "chat_template": null, "hf_chat_template_name": null, "completion_template": null, "file_storage_path": "sglang_storage", "enable_cache_report": false, "reasoning_parser": null, "default_chat_template_kwargs": null, "strip_thinking_cache": false, "enable_strict_thinking": false, "tool_call_parser": null, "tool_server": null, "sampling_defaults": "model", "asr_max_buffer_seconds": 60, "asr_max_concurrent_sessions": 32, "preferred_sampling_params": null, "allow_auto_truncate": false, "stream_interval": 1, "batch_notify_size": 16, "stream_response_default_include_usage": false, "incremental_streaming_output": false, "enable_streaming_session": false, "enable_session_radix_cache": false, "log_level": "info", "log_level_http": null, "log_requests": false, "log_requests_level": 2, "log_requests_format": "text", "log_requests_target": null, "uvicorn_access_log_exclude_prefixes": [], "crash_dump_folder": null, "show_time_cost": false, "enable_metrics": false, "smg_http_sidecar_port": null, "enable_mfu_metrics": false, "enable_metrics_for_all_schedulers": false, "load_snapshot_publish_interval": 15, "tokenizer_metrics_custom_labels_header": "x-custom-labels", "tokenizer_metrics_allowed_custom_labels": null, "extra_metric_labels": null, "bucket_time_to_first_token": null, "bucket_inter_token_latency": null, "bucket_e2e_request_latency": null, "prompt_tokens_buckets": null, "generation_tokens_buckets": null, "gc_warning_threshold_secs": 0.0, "decode_log_interval": 40, "enable_request_time_stats_logging": false, "kv_events_config": null, "enable_forward_pass_metrics": false, "forward_pass_metrics_worker_id": "", "forward_pass_metrics_ipc_name": null, "enable_trace": false, "trace_modules": "request", "otlp_traces_endpoint": "localhost:4317", "export_metrics_to_file": false, "export_metrics_to_file_dir": null, "stat_loggers": null, "constrained_json_whitespace_pattern": null, "constrained_json_disable_any_whitespace": false, "attention_backend": "triton", "decode_attention_backend": null, "prefill_attention_backend": null, "sampling_backend": "flashinfer", "grammar_backend": "xgrammar", "radix_cache_backend": null, "mm_attention_backend": null, "fp8_gemm_runner_backend": "auto", "fp4_gemm_runner_backend": "auto", "bf16_gemm_backend": "auto", "dsa_prefill_backend": null, "dsa_decode_backend": null, "dsa_paged_mqa_logits_backend": "auto", "dsa_topk_backend": "sgl-kernel", "disable_flashinfer_autotune": false, "flashinfer_autotune_skip_ops": null, "mamba_backend": "triton", "cuda_graph_config": {"decode": {"backend": "full", "max_bs": 16, "bs": [1, 2, 4, 8, 12, 16], "tc_compiler": "eager", "full_prefill_max_req": null}, "prefill": {"backend": "disabled", "max_bs": 2048, "bs": [4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], "tc_compiler": "eager", "full_prefill_max_req": null}}, "cuda_graph_backend_decode": null, "cuda_graph_backend_prefill": null, "cuda_graph_max_bs_decode": 16, "cuda_graph_max_bs_prefill": null, "cuda_graph_bs_decode": null, "cuda_graph_bs_prefill": null, "cuda_graph_tc_compiler": null, "disable_prefill_cuda_graph": false, "disable_decode_cuda_graph": false, "disable_cuda_graph": false, "disable_cuda_graph_padding": false, "enable_profile_cuda_graph": false, "enable_cudagraph_gc": false, "debug_cuda_graph": false, "enable_layerwise_nvtx_marker": false, "enable_nccl_nvls": false, "enable_symm_mem": false, "triton_attention_reduce_in_fp32": false, "triton_attention_num_kv_splits": 8, "triton_attention_split_tile_size": null, "flashinfer_mla_disable_ragged": false, "enable_fused_qk_norm_rope": false, "enable_precise_embedding_interpolation": false, "enable_fused_moe_sum_all_reduce": false, "enable_deepseek_v4_fp4_indexer": false, "disable_custom_all_reduce": false, "enable_mscclpp": false, "enable_torch_symm_mem": false, "enable_scattered_sconv": false, "pre_warm_nccl": false, "enable_quant_communications": false, "enable_flashinfer_allreduce_fusion": false, "enforce_disable_flashinfer_allreduce_fusion": false, "flashinfer_allreduce_fusion_backend": null, "enable_aiter_allreduce_fusion": false, "enable_torch_compile": false, "enable_torch_compile_debug_mode": false, "torch_compile_max_bs": 32, "torchao_config": "", "speculative_algorithm": null, "speculative_draft_model_path": null, "speculative_draft_model_revision": null, "speculative_draft_load_format": null, "speculative_num_steps": null, "speculative_eagle_topk": null, "speculative_num_draft_tokens": null, "speculative_dflash_block_size": null, "speculative_dspark_block_size": null, "speculative_dspark_sps_table_path": null, "speculative_dspark_confidence_sts_path": null, "speculative_dspark_align_verify_tokens_to_graph_tier": false, "speculative_accept_threshold_single": 1.0, "speculative_accept_threshold_acc": 1.0, "speculative_use_rejection_sampling": false, "speculative_token_map": null, "speculative_attention_mode": "prefill", "speculative_draft_attention_backend": null, "speculative_draft_window_size": null, "speculative_moe_runner_backend": "marlin", "speculative_moe_a2a_backend": null, "speculative_draft_model_quantization": null, "speculative_skip_dp_mlp_sync": false, "enable_multi_layer_eagle": false, "speculative_adaptive": false, "speculative_adaptive_config": null, "decoupled_spec_bind_endpoint": null, "decoupled_spec_connect_endpoints": null, "decoupled_spec_rank": null, "decoupled_spec_role": "null", "spec_trace_dir": null, "speculative_ngram_min_bfs_breadth": 1, "speculative_ngram_max_bfs_breadth": 10, "speculative_ngram_match_type": "BFS", "speculative_ngram_max_trie_depth": 18, "speculative_ngram_capacity": 10000000, "speculative_ngram_external_corpus_path": null, "speculative_ngram_external_sam_budget": 0, "speculative_ngram_external_corpus_max_tokens": 10000000, "ep_size": 32, "moe_a2a_backend": "none", "moe_runner_backend": "marlin", "flashinfer_mxfp4_moe_precision": "default", "deepep_mode": "auto", "fuseep_mode": 2, "deepep_dispatcher_output_dtype": "auto", "ep_num_redundant_experts": 0, "ep_dispatch_algorithm": null, "init_expert_location": "trivial", "enable_eplb": false, "eplb_algorithm": "auto", "eplb_rebalance_num_iterations": 1000, "eplb_rebalance_layers_per_chunk": null, "eplb_min_rebalancing_utilization_threshold": 1.0, "expert_distribution_recorder_mode": null, "expert_distribution_recorder_buffer_size": 1000, "enable_expert_distribution_metrics": false, "deepep_config": null, "moe_dense_tp_size": null, "elastic_ep_backend": null, "enable_elastic_expert_backup": false, "mooncake_ib_device": null, "enable_waterfill": false, "ep_join_mode": null, "ep_join_rank_offset": 0, "elastic_ep_initial_size": null, "max_ep_size": null, "elastic_ep_scale_timeout": 600, "elastic_ep_rejoin": false, "disable_flashinfer_cutlass_moe_fp4_allgather": false, "disable_shared_experts_fusion": false, "enforce_shared_experts_fusion": false, "max_mamba_cache_size": null, "mamba_ssm_dtype": null, "mamba_max_states_per_path": -1, "enable_mamba_cache_stochastic_rounding": false, "mamba_cache_philox_rounds": 0, "mamba_full_memory_ratio": 0.21, "mamba_radix_cache_strategy": "extra_buffer_lazy", "uses_mamba_radix_cache": false, "mamba_track_interval": 256, "enable_int8_mamba_checkpoint": false, "int8_mamba_ckpt_size": null, "linear_attn_backend": "triton", "linear_attn_decode_backend": null, "linear_attn_prefill_backend": null, "linear_attn_verify_backend": null, "enable_linear_replayssm": false, "linear_replayssm_cache_len": 16, "enable_linear_replayssm_spec": false, "enable_hierarchical_cache": false, "hicache_ratio": 2.0, "hicache_size": 0, "hicache_write_policy": "write_through", "hicache_io_backend": "kernel", "hicache_mem_layout": "page_first", "hicache_storage_backend": null, "hicache_storage_prefetch_policy": "timeout", "hicache_storage_backend_extra_config": null, "enable_hisparse": false, "hisparse_config": null, "enable_broadcast_mm_inputs_process": false, "enable_prefix_mm_cache": false, "mm_enable_dp_encoder": false, "mm_process_config": {}, "mm_processor_worker_num": 0, "mm_io_worker_num": 0, "limit_mm_data_per_request": null, "enable_mm_global_cache": false, "disable_fast_image_processor": false, "mm_feature_transport": "cpu", "keep_mm_feature_on_device": false, "enable_lora": null, "enable_lora_overlap_loading": null, "max_lora_rank": null, "lora_target_modules": null, "lora_paths": null, "max_loaded_loras": null, "max_loras_per_batch": 8, "lora_eviction_policy": "lru", "lora_backend": "csgmv", "max_lora_chunk_size": 16, "experts_shared_outer_loras": null, "lora_use_virtual_experts": false, "lora_strict_loading": false, "lora_drain_wait_threshold": 0.0, "enable_two_batch_overlap": false, "enable_single_batch_overlap": false, "tbo_token_distribution_threshold": 0.48, "cpu_offload_gb": 0, "offload_group_size": -1, "offload_num_in_group": 1, "offload_prefetch_step": 1, "offload_mode": "cpu", "enable_lmcache": false, "lmcache_config_file": null, "enable_flexkv": false, "flexkv_config_file": null, "kt_weight_path": null, "kt_method": "AMXINT4", "kt_cpuinfer": null, "kt_threadpool_count": 2, "kt_num_gpu_experts": null, "kt_max_deferred_experts_per_token": null, "dllm_algorithm": null, "dllm_algorithm_config": null, "dllm_fdfo": true, "disaggregation_mode": "decode", "disaggregation_transfer_backend": "mooncake", "disaggregation_bootstrap_port": 28800, "disaggregation_ib_device": "mlx5_0,mlx5_1,mlx5_2,mlx5_3", "disaggregation_decode_enable_radix_cache": false, "disaggregation_decode_enable_offload_kvcache": false, "num_reserved_decode_tokens": 512, "disaggregation_decode_extra_slots": 0, "disaggregation_decode_polling_interval": 1, "optimistic_prefill_attempts": 0, "encoder_only": false, "language_only": false, "encoder_transfer_backend": "zmq_to_tokenizer", "encoder_urls": [], "encoder_bootstrap_port": 8997, "encoder_register_urls": [], "enable_adaptive_dispatch_to_encoder": false, "enable_pdmux": false, "pdmux_config_path": null, "sm_group_num": 8, "custom_weight_loader": [], "weight_loader_disable_mmap": false, "weight_loader_prefetch_checkpoints": false, "weight_loader_prefetch_num_threads": 4, "weight_loader_drop_cache_after_load": false, "remote_instance_weight_loader_seed_instance_ip": null, "remote_instance_weight_loader_seed_instance_service_port": null, "remote_instance_weight_loader_send_weights_group_ports": null, "remote_instance_weight_loader_backend": "nccl", "remote_instance_weight_loader_start_seed_via_transfer_engine": false, "engine_info_bootstrap_port": 6789, "modelexpress_config": null, "download_dir": null, "model_checksum": null, "delete_ckpt_after_loading": false, "decrypted_config_file": null, "decrypted_draft_config_file": null, "checkpoint_engine_wait_weights_before_ready": false, "enable_prefill_delayer": false, "prefill_delayer_max_delay_passes": 30, "prefill_delayer_token_usage_low_watermark": null, "prefill_delayer_forward_passes_buckets": null, "prefill_delayer_wait_seconds_buckets": null, "prefill_delayer_queue_min_ratio": null, "prefill_delayer_max_delay_ms": null, "min_free_slots_delay": null, "enable_deterministic_inference": false, "rl_on_policy_target": null, "kv_canary": "none", "kv_canary_real_data": "none", "kv_canary_sweep_interval": 0, "enable_dynamic_batch_tokenizer": false, "dynamic_batch_tokenizer_batch_size": 32, "dynamic_batch_tokenizer_batch_timeout": 0.002, "enable_tokenizer_batch_encode": false, "disable_tokenizer_batch_decode": false, "debug_tensor_dump_output_folder": null, "debug_tensor_dump_layers": null, "debug_tensor_dump_input_file": null, "enable_memory_saver": false, "enable_weights_cpu_backup": false, "enable_draft_weights_cpu_backup": false, "enable_custom_logit_processor": false, "enable_return_hidden_states": false, "enable_return_routed_experts": false, "enable_return_indexer_topk": false, "disable_outlines_disk_cache": false, "enable_mis": false, "weight_cache_mode": "off", "weight_cache_socket": null, "weight_cache_timeout": 1800, "forward_hooks": null, "msprobe_dump_config": null, "status": "ready", "max_total_num_tokens": 440256, "max_req_input_len": 440250, "internal_states": [{"model_path": "/data/hf_models/Kimi-K3", "tokenizer_path": "/data/hf_models/Kimi-K3", "tokenizer_mode": "auto", "tokenizer_backend": "huggingface", "tokenizer_worker_num": 1, "detokenizer_worker_num": 1, "skip_tokenizer_init": false, "load_format": "auto", "model_loader_extra_config": "{}", "trust_remote_code": true, "context_length": null, "is_embedding": false, "enable_multimodal": null, "revision": null, "model_impl": "auto", "model_config_parser": "auto", "json_model_override_args": "{}", "dtype": "auto", "quantization": null, "quantization_param_path": null, "kv_cache_dtype": "auto", "enable_fp32_lm_head": false, "modelopt_quant": null, "modelopt_checkpoint_restore_path": null, "modelopt_checkpoint_save_path": null, "modelopt_export_path": null, "quantize_and_serve": false, "rl_quant_profile": null, "enable_tf32_matmul": false, "mem_fraction_static": 0.88, "max_running_requests": null, "max_queued_requests": null, "max_total_tokens": null, "chunked_prefill_size": 8192, "enable_dynamic_chunking": false, "max_prefill_tokens": 16384, "prefill_max_requests": null, "schedule_policy": "fcfs", "enable_priority_scheduling": false, "disable_priority_preemption": false, "default_priority_value": null, "abort_on_priority_when_disabled": false, "schedule_low_priority_values_first": false, "priority_scheduling_preemption_threshold": 10, "retraction_policy": "length", "schedule_conservativeness": 1.0, "page_size": 64, "swa_full_tokens_ratio": 0.8, "disable_hybrid_swa_memory": false, "radix_eviction_policy": "lru", "prefill_only_disable_kv_cache": false, "disable_radix_cache": true, "enable_page_major_kv_layout": false, "enable_unified_memory": false, "disable_chunked_prefix_cache": true, "disable_overlap_schedule": false, "num_continuous_decode_steps": 1, "scheduler_recv_interval": 1, "enable_mixed_chunk": false, "nccl_port": null, "dist_timeout": 3600, "dist_init_addr": "174.1.60.5:20000", "nnodes": 4, "node_rank": 0, "tp_size": 32, "dcp_size": 1, "pp_size": 1, "pp_max_micro_batch_size": 181, "pp_async_batch_depth": 0, "dp_size": 1, "load_balance_method": "round_robin", "attn_cp_size": 1, "moe_dp_size": 1, "dcp_comm_backend": "ag_rs", "dcp_replicate_q_proj": null, "dwdp_size": 1, "enable_prefill_cp": false, "cp_strategy": null, "enable_dsa_cache_layer_split": false, "enable_dsa_prefill_context_parallel": false, "dsa_prefill_cp_mode": "round-robin-split", "enable_prefill_context_parallel": false, "prefill_cp_mode": "in-seq-split", "enable_cp_decode_attn_tp": false, "enable_dp_attention": false, "enable_dp_attention_local_control_broadcast": false, "enable_dp_lm_head": false, "enable_attn_tp_input_scattered": false, "disable_attn_tp_gather": false, "enable_p2p_check": false, "device": "cuda", "base_gpu_id": 0, "gpu_id_step": 1, "random_seed": 369235288, "watchdog_timeout": 300, "soft_watchdog_timeout": null, "sleep_on_idle": false, "use_ray": false, "numa_node": null, "gc_threshold": null, "host": "0.0.0.0", "port": 30000, "fastapi_root_path": "", "smg_grpc_mode": false, "grpc_mode": false, "grpc_port": null, "sidecar": null, "sidecar_args": null, "skip_server_warmup": false, "warmups": null, "enable_http2": false, "ssl_keyfile": null, "ssl_certfile": null, "ssl_ca_certs": null, "ssl_keyfile_password": null, "enable_ssl_refresh": false, "api_key": null, "admin_api_key": null, "served_model_name": "kimi-k3", "weight_version": "default", "chat_template": null, "hf_chat_template_name": null, "completion_template": null, "file_storage_path": "sglang_storage", "enable_cache_report": false, "reasoning_parser": null, "default_chat_template_kwargs": null, "strip_thinking_cache": false, "enable_strict_thinking": false, "tool_call_parser": null, "tool_server": null, "sampling_defaults": "model", "asr_max_buffer_seconds": 60, "asr_max_concurrent_sessions": 32, "preferred_sampling_params": null, "allow_auto_truncate": false, "stream_interval": 1, "batch_notify_size": 16, "stream_response_default_include_usage": false, "incremental_streaming_output": false, "enable_streaming_session": false, "enable_session_radix_cache": false, "log_level": "info", "log_level_http": null, "log_requests": false, "log_requests_level": 2, "log_requests_format": "text", "log_requests_target": null, "uvicorn_access_log_exclude_prefixes": [], "crash_dump_folder": null, "show_time_cost": false, "enable_metrics": false, "smg_http_sidecar_port": null, "enable_mfu_metrics": false, "enable_metrics_for_all_schedulers": false, "load_snapshot_publish_interval": 15, "tokenizer_metrics_custom_labels_header": "x-custom-labels", "tokenizer_metrics_allowed_custom_labels": null, "extra_metric_labels": null, "bucket_time_to_first_token": null, "bucket_inter_token_latency": null, "bucket_e2e_request_latency": null, "prompt_tokens_buckets": null, "generation_tokens_buckets": null, "gc_warning_threshold_secs": 0.0, "decode_log_interval": 40, "enable_request_time_stats_logging": false, "kv_events_config": null, "enable_forward_pass_metrics": false, "forward_pass_metrics_worker_id": "", "forward_pass_metrics_ipc_name": null, "enable_trace": false, "trace_modules": "request", "otlp_traces_endpoint": "localhost:4317", "export_metrics_to_file": false, "export_metrics_to_file_dir": null, "stat_loggers": null, "constrained_json_whitespace_pattern": null, "constrained_json_disable_any_whitespace": false, "attention_backend": "triton", "decode_attention_backend": null, "prefill_attention_backend": null, "sampling_backend": "flashinfer", "grammar_backend": "xgrammar", "radix_cache_backend": null, "mm_attention_backend": null, "fp8_gemm_runner_backend": "auto", "fp4_gemm_runner_backend": "auto", "bf16_gemm_backend": "auto", "dsa_prefill_backend": null, "dsa_decode_backend": null, "dsa_paged_mqa_logits_backend": "auto", "dsa_topk_backend": "sgl-kernel", "disable_flashinfer_autotune": false, "flashinfer_autotune_skip_ops": null, "mamba_backend": "triton", "cuda_graph_config": {"decode": {"backend": "full", "max_bs": 16, "bs": [1, 2, 4, 8, 12, 16], "tc_compiler": "eager", "full_prefill_max_req": null}, "prefill": {"backend": "disabled", "max_bs": 2048, "bs": [4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], "tc_compiler": "eager", "full_prefill_max_req": null}}, "cuda_graph_backend_decode": null, "cuda_graph_backend_prefill": null, "cuda_graph_max_bs_decode": 16, "cuda_graph_max_bs_prefill": null, "cuda_graph_bs_decode": null, "cuda_graph_bs_prefill": null, "cuda_graph_tc_compiler": null, "disable_prefill_cuda_graph": false, "disable_decode_cuda_graph": false, "disable_cuda_graph": false, "disable_cuda_graph_padding": false, "enable_profile_cuda_graph": false, "enable_cudagraph_gc": false, "debug_cuda_graph": false, "enable_layerwise_nvtx_marker": false, "enable_nccl_nvls": false, "enable_symm_mem": false, "triton_attention_reduce_in_fp32": false, "triton_attention_num_kv_splits": 8, "triton_attention_split_tile_size": null, "flashinfer_mla_disable_ragged": false, "enable_fused_qk_norm_rope": false, "enable_precise_embedding_interpolation": false, "enable_fused_moe_sum_all_reduce": false, "enable_deepseek_v4_fp4_indexer": false, "disable_custom_all_reduce": false, "enable_mscclpp": false, "enable_torch_symm_mem": false, "enable_scattered_sconv": false, "pre_warm_nccl": false, "enable_quant_communications": false, "enable_flashinfer_allreduce_fusion": false, "enforce_disable_flashinfer_allreduce_fusion": false, "flashinfer_allreduce_fusion_backend": null, "enable_aiter_allreduce_fusion": false, "enable_torch_compile": false, "enable_torch_compile_debug_mode": false, "torch_compile_max_bs": 32, "torchao_config": "", "speculative_algorithm": null, "speculative_draft_model_path": null, "speculative_draft_model_revision": null, "speculative_draft_load_format": null, "speculative_num_steps": null, "speculative_eagle_topk": null, "speculative_num_draft_tokens": null, "speculative_dflash_block_size": null, "speculative_dspark_block_size": null, "speculative_dspark_sps_table_path": null, "speculative_dspark_confidence_sts_path": null, "speculative_dspark_align_verify_tokens_to_graph_tier": false, "speculative_accept_threshold_single": 1.0, "speculative_accept_threshold_acc": 1.0, "speculative_use_rejection_sampling": false, "speculative_token_map": null, "speculative_attention_mode": "prefill", "speculative_draft_attention_backend": null, "speculative_draft_window_size": null, "speculative_moe_runner_backend": "marlin", "speculative_moe_a2a_backend": null, "speculative_draft_model_quantization": null, "speculative_skip_dp_mlp_sync": false, "enable_multi_layer_eagle": false, "speculative_adaptive": false, "speculative_adaptive_config": null, "decoupled_spec_bind_endpoint": null, "decoupled_spec_connect_endpoints": null, "decoupled_spec_rank": null, "decoupled_spec_role": "null", "spec_trace_dir": null, "speculative_ngram_min_bfs_breadth": 1, "speculative_ngram_max_bfs_breadth": 10, "speculative_ngram_match_type": "BFS", "speculative_ngram_max_trie_depth": 18, "speculative_ngram_capacity": 10000000, "speculative_ngram_external_corpus_path": null, "speculative_ngram_external_sam_budget": 0, "speculative_ngram_external_corpus_max_tokens": 10000000, "ep_size": 32, "moe_a2a_backend": "none", "moe_runner_backend": "marlin", "flashinfer_mxfp4_moe_precision": "default", "deepep_mode": "auto", "fuseep_mode": 2, "deepep_dispatcher_output_dtype": "auto", "ep_num_redundant_experts": 0, "ep_dispatch_algorithm": null, "init_expert_location": "trivial", "enable_eplb": false, "eplb_algorithm": "auto", "eplb_rebalance_num_iterations": 1000, "eplb_rebalance_layers_per_chunk": null, "eplb_min_rebalancing_utilization_threshold": 1.0, "expert_distribution_recorder_mode": null, "expert_distribution_recorder_buffer_size": 1000, "enable_expert_distribution_metrics": false, "deepep_config": null, "moe_dense_tp_size": null, "elastic_ep_backend": null, "enable_elastic_expert_backup": false, "mooncake_ib_device": null, "enable_waterfill": false, "ep_join_mode": null, "ep_join_rank_offset": 0, "elastic_ep_initial_size": null, "max_ep_size": null, "elastic_ep_scale_timeout": 600, "elastic_ep_rejoin": false, "disable_flashinfer_cutlass_moe_fp4_allgather": false, "disable_shared_experts_fusion": false, "enforce_shared_experts_fusion": false, "max_mamba_cache_size": 181, "mamba_ssm_dtype": null, "mamba_max_states_per_path": -1, "enable_mamba_cache_stochastic_rounding": false, "mamba_cache_philox_rounds": 0, "mamba_full_memory_ratio": 0.21, "mamba_radix_cache_strategy": "extra_buffer_lazy", "uses_mamba_radix_cache": false, "mamba_track_interval": 256, "enable_int8_mamba_checkpoint": false, "int8_mamba_ckpt_size": null, "linear_attn_backend": "triton", "linear_attn_decode_backend": null, "linear_attn_prefill_backend": null, "linear_attn_verify_backend": null, "enable_linear_replayssm": false, "linear_replayssm_cache_len": 16, "enable_linear_replayssm_spec": false, "enable_hierarchical_cache": false, "hicache_ratio": 2.0, "hicache_size": 0, "hicache_write_policy": "write_through", "hicache_io_backend": "kernel", "hicache_mem_layout": "page_first", "hicache_storage_backend": null, "hicache_storage_prefetch_policy": "timeout", "hicache_storage_backend_extra_config": null, "enable_hisparse": false, "hisparse_config": null, "enable_broadcast_mm_inputs_process": false, "enable_prefix_mm_cache": false, "mm_enable_dp_encoder": false, "mm_process_config": {}, "mm_processor_worker_num": 0, "mm_io_worker_num": 0, "limit_mm_data_per_request": null, "enable_mm_global_cache": false, "disable_fast_image_processor": false, "mm_feature_transport": "cpu", "keep_mm_feature_on_device": false, "enable_lora": null, "enable_lora_overlap_loading": null, "max_lora_rank": null, "lora_target_modules": null, "lora_paths": null, "max_loaded_loras": null, "max_loras_per_batch": 8, "lora_eviction_policy": "lru", "lora_backend": "csgmv", "max_lora_chunk_size": 16, "experts_shared_outer_loras": null, "lora_use_virtual_experts": false, "lora_strict_loading": false, "lora_drain_wait_threshold": 0.0, "enable_two_batch_overlap": false, "enable_single_batch_overlap": false, "tbo_token_distribution_threshold": 0.48, "cpu_offload_gb": 0, "offload_group_size": -1, "offload_num_in_group": 1, "offload_prefetch_step": 1, "offload_mode": "cpu", "enable_lmcache": false, "lmcache_config_file": null, "enable_flexkv": false, "flexkv_config_file": null, "kt_weight_path": null, "kt_method": "AMXINT4", "kt_cpuinfer": null, "kt_threadpool_count": 2, "kt_num_gpu_experts": null, "kt_max_deferred_experts_per_token": null, "dllm_algorithm": null, "dllm_algorithm_config": null, "dllm_fdfo": true, "disaggregation_mode": "decode", "disaggregation_transfer_backend": "mooncake", "disaggregation_bootstrap_port": 28800, "disaggregation_ib_device": "mlx5_0,mlx5_1,mlx5_2,mlx5_3", "disaggregation_decode_enable_radix_cache": false, "disaggregation_decode_enable_offload_kvcache": false, "num_reserved_decode_tokens": 512, "disaggregation_decode_extra_slots": 0, "disaggregation_decode_polling_interval": 1, "optimistic_prefill_attempts": 0, "encoder_only": false, "language_only": false, "encoder_transfer_backend": "zmq_to_tokenizer", "encoder_urls": [], "encoder_bootstrap_port": 8997, "encoder_register_urls": [], "enable_adaptive_dispatch_to_encoder": false, "enable_pdmux": false, "pdmux_config_path": null, "sm_group_num": 8, "custom_weight_loader": [], "weight_loader_disable_mmap": false, "weight_loader_prefetch_checkpoints": false, "weight_loader_prefetch_num_threads": 4, "weight_loader_drop_cache_after_load": false, "remote_instance_weight_loader_seed_instance_ip": null, "remote_instance_weight_loader_seed_instance_service_port": null, "remote_instance_weight_loader_send_weights_group_ports": null, "remote_instance_weight_loader_backend": "nccl", "remote_instance_weight_loader_start_seed_via_transfer_engine": false, "engine_info_bootstrap_port": 6789, "modelexpress_config": null, "download_dir": null, "model_checksum": null, "delete_ckpt_after_loading": false, "decrypted_config_file": null, "decrypted_draft_config_file": null, "checkpoint_engine_wait_weights_before_ready": false, "enable_prefill_delayer": false, "prefill_delayer_max_delay_passes": 30, "prefill_delayer_token_usage_low_watermark": null, "prefill_delayer_forward_passes_buckets": null, "prefill_delayer_wait_seconds_buckets": null, "prefill_delayer_queue_min_ratio": null, "prefill_delayer_max_delay_ms": null, "min_free_slots_delay": null, "enable_deterministic_inference": false, "rl_on_policy_target": null, "kv_canary": "none", "kv_canary_real_data": "none", "kv_canary_sweep_interval": 0, "enable_dynamic_batch_tokenizer": false, "dynamic_batch_tokenizer_batch_size": 32, "dynamic_batch_tokenizer_batch_timeout": 0.002, "enable_tokenizer_batch_encode": false, "disable_tokenizer_batch_decode": false, "debug_tensor_dump_output_folder": null, "debug_tensor_dump_layers": null, "debug_tensor_dump_input_file": null, "enable_memory_saver": false, "enable_weights_cpu_backup": false, "enable_draft_weights_cpu_backup": false, "enable_custom_logit_processor": false, "enable_return_hidden_states": false, "enable_return_routed_experts": false, "enable_return_indexer_topk": false, "disable_outlines_disk_cache": false, "enable_mis": false, "weight_cache_mode": "off", "weight_cache_socket": null, "weight_cache_timeout": 1800, "forward_hooks": null, "msprobe_dump_config": null, "_resolved_overrides": [["_sampling_backend_default", {"sampling_backend": "flashinfer"}], ["_attention_backend_default", {"attention_backend": "triton"}], ["_data_parallelism_defaults", {"enable_dp_attention": false, "enable_dp_lm_head": false}], ["_speculative_moe_runner_default", {"speculative_moe_runner_backend": "marlin"}]], "grpc_worker_threads": 4, "_quantization_explicitly_unset": false, "_cuda_graph_config_locked": [["decode", "max_bs"]], "_declarations_materialized": true, "_runtime_mutations": [["model_runner.chunked_prefix_cache_gate", {"disable_chunked_prefix_cache": true}], ["mamba_pool.memory_budget", {"max_mamba_cache_size": 181}]], "_in_override": false, "_mx_config_cache": {}, "max_speculative_num_draft_tokens": null, "last_gen_throughput": 26.412623015816244, "memory_usage": {"weight": 58.57, "kvcache": 11.34, "token_capacity": 440256, "graph": 0.6}, "effective_max_running_requests_per_dp": 181}], "version": "0.5.16", "kv_events": null}]}, "duration": 27.523451966932043, "completed": 1, "total_input_tokens": 16384, "total_input_text_tokens": 16384, "total_input_vision_tokens": 0, "total_output_tokens": 512, "total_output_tokens_retokenized": 512, "request_throughput": 0.03633265192176645, "input_throughput": 595.2741690862215, "output_throughput": 18.60231778394442, "total_throughput": 613.8764868701659, "mean_e2e_latency_ms": 27454.693216132, "median_e2e_latency_ms": 27454.693216132, "std_e2e_latency_ms": 0.0, "p90_e2e_latency_ms": 27454.693216132, "p95_e2e_latency_ms": 27454.693216132, "p99_e2e_latency_ms": 27454.693216132, "mean_ttft_ms": 7701.911773066968, "median_ttft_ms": 7701.911773066968, "std_ttft_ms": 0.0, "p90_ttft_ms": 7701.911773066968, "p95_ttft_ms": 7701.911773066968, "p99_ttft_ms": 7701.911773066968, "mean_tpot_ms": 38.655149595039205, "median_tpot_ms": 38.655149595039205, "std_tpot_ms": 0.0, "p90_tpot_ms": 38.655149595039205, "p95_tpot_ms": 38.655149595039205, "p99_tpot_ms": 38.655149595039205, "mean_itl_ms": 38.73088652742844, "median_itl_ms": 38.19326113443822, "std_itl_ms": 0.9328015917220247, "p90_itl_ms": 39.95937877334654, "p95_itl_ms": 40.03999005071819, "p99_itl_ms": 40.276281335391104, "concurrency": 0.9975018122406065, "accept_length": null, "max_output_tokens_per_s": 27.0, "max_concurrent_requests": 1, "input_lens": [16384], "output_lens": [512], "ttfts": [7.701911773066968], "itls": [[0.04083035001531243, 0.039806921035051346, 0.04024836583994329, 0.03981207404285669, 0.04004489514045417, 0.04006541380658746, 0.03984395717270672, 0.039881554897874594, 0.03979184199124575, 0.03990529803559184, 0.039889982901513577, 0.03989713708870113, 0.03972071106545627, 0.040088125970214605, 0.03981506102718413, 0.039867267943918705, 0.0396932871080935, 0.039750561816617846, 0.0398527851793915, 0.039994987891986966, 0.03972077392973006, 0.04015325801447034, 0.039877356961369514, 0.039797938195988536, 0.03975668898783624, 0.0399720489513129, 0.03984465589746833, 0.04009190411306918, 0.039891825057566166, 0.04020225792191923, 0.0398714249022305, 0.03985334117896855, 0.03995984885841608, 0.039944007992744446, 0.039914957946166396, 0.03997808205895126, 0.03995979996398091, 0.040014917962253094, 0.04015887319110334, 0.03988239192403853, 0.039850428933277726, 0.03993461304344237, 0.040033994941040874, 0.03987303399480879, 0.03998301504179835, 0.04011787101626396, 0.04001436196267605, 0.03981609712354839, 0.03982728999108076, 0.03981096995994449, 0.039810518035665154, 0.039841328049078584, 0.03999524493701756, 0.04033840890042484, 0.03981981216929853, 0.039846592815592885, 0.03973760805092752, 0.03981906292028725, 0.039803877007216215, 0.039872230030596256, 0.0397039451636374, 0.040212797932326794, 0.04012760892510414, 0.04042372899129987, 0.03994061006233096, 0.039528314024209976, 0.0398298860527575, 0.03988945297896862, 0.0400556898675859, 0.03988743317313492, 0.040032679913565516, 0.03994143195450306, 0.03982814005576074, 0.04170284094288945, 0.03803424700163305, 0.03991949395276606, 0.03996253595687449, 0.0399691560305655, 0.039955497020855546, 0.03999042697250843, 0.04003356001339853, 0.03991758404299617, 0.03985053300857544, 0.03997340006753802, 0.03983691707253456, 0.03993424796499312, 0.03991809603758156, 0.039893379900604486, 0.03977932594716549, 0.03993987198919058, 0.03983458806760609, 0.03990445705130696, 0.03984037693589926, 0.039967257995158434, 0.03985377796925604, 0.039838159922510386, 0.03981524216942489, 0.039974415907636285, 0.039763608016073704, 0.04008330008946359, 0.03971516899764538, 0.03986285883001983, 0.03993189218454063, 0.03958199080079794, 0.04006678913719952, 0.0399901419878006, 0.03992911288514733, 0.04027904220856726, 0.03955029486678541, 0.039887976134195924, 0.039806320797652006, 0.03987939702346921, 0.0397072471678257, 0.0399191549513489, 0.03983283694833517, 0.039945435943081975, 0.03983319201506674, 0.03993722004815936, 0.039941771887242794, 0.03993911622092128, 0.03980040084570646, 0.03987839608453214, 0.03981118486262858, 0.03998718620277941, 0.039742944994941354, 0.04001978994347155, 0.03989219502545893, 0.03994741686619818, 0.03977451310493052, 0.039904935052618384, 0.039871356915682554, 0.03995933197438717, 0.03975434694439173, 0.03988255513831973, 0.040010878816246986, 0.040141766192391515, 0.03951797494664788, 0.03983631287701428, 0.03988760709762573, 0.0400468239095062, 0.039625395089387894, 0.03994750208221376, 0.03972610295750201, 0.040013284888118505, 0.03973602713085711, 0.03979067294858396, 0.039858279982581735, 0.039932827930897474, 0.03970248601399362, 0.04009915399365127, 0.03986002109013498, 0.03987385402433574, 0.04019812401384115, 0.03964482294395566, 0.039856780087575316, 0.03988133883103728, 0.039753122022375464, 0.039920883951708674, 0.039847328094765544, 0.039940994000062346, 0.03976979502476752, 0.04028426902368665, 0.03993350802920759, 0.03993080789223313, 0.03991145803593099, 0.03996794600971043, 0.03986114892177284, 0.039957434171810746, 0.03981240885332227, 0.03979864902794361, 0.040152653120458126, 0.0398433399386704, 0.03979198099114001, 0.0399195090867579, 0.039843950886279345, 0.03991010901518166, 0.03953455504961312, 0.039924607845023274, 0.040070487186312675, 0.03999910689890385, 0.03981633693911135, 0.03984182816930115, 0.03956749779172242, 0.039325586054474115, 0.03921738709323108, 0.039390712045133114, 0.03931170399300754, 0.03919725585728884, 0.039212661096826196, 0.03942238399758935, 0.03927897987887263, 0.03899491811171174, 0.03881128807552159, 0.03875023382715881, 0.038990556029602885, 0.03873132006265223, 0.038665084866806865, 0.03873454011045396, 0.03867092798464, 0.03866072907112539, 0.03858735994435847, 0.0387242219876498, 0.03860955708660185, 0.03877656301483512, 0.038651238894090056, 0.038767260033637285, 0.03865952300839126, 0.03873041388578713, 0.038636347046121955, 0.038714302936568856, 0.038667993154376745, 0.038665423868224025, 0.03855616203509271, 0.03853716398589313, 0.03822874906472862, 0.03863983997143805, 0.0379248580429703, 0.038125060964375734, 0.03813764313235879, 0.03820766392163932, 0.038147432962432504, 0.038300107000395656, 0.038114351918920875, 0.03831721097230911, 0.03822934813797474, 0.03817507205531001, 0.03814107598736882, 0.03813586803153157, 0.038180907955393195, 0.038173602893948555, 0.03819765197113156, 0.03823085408657789, 0.03814362594857812, 0.038269539130851626, 0.038104990031570196, 0.03819922590628266, 0.03809981304220855, 0.03847339004278183, 0.03806163580156863, 0.03797810315154493, 0.03811655193567276, 0.03822488198056817, 0.03819992393255234, 0.03819547616876662, 0.03815723187290132, 0.038195664063096046, 0.03819975908845663, 0.03815278387628496, 0.03810095600783825, 0.03824832709506154, 0.038080518832430243, 0.03815715410746634, 0.03812673594802618, 0.03821625513955951, 0.03821930778212845, 0.03813746804371476, 0.0381540241651237, 0.03820335795171559, 0.038169476902112365, 0.0381889920681715, 0.03817570605315268, 0.038171401945874095, 0.0381712110247463, 0.03813642496243119, 0.03815454803407192, 0.03813593205995858, 0.038112003821879625, 0.038212264189496636, 0.03815942979417741, 0.03826577216386795, 0.03821524395607412, 0.038175917929038405, 0.0382415649946779, 0.03828978515230119, 0.038110457826405764, 0.03821987914852798, 0.038246218813583255, 0.03816439607180655, 0.03826757613569498, 0.03812349890358746, 0.038078540936112404, 0.038237032014876604, 0.038191046100109816, 0.038141411961987615, 0.03807720495387912, 0.0382329230196774, 0.03848673007450998, 0.0381049239076674, 0.038118692114949226, 0.03814377589151263, 0.03809043695218861, 0.03830795013345778, 0.03822628501802683, 0.038121853955090046, 0.03819987690076232, 0.03816125704906881, 0.038065752014517784, 0.03814074210822582, 0.03801393182948232, 0.038078453158959746, 0.038003708934411407, 0.03807434788905084, 0.03800998907536268, 0.03800858906470239, 0.0379820188973099, 0.03801767807453871, 0.03798158885911107, 0.0387815001886338, 0.03801312786526978, 0.03801161306910217, 0.03793519106693566, 0.03803338296711445, 0.038007644936442375, 0.03795997309498489, 0.03783410391770303, 0.037973155034705997, 0.037979425862431526, 0.038017732091248035, 0.03809530008584261, 0.038087477907538414, 0.03830961603671312, 0.03775124694220722, 0.038057960104197264, 0.0380217470228672, 0.037962338887155056, 0.03819730691611767, 0.03788514412008226, 0.037983539048582315, 0.037901060888543725, 0.037958678090944886, 0.03792150295339525, 0.037934275111183524, 0.03804452484473586, 0.03809245000593364, 0.03790992312133312, 0.037968048825860023, 0.03792752209119499, 0.037924807984381914, 0.03802611096762121, 0.037991485092788935, 0.037958252942189574, 0.037955495063215494, 0.0379452770575881, 0.03792221285402775, 0.03793048206716776, 0.038048171903938055, 0.03787084901705384, 0.03800319298170507, 0.03786134603433311, 0.03796209115535021, 0.03792174579575658, 0.037902448093518615, 0.03793382295407355, 0.03802823298610747, 0.0378846931271255, 0.03806462185457349, 0.037914276123046875, 0.03800813597626984, 0.037913325941190124, 0.037793420953676105, 0.039528438122943044, 0.03664518194273114, 0.03785427799448371, 0.03805864602327347, 0.037814140086993575, 0.0380529728718102, 0.03804642800241709, 0.03794725798070431, 0.037956916028633714, 0.038014082005247474, 0.037981115048751235, 0.037906656973063946, 0.03807410807348788, 0.037832233821973205, 0.03802544507198036, 0.03795806411653757, 0.03793817386031151, 0.037966476986184716, 0.0380381119903177, 0.03787209698930383, 0.03794022602960467, 0.03794430801644921, 0.038029465125873685, 0.038025602931156754, 0.03794296900741756, 0.03788586286827922, 0.038625311106443405, 0.037625492084771395, 0.03784615686163306, 0.03785931202583015, 0.03785987407900393, 0.03784955898299813, 0.03780809696763754, 0.03788955393247306, 0.038073400035500526, 0.03761594300158322, 0.037838489981368184, 0.03788116597570479, 0.037814453011378646, 0.037921556970104575, 0.03772356314584613, 0.037884467048570514, 0.03783004800789058, 0.03788671689108014, 0.03797538811340928, 0.03775185393169522, 0.037800034042447805, 0.03785351989790797, 0.03776058810763061, 0.037823939928784966, 0.03787536802701652, 0.037934087915346026, 0.03785013314336538, 0.03790527884848416, 0.037830319022759795, 0.03791004209779203, 0.03782674600370228, 0.03786582499742508, 0.03775250096805394, 0.0378405989613384, 0.03776222304441035, 0.03795375698246062, 0.03786348085850477, 0.0379209010861814, 0.03771954006515443, 0.03794064186513424, 0.03783560311421752, 0.037853949004784226, 0.03784101293422282, 0.03786959405988455, 0.037826152984052896, 0.03794732899405062, 0.03779317298904061, 0.03782378998585045, 0.03786364197731018, 0.03785557299852371, 0.03793261316604912, 0.037899096962064505, 0.037913725012913346, 0.03787644882686436, 0.037871441105380654, 0.03813351597636938, 0.03749885899014771, 0.03788974112831056, 0.037672639824450016, 0.03791048796847463, 0.03778552915900946, 0.03793626092374325, 0.03767412202432752, 0.03786709997802973, 0.037888099905103445, 0.037838558200746775, 0.03786235488951206, 0.037622767966240644, 0.03809811803512275, 0.037938015069812536, 0.037761881947517395, 0.03791664890013635, 0.03777698101475835, 0.037898272974416614, 0.037721790140494704, 0.03783384198322892, 0.03777735005132854, 0.03845384903252125, 0.03739743190817535, 0.03778820997104049, 0.03792793699540198, 0.03777981293387711, 0.03773873904719949, 0.037926501128822565, 0.037915224907919765, 0.03785373899154365, 0.03781453403644264, 0.03784484090283513, 0.037772834999486804, 0.037913073087111115, 0.03776575392112136, 0.03786557703278959, 0.03766468702815473, 0.037972931982949376, 0.03780396212823689, 0.03789454884827137, 0.03779123304411769, 0.037896068999543786, 0.03783813398331404, 0.03785294690169394, 0.037812387105077505, 0.037921993993222713, 0.03773335297591984, 0.03792384103871882, 0.037789541063830256, 0.037878297036513686, 0.03789319400675595, 0.037786236964166164, 0.03780416282825172, 0.03791103116236627, 0.03786896984092891, 0.03789918008260429, 0.03778183204121888, 0.03784469095990062, 0.038375322008505464, 0.037897228030487895, 0.037819087971001863, 0.037856003968045115, 0.03783592092804611, 0.037882712204009295, 0.03775224299170077, 0.03788603190332651, 0.03771391394548118, 0.03933928208425641]], "generated_texts": ["\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-"], "errors": [""]}
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-20260827-1325/bench/decode_16k_512_c1.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-20260827-1325/bench/decode_16k_512_c1.log
new file mode 100644
index 0000000..4ab8655
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-20260827-1325/bench/decode_16k_512_c1.log
@@ -0,0 +1,71 @@
+/usr/local/lib/python3.12/dist-packages/torchao/quantization/quant_api.py:1731: SyntaxWarning: invalid escape sequence '\.'
+  """Configuration class for applying different quantization configs to modules or parameters based on their fully qualified names (FQNs).
+/sgl-workspace/sglang/python/sglang/kernels/ops/attention/fla/utils.py:223: UserWarning: Triton is not supported on current platform, roll back to CPU.
+  warnings.warn(
+/sgl-workspace/sglang/python/sglang/srt/layers/quantization/awq/awq.py:52: UserWarning: Only CUDA, HIP and XPU support AWQ currently.
+  warnings.warn(f"Only CUDA, HIP and XPU support AWQ currently.")
+/sgl-workspace/sglang/python/sglang/srt/layers/quantization/gguf.py:69: UserWarning: Only CUDA, MUSA and NPU support GGUF quantization currently.
+  warnings.warn(f"Only CUDA, MUSA and NPU support GGUF quantization currently.")
+benchmark_args=Namespace(backend='sglang', base_url=None, host='174.1.60.5', port=31000, ready_check_timeout_sec=60, dataset_name='random-ids', dataset_path='', dataset_offset=0, agentic_max_turns=None, speed_bench_category=None, speed_bench_output_len=512, model=None, served_model_name=None, tokenizer='/data/hf_models/Kimi-K3', num_prompts=1, sharegpt_output_len=None, sharegpt_context_len=None, random_input_len=16384, random_output_len=512, random_range_ratio=1.0, image_count=1, image_resolution='1080p', random_image_count=False, image_format='jpeg', image_content='random', request_rate=10000.0, use_trace_timestamps=False, max_concurrency=1, output_file='/data/hzy/sskj/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-20260827-1325/bench/decode_16k_512_c1.jsonl', output_details=True, print_requests=False, disable_tqdm=True, disable_stream=False, return_logprob=False, top_logprobs_num=0, token_ids_logprob=None, logprob_start_len=-1, return_routed_experts=False, cache_report=False, seed=42, disable_ignore_eos=False, temperature=0.0, top_p=1.0, extra_request_body=None, apply_chat_template=False, profile=False, plot_throughput=False, profile_activities=['CPU', 'GPU'], profile_start_step=None, profile_steps=None, profile_num_steps=None, profile_by_stage=False, profile_stages=None, profile_output_dir=None, profile_prefix=None, lora_name=None, lora_request_distribution='uniform', lora_zipf_alpha=1.5, prompt_suffix='', pd_separated=False, profile_prefill_url=None, profile_decode_url=None, flush_cache=False, warmup_requests=0, tokenize_prompt=False, gsp_num_groups=64, gsp_prompts_per_group=16, gsp_system_prompt_len=2048, gsp_question_len=128, gsp_output_len=256, gsp_range_ratio=1.0, gsp_fast_prepare=False, gsp_send_routing_key=False, gsp_num_turns=1, gsp_ordered=False, gsp_group_distribution='uniform', gsp_zipf_alpha=None, mooncake_slowdown_factor=1.0, mooncake_num_rounds=1, mooncake_workload='conversation', fake_prefill=False, tag=None, header=None)
+Waiting up to 60s for http://174.1.60.5:31000/v1/models to become ready...
+Server ready in 0.0s.
+'[Errno 101] Network is unreachable' thrown while requesting HEAD https://huggingface.co/kimi-k3/resolve/main/config.json
+Retrying in 1s [Retry 1/5].
+'[Errno 101] Network is unreachable' thrown while requesting HEAD https://huggingface.co/kimi-k3/resolve/main/config.json
+Retrying in 1s [Retry 1/5].
+Fail to load tokenizer config with error=Can't load the configuration of 'kimi-k3'. If you were trying to load it from 'https://huggingface.co/models', make sure you don't have a local directory with the same name. Otherwise, make sure 'kimi-k3' is the correct path to a directory containing a config.json file
+
+WARNING It is recommended to use the `Chat` or `Instruct` model for benchmarking.
+Because when the tokenizer counts the output tokens, if there is gibberish, it might count incorrectly.
+
+Namespace(backend='sglang', base_url=None, host='174.1.60.5', port=31000, ready_check_timeout_sec=60, dataset_name='random-ids', dataset_path='', dataset_offset=0, agentic_max_turns=None, speed_bench_category=None, speed_bench_output_len=512, model='kimi-k3', served_model_name=None, tokenizer='/data/hf_models/Kimi-K3', num_prompts=1, sharegpt_output_len=None, sharegpt_context_len=None, random_input_len=16384, random_output_len=512, random_range_ratio=1.0, image_count=1, image_resolution='1080p', random_image_count=False, image_format='jpeg', image_content='random', request_rate=10000.0, use_trace_timestamps=False, max_concurrency=1, output_file='/data/hzy/sskj/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-20260827-1325/bench/decode_16k_512_c1.jsonl', output_details=True, print_requests=False, disable_tqdm=True, disable_stream=False, return_logprob=False, top_logprobs_num=0, token_ids_logprob=None, logprob_start_len=-1, return_routed_experts=False, cache_report=False, seed=42, disable_ignore_eos=False, temperature=0.0, top_p=1.0, extra_request_body=None, apply_chat_template=False, profile=False, plot_throughput=False, profile_activities=['CPU', 'GPU'], profile_start_step=None, profile_steps=None, profile_num_steps=None, profile_by_stage=False, profile_stages=None, profile_output_dir=None, profile_prefix=None, lora_name=None, lora_request_distribution='uniform', lora_zipf_alpha=1.5, prompt_suffix='', pd_separated=False, profile_prefill_url=None, profile_decode_url=None, flush_cache=False, warmup_requests=0, tokenize_prompt=False, gsp_num_groups=64, gsp_prompts_per_group=16, gsp_system_prompt_len=2048, gsp_question_len=128, gsp_output_len=256, gsp_range_ratio=1.0, gsp_fast_prepare=False, gsp_send_routing_key=False, gsp_num_turns=1, gsp_ordered=False, gsp_group_distribution='uniform', gsp_zipf_alpha=None, mooncake_slowdown_factor=1.0, mooncake_num_rounds=1, mooncake_workload='conversation', fake_prefill=False, tag=None, header=None)
+
+#Input tokens: 16384
+#Output tokens: 512
+Starting warmup with 0 sequences...
+Warmup completed with 0 sequences. Starting main benchmark run...
+Calling super().encode with {'add_special_tokens': False}
+
+============ Serving Benchmark Result ============
+Backend:                                 sglang    
+Traffic request rate:                    10000.0   
+Max request concurrency:                 1         
+Successful requests:                     1         
+Benchmark duration (s):                  27.52     
+Total input tokens:                      16384     
+Total input text tokens:                 16384     
+Total generated tokens:                  512       
+Total generated tokens (retokenized):    512       
+Request throughput (req/s):              0.04      
+Input token throughput (tok/s):          595.27    
+Output token throughput (tok/s):         18.60     
+Peak output token throughput (tok/s):    27.00     
+Peak concurrent requests:                1         
+Total token throughput (tok/s):          613.88    
+Concurrency:                             1.00      
+----------------End-to-End Latency----------------
+Mean E2E Latency (ms):                   27454.69  
+Median E2E Latency (ms):                 27454.69  
+P90 E2E Latency (ms):                    27454.69  
+P95 E2E Latency (ms):                    27454.69  
+P99 E2E Latency (ms):                    27454.69  
+---------------Time to First Token----------------
+Mean TTFT (ms):                          7701.91   
+Median TTFT (ms):                        7701.91   
+P90 TTFT (ms):                           7701.91   
+P95 TTFT (ms):                           7701.91   
+P99 TTFT (ms):                           7701.91   
+-----Time per Output Token (excl. 1st token)------
+Mean TPOT (ms):                          38.66     
+Median TPOT (ms):                        38.66     
+P90 TPOT (ms):                           38.66     
+P95 TPOT (ms):                           38.66     
+P99 TPOT (ms):                           38.66     
+---------------Inter-Token Latency----------------
+Mean ITL (ms):                           38.73     
+Median ITL (ms):                         38.19     
+P90 ITL (ms):                            39.96     
+P95 ITL (ms):                            40.04     
+P99 ITL (ms):                            40.28     
+Max ITL (ms):                            41.70     
+==================================================
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-20260827-1325/logs/manual_d_node0.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-20260827-1325/logs/manual_d_node0.log
new file mode 100644
index 0000000..e8a66b8
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-20260827-1325/logs/manual_d_node0.log
@@ -0,0 +1,1642 @@
+/usr/local/lib/python3.12/dist-packages/torchao/quantization/quant_api.py:1731: SyntaxWarning: invalid escape sequence '\.'
+  """Configuration class for applying different quantization configs to modules or parameters based on their fully qualified names (FQNs).
+/sgl-workspace/sglang/python/sglang/launch_server.py:56: UserWarning: 'python -m sglang.launch_server' is still supported, but 'sglang serve' is the recommended entrypoint.
+  Example: sglang serve --model-path <model> [options]
+  warnings.warn(
+[2026-08-27 04:36:08] KV cache is forced as chunk cache for decode server
+[2026-08-27 04:36:11] Breakable CUDA graph is incompatible with MLA attention; disabling prefill CUDA graph.
+[2026-08-27 04:36:12] Attention backend not specified. Use triton backend by default.
+[2026-08-27 04:36:14] server_args=ServerArgs(model_path='/data/hf_models/Kimi-K3', tokenizer_path='/data/hf_models/Kimi-K3', tokenizer_mode='auto', tokenizer_backend='huggingface', tokenizer_worker_num=1, detokenizer_worker_num=1, skip_tokenizer_init=False, load_format='auto', model_loader_extra_config='{}', trust_remote_code=True, context_length=None, is_embedding=False, enable_multimodal=None, revision=None, model_impl='auto', model_config_parser='auto', json_model_override_args='{}', dtype='auto', quantization=None, quantization_param_path=None, kv_cache_dtype='auto', enable_fp32_lm_head=False, modelopt_quant=None, modelopt_checkpoint_restore_path=None, modelopt_checkpoint_save_path=None, modelopt_export_path=None, quantize_and_serve=False, rl_quant_profile=None, enable_tf32_matmul=False, mem_fraction_static=0.88, max_running_requests=None, max_queued_requests=None, max_total_tokens=None, chunked_prefill_size=8192, enable_dynamic_chunking=False, max_prefill_tokens=16384, prefill_max_requests=None, schedule_policy='fcfs', enable_priority_scheduling=False, disable_priority_preemption=False, default_priority_value=None, abort_on_priority_when_disabled=False, schedule_low_priority_values_first=False, priority_scheduling_preemption_threshold=10, retraction_policy='length', schedule_conservativeness=1.0, page_size=64, swa_full_tokens_ratio=0.8, disable_hybrid_swa_memory=False, radix_eviction_policy='lru', prefill_only_disable_kv_cache=False, disable_radix_cache=True, enable_page_major_kv_layout=False, enable_unified_memory=False, disable_chunked_prefix_cache=False, disable_overlap_schedule=False, num_continuous_decode_steps=1, scheduler_recv_interval=1, enable_mixed_chunk=False, nccl_port=None, dist_timeout=3600, dist_init_addr='174.1.60.5:20000', nnodes=4, node_rank=0, tp_size=32, dcp_size=1, pp_size=1, pp_max_micro_batch_size=None, pp_async_batch_depth=0, dp_size=1, load_balance_method='round_robin', attn_cp_size=1, moe_dp_size=1, dcp_comm_backend='ag_rs', dcp_replicate_q_proj=None, dwdp_size=1, enable_prefill_cp=False, cp_strategy=None, enable_dsa_cache_layer_split=False, enable_dsa_prefill_context_parallel=False, dsa_prefill_cp_mode='round-robin-split', enable_prefill_context_parallel=False, prefill_cp_mode='in-seq-split', enable_cp_decode_attn_tp=False, enable_dp_attention=False, enable_dp_attention_local_control_broadcast=False, enable_dp_lm_head=False, enable_attn_tp_input_scattered=False, disable_attn_tp_gather=False, enable_p2p_check=False, device='cuda', base_gpu_id=0, gpu_id_step=1, random_seed=369235288, watchdog_timeout=300, soft_watchdog_timeout=None, sleep_on_idle=False, use_ray=False, custom_sigquit_handler=None, numa_node=None, gc_threshold=None, host='0.0.0.0', port=30000, fastapi_root_path='', smg_grpc_mode=False, grpc_mode=False, grpc_port=None, sidecar=None, sidecar_args=None, skip_server_warmup=False, warmups=None, enable_http2=False, ssl_keyfile=None, ssl_certfile=None, ssl_ca_certs=None, ssl_keyfile_password=None, enable_ssl_refresh=False, api_key=None, admin_api_key=None, served_model_name='kimi-k3', weight_version='default', chat_template=None, hf_chat_template_name=None, completion_template=None, file_storage_path='sglang_storage', enable_cache_report=False, reasoning_parser=None, default_chat_template_kwargs=None, strip_thinking_cache=False, enable_strict_thinking=False, tool_call_parser=None, tool_server=None, sampling_defaults='model', asr_max_buffer_seconds=60, asr_max_concurrent_sessions=32, preferred_sampling_params=None, allow_auto_truncate=False, stream_interval=1, batch_notify_size=16, stream_response_default_include_usage=False, incremental_streaming_output=False, enable_streaming_session=False, enable_session_radix_cache=False, log_level='info', log_level_http=None, log_requests=False, log_requests_level=2, log_requests_format='text', log_requests_target=None, uvicorn_access_log_exclude_prefixes=[], crash_dump_folder=None, show_time_cost=False, enable_metrics=False, smg_http_sidecar_port=None, enable_mfu_metrics=False, enable_metrics_for_all_schedulers=False, load_snapshot_publish_interval=15, tokenizer_metrics_custom_labels_header='x-custom-labels', tokenizer_metrics_allowed_custom_labels=None, extra_metric_labels=None, bucket_time_to_first_token=None, bucket_inter_token_latency=None, bucket_e2e_request_latency=None, prompt_tokens_buckets=None, generation_tokens_buckets=None, gc_warning_threshold_secs=0.0, decode_log_interval=40, enable_request_time_stats_logging=False, kv_events_config=None, enable_forward_pass_metrics=False, forward_pass_metrics_worker_id='', forward_pass_metrics_ipc_name=None, enable_trace=False, trace_modules='request', otlp_traces_endpoint='localhost:4317', export_metrics_to_file=False, export_metrics_to_file_dir=None, stat_loggers=None, constrained_json_whitespace_pattern=None, constrained_json_disable_any_whitespace=False, attention_backend='triton', decode_attention_backend=None, prefill_attention_backend=None, sampling_backend='flashinfer', grammar_backend='xgrammar', radix_cache_backend=None, mm_attention_backend=None, fp8_gemm_runner_backend='auto', fp4_gemm_runner_backend='auto', bf16_gemm_backend='auto', dsa_prefill_backend=None, dsa_decode_backend=None, dsa_paged_mqa_logits_backend='auto', dsa_topk_backend='sgl-kernel', disable_flashinfer_autotune=False, flashinfer_autotune_skip_ops=None, mamba_backend='triton', cuda_graph_config=CudaGraphConfig(decode=PhaseConfig(backend='full', max_bs=16, bs=[1, 2, 4, 8, 12, 16], tc_compiler='eager', full_prefill_max_req=None), prefill=PhaseConfig(backend='disabled', max_bs=2048, bs=[4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], tc_compiler='eager', full_prefill_max_req=None)), cuda_graph_backend_decode=None, cuda_graph_backend_prefill=None, cuda_graph_max_bs_decode=16, cuda_graph_max_bs_prefill=None, cuda_graph_bs_decode=None, cuda_graph_bs_prefill=None, cuda_graph_tc_compiler=None, disable_prefill_cuda_graph=False, disable_decode_cuda_graph=False, disable_cuda_graph=False, disable_cuda_graph_padding=False, enable_profile_cuda_graph=False, enable_cudagraph_gc=False, debug_cuda_graph=False, enable_layerwise_nvtx_marker=False, enable_nccl_nvls=False, enable_symm_mem=False, triton_attention_reduce_in_fp32=False, triton_attention_num_kv_splits=8, triton_attention_split_tile_size=None, flashinfer_mla_disable_ragged=False, enable_fused_qk_norm_rope=False, enable_precise_embedding_interpolation=False, enable_fused_moe_sum_all_reduce=False, enable_deepseek_v4_fp4_indexer=False, disable_custom_all_reduce=False, enable_mscclpp=False, enable_torch_symm_mem=False, enable_scattered_sconv=False, pre_warm_nccl=False, enable_quant_communications=False, enable_flashinfer_allreduce_fusion=False, enforce_disable_flashinfer_allreduce_fusion=False, flashinfer_allreduce_fusion_backend=None, enable_aiter_allreduce_fusion=False, enable_torch_compile=False, enable_torch_compile_debug_mode=False, torch_compile_max_bs=32, torchao_config='', speculative_algorithm=None, speculative_draft_model_path=None, speculative_draft_model_revision=None, speculative_draft_load_format=None, speculative_num_steps=None, speculative_eagle_topk=None, speculative_num_draft_tokens=None, speculative_dflash_block_size=None, speculative_dspark_block_size=None, speculative_dspark_sps_table_path=None, speculative_dspark_confidence_sts_path=None, speculative_dspark_align_verify_tokens_to_graph_tier=False, speculative_accept_threshold_single=1.0, speculative_accept_threshold_acc=1.0, speculative_use_rejection_sampling=False, speculative_token_map=None, speculative_attention_mode='prefill', speculative_draft_attention_backend=None, speculative_draft_window_size=None, speculative_moe_runner_backend='marlin', speculative_moe_a2a_backend=None, speculative_draft_model_quantization=None, speculative_skip_dp_mlp_sync=False, enable_multi_layer_eagle=False, speculative_adaptive=False, speculative_adaptive_config=None, decoupled_spec_bind_endpoint=None, decoupled_spec_connect_endpoints=None, decoupled_spec_rank=None, decoupled_spec_role='null', spec_trace_dir=None, speculative_ngram_min_bfs_breadth=1, speculative_ngram_max_bfs_breadth=10, speculative_ngram_match_type='BFS', speculative_ngram_max_trie_depth=18, speculative_ngram_capacity=10000000, speculative_ngram_external_corpus_path=None, speculative_ngram_external_sam_budget=0, speculative_ngram_external_corpus_max_tokens=10000000, ep_size=32, moe_a2a_backend='none', moe_runner_backend='marlin', flashinfer_mxfp4_moe_precision='default', deepep_mode='auto', fuseep_mode=2, deepep_dispatcher_output_dtype='auto', ep_num_redundant_experts=0, ep_dispatch_algorithm=None, init_expert_location='trivial', enable_eplb=False, eplb_algorithm='auto', eplb_rebalance_num_iterations=1000, eplb_rebalance_layers_per_chunk=None, eplb_min_rebalancing_utilization_threshold=1.0, expert_distribution_recorder_mode=None, expert_distribution_recorder_buffer_size=1000, enable_expert_distribution_metrics=False, deepep_config=None, moe_dense_tp_size=None, elastic_ep_backend=None, enable_elastic_expert_backup=False, mooncake_ib_device=None, enable_waterfill=False, ep_join_mode=None, ep_join_rank_offset=0, elastic_ep_initial_size=None, max_ep_size=None, elastic_ep_scale_timeout=600, elastic_ep_rejoin=False, disable_flashinfer_cutlass_moe_fp4_allgather=False, disable_shared_experts_fusion=False, enforce_shared_experts_fusion=False, max_mamba_cache_size=None, mamba_ssm_dtype=None, mamba_max_states_per_path=-1, enable_mamba_cache_stochastic_rounding=False, mamba_cache_philox_rounds=0, mamba_full_memory_ratio=0.21, mamba_radix_cache_strategy='extra_buffer_lazy', uses_mamba_radix_cache=False, mamba_track_interval=256, enable_int8_mamba_checkpoint=False, int8_mamba_ckpt_size=None, linear_attn_backend='triton', linear_attn_decode_backend=None, linear_attn_prefill_backend=None, linear_attn_verify_backend=None, enable_linear_replayssm=False, linear_replayssm_cache_len=16, enable_linear_replayssm_spec=False, enable_hierarchical_cache=False, hicache_ratio=2.0, hicache_size=0, hicache_write_policy='write_through', hicache_io_backend='kernel', hicache_mem_layout='page_first', hicache_storage_backend=None, hicache_storage_prefetch_policy='timeout', hicache_storage_backend_extra_config=None, enable_hisparse=False, hisparse_config=None, enable_broadcast_mm_inputs_process=False, enable_prefix_mm_cache=False, mm_enable_dp_encoder=False, mm_process_config={}, mm_processor_worker_num=0, mm_io_worker_num=0, limit_mm_data_per_request=None, enable_mm_global_cache=False, disable_fast_image_processor=False, mm_feature_transport='cpu', keep_mm_feature_on_device=False, enable_lora=None, enable_lora_overlap_loading=None, max_lora_rank=None, lora_target_modules=None, lora_paths=None, max_loaded_loras=None, max_loras_per_batch=8, lora_eviction_policy='lru', lora_backend='csgmv', max_lora_chunk_size=16, experts_shared_outer_loras=None, lora_use_virtual_experts=False, lora_strict_loading=False, lora_drain_wait_threshold=0.0, enable_two_batch_overlap=False, enable_single_batch_overlap=False, tbo_token_distribution_threshold=0.48, cpu_offload_gb=0, offload_group_size=-1, offload_num_in_group=1, offload_prefetch_step=1, offload_mode='cpu', enable_lmcache=False, lmcache_config_file=None, enable_flexkv=False, flexkv_config_file=None, kt_weight_path=None, kt_method='AMXINT4', kt_cpuinfer=None, kt_threadpool_count=2, kt_num_gpu_experts=None, kt_max_deferred_experts_per_token=None, dllm_algorithm=None, dllm_algorithm_config=None, dllm_fdfo=True, disaggregation_mode='decode', disaggregation_transfer_backend='mooncake', disaggregation_bootstrap_port=28800, disaggregation_ib_device='mlx5_0,mlx5_1,mlx5_2,mlx5_3', disaggregation_decode_enable_radix_cache=False, disaggregation_decode_enable_offload_kvcache=False, num_reserved_decode_tokens=512, disaggregation_decode_extra_slots=0, disaggregation_decode_polling_interval=1, optimistic_prefill_attempts=0, encoder_only=False, language_only=False, encoder_transfer_backend='zmq_to_tokenizer', encoder_urls=[], encoder_bootstrap_port=8997, encoder_register_urls=[], enable_adaptive_dispatch_to_encoder=False, enable_pdmux=False, pdmux_config_path=None, sm_group_num=8, custom_weight_loader=[], weight_loader_disable_mmap=False, weight_loader_prefetch_checkpoints=False, weight_loader_prefetch_num_threads=4, weight_loader_drop_cache_after_load=False, remote_instance_weight_loader_seed_instance_ip=None, remote_instance_weight_loader_seed_instance_service_port=None, remote_instance_weight_loader_send_weights_group_ports=None, remote_instance_weight_loader_backend='nccl', remote_instance_weight_loader_start_seed_via_transfer_engine=False, engine_info_bootstrap_port=6789, modelexpress_config=None, download_dir=None, model_checksum=None, delete_ckpt_after_loading=False, decrypted_config_file=None, decrypted_draft_config_file=None, checkpoint_engine_wait_weights_before_ready=False, enable_prefill_delayer=False, prefill_delayer_max_delay_passes=30, prefill_delayer_token_usage_low_watermark=None, prefill_delayer_forward_passes_buckets=None, prefill_delayer_wait_seconds_buckets=None, prefill_delayer_queue_min_ratio=None, prefill_delayer_max_delay_ms=None, min_free_slots_delay=None, enable_deterministic_inference=False, rl_on_policy_target=None, kv_canary='none', kv_canary_real_data='none', kv_canary_sweep_interval=0, enable_dynamic_batch_tokenizer=False, dynamic_batch_tokenizer_batch_size=32, dynamic_batch_tokenizer_batch_timeout=0.002, enable_tokenizer_batch_encode=False, disable_tokenizer_batch_decode=False, debug_tensor_dump_output_folder=None, debug_tensor_dump_layers=None, debug_tensor_dump_input_file=None, enable_memory_saver=False, enable_weights_cpu_backup=False, enable_draft_weights_cpu_backup=False, enable_custom_logit_processor=False, enable_return_hidden_states=False, enable_return_routed_experts=False, enable_return_indexer_topk=False, disable_outlines_disk_cache=False, enable_mis=False, weight_cache_mode='off', weight_cache_socket=None, weight_cache_timeout=1800, forward_hooks=None, msprobe_dump_config=None)
+[2026-08-27 04:36:14] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:14] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:14] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:14] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:14] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:14] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:14] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:14] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:17] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:17] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:18] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:19] Multimodal processor concurrency enabled with 2 isolated worker threads (auto).
+[2026-08-27 04:36:19] No HuggingFace chat template found
+[2026-08-27 04:36:19] No chat template found, defaulting to 'string' content format
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:34 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:34 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:34 TP4 EP4] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:34 TP4 EP4] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:34 TP6 EP6] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:34 TP6 EP6] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:34] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:34] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:34 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:34 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+/sgl-workspace/sglang/python/sglang/srt/utils/hf_transformers/tokenizer.py:423: UserWarning: Using a slow tokenizer. This might cause a significant slowdown. Consider using a fast tokenizer instead.
+  warnings.warn(
+[2026-08-27 04:36:34] Applying special tokens cache patch for Kimi tokenizer: <class 'transformers_modules.Kimi_hyphen_K3.caf40b2773f9029e.tokenization_kimi.TikTokenTokenizer'>
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:34 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:34 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:34 TP7 EP7] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:34 TP7 EP7] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:34 TP5 EP5] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:34 TP5 EP5] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:35 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:35 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:36 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:36 TP6 EP6] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:36 TP4 EP4] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:36 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:36 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:36 TP7 EP7] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:36 TP5 EP5] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:37 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:37.190651   236 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:37.190778   236 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:37.190796   236 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.5 port: 12001
+I0827 04:36:37.190856   236 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.5:16500
+I0827 04:36:37.191038   236 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:37.194572   241 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:37.194692   241 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:37.194712   241 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.5 port: 12001
+I0827 04:36:37.194767   241 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.5:15678
+I0827 04:36:37.194965   241 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:37.219420   236 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:37.223201   241 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:37.227941   236 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:37.231628   241 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:37.236233   236 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:37.239850   241 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:37.244647   236 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:37.248145   241 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:37.248418   236 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:37.248464   236 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:37.248525   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.250921   241 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:37.250947   241 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:37.251004   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.255973   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:37.257668   236 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:05
+I0827 04:36:37.257705   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.283210   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:37.283238   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:37.284873   236 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:05
+I0827 04:36:37.284902   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.284904   241 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:05
+I0827 04:36:37.284934   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.293713   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:37.295406   241 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:05
+I0827 04:36:37.295433   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.311285   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:37.312896   236 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:05
+I0827 04:36:37.312925   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.327399   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:37.328944   241 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:05
+I0827 04:36:37.328972   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.339579   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:37.341301   236 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:05
+I0827 04:36:37.341434   236 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:37 TP1 EP1] Init torch distributed begin.
+I0827 04:36:37.359290   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:37.360814   241 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:05
+I0827 04:36:37.360936   241 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:37 TP6 EP6] Init torch distributed begin.
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:37.366210   239 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:37.366365   239 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:37.366390   239 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.5 port: 12001
+I0827 04:36:37.366467   239 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.5:15787
+I0827 04:36:37.366704   239 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:37.395485   239 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:37.402652   239 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:37.409775   239 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:37.434130   239 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:37.436257   239 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:37.436282   239 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:37.436331   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.442425   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:37.443729   239 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:05
+I0827 04:36:37.443749   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.448421   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:37.449750   239 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:05
+I0827 04:36:37.449769   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.454622   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:37.455952   239 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:05
+I0827 04:36:37.455972   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.463356   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:37.464902   239 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:05
+I0827 04:36:37.465021   239 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:37 TP4 EP4] Init torch distributed begin.
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:37.510618   238 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:37.510744   238 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:37.510763   238 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.5 port: 12001
+I0827 04:36:37.510824   238 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.5:16205
+I0827 04:36:37.511003   238 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:37.524268   242 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:37.524391   242 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:37.524411   242 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.5 port: 12001
+I0827 04:36:37.524524   242 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.5:16605
+I0827 04:36:37.524709   242 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:37.528693   235 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:37.528817   235 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:37.528836   235 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.5 port: 12001
+I0827 04:36:37.528893   235 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.5:16672
+I0827 04:36:37.529069   235 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:37.535192   238 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:37.535425   235 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:37.541376   235 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:37.542625   238 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:37.547586   235 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:37.549854   238 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:37.551366   242 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:37.553905   235 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:37.556406   235 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:37.556468   235 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:37.556523   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.558794   242 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:37.563313   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:37.566102   242 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:37.573271   242 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:37.574895   238 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:37.576385   242 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:37.576412   242 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:37.576478   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.577947   238 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:37.577976   238 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:37.578033   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.583841   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:37.585472   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:37.585623   242 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:05
+I0827 04:36:37.585656   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.587072   238 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:05
+I0827 04:36:37.587105   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.594069   235 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:05
+I0827 04:36:37.594128   235 rdma_context.cpp:205] Using SIEVE endpoint store
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:37.598918   240 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:37.599033   240 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:37.599050   240 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.5 port: 12001
+I0827 04:36:37.599107   240 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.5:15428
+I0827 04:36:37.599270   240 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:37.605341   240 topology.cpp:127] Device mlx5_0 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:37.606741   237 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:37.606841   237 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:37.606855   237 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.5 port: 12001
+I0827 04:36:37.606901   237 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.5:16531
+I0827 04:36:37.607028   237 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:37.611215   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:37.611989   237 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:37.612797   242 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:05
+I0827 04:36:37.612828   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.615110   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:37.616685   238 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:05
+I0827 04:36:37.616714   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.619258   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:37.620560   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:37.620851   235 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:05
+I0827 04:36:37.620885   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.622239   242 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:05
+I0827 04:36:37.622268   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.624459   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:37.625998   238 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:05
+I0827 04:36:37.626025   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.630085   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:37.630686   240 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:37.631680   242 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:05
+I0827 04:36:37.631806   242 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:37 TP7 EP7] Init torch distributed begin.
+I0827 04:36:37.633723   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:37.635268   238 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:05
+I0827 04:36:37.635390   238 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:37 TP3 EP3] Init torch distributed begin.
+I0827 04:36:37.642561   237 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:37.649834   237 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:37.651238   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:37.652786   235 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:05
+I0827 04:36:37.652813   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.657148   237 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:37.658787   240 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:37.659991   237 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:37.660017   237 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:37.660071   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.667523   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:37.669195   240 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:37.669226   237 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:05
+I0827 04:36:37.669257   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.672511   240 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:37.672547   240 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:37.672606   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.677067   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:37.678661   237 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:05
+I0827 04:36:37.678689   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.682178   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:37.683274   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:37.683740   240 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:05
+I0827 04:36:37.683773   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.684834   235 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:05
+I0827 04:36:37.684954   235 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:37 TP0 EP0] Init torch distributed begin.
+I0827 04:36:37.703369   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:37.704975   237 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:05
+I0827 04:36:37.705003   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.711591   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:37.713003   240 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:05
+I0827 04:36:37.713030   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.735574   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:37.737313   237 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:05
+I0827 04:36:37.737439   237 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:37 TP2 EP2] Init torch distributed begin.
+I0827 04:36:37.740485   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:37.741588   240 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:05
+I0827 04:36:37.741605   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.766494   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:37.767982   240 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:05
+I0827 04:36:37.768137   240 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:37 TP5 EP5] Init torch distributed begin.
+[2026-08-27 04:36:46 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:36:48 TP0 EP0] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP1 EP1] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP2 EP2] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP5 EP5] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP4 EP4] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP3 EP3] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP6 EP6] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP7 EP7] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP0 EP0] Init torch distributed ends. elapsed=10.55 s, mem usage=0.40 GB
+[2026-08-27 04:36:48 TP7 EP7] Init torch distributed ends. elapsed=10.61 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP6 EP6] Init torch distributed ends. elapsed=10.88 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP4 EP4] Init torch distributed ends. elapsed=10.77 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP3 EP3] Init torch distributed ends. elapsed=10.60 s, mem usage=0.40 GB
+[2026-08-27 04:36:48 TP1 EP1] Init torch distributed ends. elapsed=10.90 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP5 EP5] Init torch distributed ends. elapsed=10.47 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP2 EP2] Init torch distributed ends. elapsed=10.50 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP7 EP7] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP4 EP4] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP1 EP1] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP6 EP6] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP7 EP7] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP4 EP4] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP5 EP5] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP6 EP6] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP3 EP3] Load weight begin. avail mem=82.19 GB
+[2026-08-27 04:36:48 TP0 EP0] Load weight begin. avail mem=82.19 GB
+[2026-08-27 04:36:48 TP5 EP5] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP2 EP2] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP4 EP4] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP7 EP7] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP6 EP6] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP4 EP4] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP7 EP7] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP6 EP6] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP5 EP5] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP5 EP5] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:50 TP1 EP1] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP0 EP0] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP2 EP2] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP3 EP3] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP5 EP5] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP4 EP4] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP6 EP6] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP7 EP7] multimem all-gather disabled because the TP group spans across nodes.
+
Multi-thread loading shards:   0% Completed | 0/96 [00:00<?, ?it/s]
Multi-thread loading shards:   1% Completed | 1/96 [00:00<00:24,  3.85it/s]
Multi-thread loading shards:   2% Completed | 2/96 [00:07<07:13,  4.61s/it]
Multi-thread loading shards:   3% Completed | 3/96 [00:09<04:44,  3.06s/it]
Multi-thread loading shards:   4% Completed | 4/96 [00:10<03:33,  2.32s/it]
Multi-thread loading shards:   5% Completed | 5/96 [00:11<02:54,  1.92s/it]
Multi-thread loading shards:   6% Completed | 6/96 [00:12<02:29,  1.66s/it]
Multi-thread loading shards:   7% Completed | 7/96 [00:13<02:13,  1.50s/it]
Multi-thread loading shards:   8% Completed | 8/96 [00:15<02:04,  1.41s/it]
Multi-thread loading shards:   9% Completed | 9/96 [00:16<01:56,  1.34s/it]
Multi-thread loading shards:  10% Completed | 10/96 [00:17<01:47,  1.25s/it]
Multi-thread loading shards:  11% Completed | 11/96 [00:18<01:44,  1.23s/it]
Multi-thread loading shards:  12% Completed | 12/96 [00:19<01:43,  1.23s/it]
Multi-thread loading shards:  14% Completed | 13/96 [00:21<01:59,  1.44s/it]
Multi-thread loading shards:  15% Completed | 14/96 [00:22<01:51,  1.36s/it]
Multi-thread loading shards:  16% Completed | 15/96 [00:24<01:45,  1.30s/it]
Multi-thread loading shards:  17% Completed | 16/96 [00:25<01:42,  1.28s/it]
Multi-thread loading shards:  18% Completed | 17/96 [00:26<01:39,  1.26s/it]
Multi-thread loading shards:  19% Completed | 18/96 [00:27<01:36,  1.24s/it]
Multi-thread loading shards:  20% Completed | 19/96 [00:28<01:32,  1.20s/it]
Multi-thread loading shards:  21% Completed | 20/96 [00:29<01:30,  1.19s/it]
Multi-thread loading shards:  22% Completed | 21/96 [00:31<01:29,  1.19s/it]
Multi-thread loading shards:  23% Completed | 22/96 [00:32<01:27,  1.18s/it]
Multi-thread loading shards:  24% Completed | 23/96 [00:33<01:26,  1.18s/it]
Multi-thread loading shards:  25% Completed | 24/96 [00:34<01:23,  1.16s/it]
Multi-thread loading shards:  26% Completed | 25/96 [00:35<01:20,  1.14s/it]
Multi-thread loading shards:  27% Completed | 26/96 [00:36<01:18,  1.13s/it]
Multi-thread loading shards:  28% Completed | 27/96 [00:37<01:19,  1.15s/it]
Multi-thread loading shards:  29% Completed | 28/96 [00:39<01:18,  1.16s/it]
Multi-thread loading shards:  30% Completed | 29/96 [00:40<01:17,  1.16s/it]
Multi-thread loading shards:  31% Completed | 30/96 [00:41<01:17,  1.18s/it]
Multi-thread loading shards:  32% Completed | 31/96 [00:42<01:16,  1.18s/it]
Multi-thread loading shards:  33% Completed | 32/96 [00:44<01:30,  1.42s/it]
Multi-thread loading shards:  34% Completed | 33/96 [00:45<01:23,  1.33s/it]
Multi-thread loading shards:  35% Completed | 34/96 [00:46<01:17,  1.26s/it]
Multi-thread loading shards:  36% Completed | 35/96 [00:48<01:14,  1.23s/it]
Multi-thread loading shards:  38% Completed | 36/96 [00:49<01:13,  1.22s/it]
Multi-thread loading shards:  39% Completed | 37/96 [00:50<01:11,  1.21s/it]
Multi-thread loading shards:  40% Completed | 38/96 [00:51<01:09,  1.20s/it]
Multi-thread loading shards:  41% Completed | 39/96 [00:52<01:07,  1.18s/it]
Multi-thread loading shards:  42% Completed | 40/96 [00:53<01:05,  1.18s/it]
Multi-thread loading shards:  43% Completed | 41/96 [00:55<01:04,  1.17s/it]
Multi-thread loading shards:  44% Completed | 42/96 [00:56<01:02,  1.17s/it]
Multi-thread loading shards:  45% Completed | 43/96 [00:57<01:03,  1.19s/it]
Multi-thread loading shards:  46% Completed | 44/96 [00:58<01:01,  1.18s/it]
Multi-thread loading shards:  47% Completed | 45/96 [00:59<00:57,  1.13s/it]
Multi-thread loading shards:  48% Completed | 46/96 [01:00<00:57,  1.15s/it]
Multi-thread loading shards:  49% Completed | 47/96 [01:02<00:57,  1.17s/it]
Multi-thread loading shards:  50% Completed | 48/96 [01:03<00:55,  1.15s/it]
Multi-thread loading shards:  51% Completed | 49/96 [01:04<00:53,  1.13s/it]
Multi-thread loading shards:  52% Completed | 50/96 [01:06<01:01,  1.34s/it]
Multi-thread loading shards:  53% Completed | 51/96 [01:07<00:58,  1.29s/it]
Multi-thread loading shards:  54% Completed | 52/96 [01:08<00:55,  1.27s/it]
Multi-thread loading shards:  55% Completed | 53/96 [01:09<00:53,  1.24s/it]
Multi-thread loading shards:  56% Completed | 54/96 [01:10<00:50,  1.21s/it]
Multi-thread loading shards:  57% Completed | 55/96 [01:12<00:49,  1.22s/it]
Multi-thread loading shards:  58% Completed | 56/96 [01:13<00:48,  1.22s/it]
Multi-thread loading shards:  59% Completed | 57/96 [01:14<00:47,  1.21s/it]
Multi-thread loading shards:  60% Completed | 58/96 [01:15<00:45,  1.20s/it]
Multi-thread loading shards:  61% Completed | 59/96 [01:16<00:44,  1.19s/it]
Multi-thread loading shards:  62% Completed | 60/96 [01:17<00:43,  1.20s/it]
Multi-thread loading shards:  64% Completed | 61/96 [01:19<00:42,  1.21s/it]
Multi-thread loading shards:  65% Completed | 62/96 [01:20<00:40,  1.19s/it]
Multi-thread loading shards:  66% Completed | 63/96 [01:21<00:40,  1.21s/it]
Multi-thread loading shards:  67% Completed | 64/96 [01:22<00:38,  1.20s/it]
Multi-thread loading shards:  68% Completed | 65/96 [01:24<00:37,  1.20s/it]
Multi-thread loading shards:  69% Completed | 66/96 [01:25<00:36,  1.20s/it]
Multi-thread loading shards:  70% Completed | 67/96 [01:26<00:35,  1.22s/it]
Multi-thread loading shards:  71% Completed | 68/96 [01:28<00:41,  1.48s/it]
Multi-thread loading shards:  72% Completed | 69/96 [01:29<00:37,  1.39s/it]
Multi-thread loading shards:  73% Completed | 70/96 [01:30<00:34,  1.34s/it]
Multi-thread loading shards:  74% Completed | 71/96 [01:32<00:32,  1.29s/it]
Multi-thread loading shards:  75% Completed | 72/96 [01:33<00:31,  1.30s/it]
Multi-thread loading shards:  76% Completed | 73/96 [01:34<00:29,  1.29s/it]
Multi-thread loading shards:  77% Completed | 74/96 [01:35<00:27,  1.27s/it]
Multi-thread loading shards:  78% Completed | 75/96 [01:37<00:26,  1.26s/it]
Multi-thread loading shards:  79% Completed | 76/96 [01:38<00:24,  1.24s/it]
Multi-thread loading shards:  80% Completed | 77/96 [01:39<00:23,  1.23s/it]
Multi-thread loading shards:  81% Completed | 78/96 [01:40<00:21,  1.19s/it]
Multi-thread loading shards:  82% Completed | 79/96 [01:41<00:20,  1.20s/it]
Multi-thread loading shards:  83% Completed | 80/96 [01:43<00:19,  1.20s/it]
Multi-thread loading shards:  84% Completed | 81/96 [01:44<00:18,  1.21s/it]
Multi-thread loading shards:  85% Completed | 82/96 [01:45<00:16,  1.20s/it]
Multi-thread loading shards:  86% Completed | 83/96 [01:46<00:15,  1.17s/it]
Multi-thread loading shards:  88% Completed | 84/96 [01:47<00:14,  1.17s/it]
Multi-thread loading shards:  89% Completed | 85/96 [01:48<00:12,  1.10s/it]
Multi-thread loading shards:  90% Completed | 86/96 [01:49<00:10,  1.05s/it]
Multi-thread loading shards:  91% Completed | 87/96 [01:50<00:09,  1.02s/it]
Multi-thread loading shards:  92% Completed | 88/96 [01:51<00:07,  1.01it/s]
Multi-thread loading shards:  93% Completed | 89/96 [01:52<00:06,  1.03it/s]
Multi-thread loading shards:  94% Completed | 90/96 [01:53<00:05,  1.04it/s]
Multi-thread loading shards:  95% Completed | 91/96 [01:54<00:04,  1.06it/s]
Multi-thread loading shards:  96% Completed | 92/96 [01:55<00:03,  1.06it/s]
Multi-thread loading shards:  97% Completed | 93/96 [01:56<00:02,  1.08it/s]
Multi-thread loading shards:  98% Completed | 94/96 [01:56<00:01,  1.46it/s]
Multi-thread loading shards: 100% Completed | 96/96 [01:56<00:00,  2.28it/s]
Multi-thread loading shards: 100% Completed | 96/96 [01:56<00:00,  1.21s/it]
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:58 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP6 EP6] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP0 EP0] Load weight end. elapsed=130.28 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP6 EP6] Load weight end. elapsed=130.29 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP3 EP3] Load weight end. elapsed=130.28 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP2 EP2] Load weight end. elapsed=130.32 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP5 EP5] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP5 EP5] Load weight end. elapsed=130.38 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP4 EP4] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP4 EP4] Load weight end. elapsed=130.48 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP7 EP7] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP7 EP7] Load weight end. elapsed=130.48 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:39:01 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:39:01 TP1 EP1] Load weight end. elapsed=132.81 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:39:01 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP6 EP6] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP6 EP6] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP5 EP5] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP5 EP5] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP7 EP7] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP7 EP7] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP4 EP4] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP4 EP4] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:02 TP0 EP0] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP7 EP7] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP6 EP6] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP5 EP5] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP1 EP1] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP4 EP4] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP3 EP3] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP2 EP2] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP7 EP7] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP4 EP4] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP6 EP6] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP5 EP5] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP0 EP0] Memory pool end. avail mem=9.18 GB
+[2026-08-27 04:39:02 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP1 EP1] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP6 EP6] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP5 EP5] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP7 EP7] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP6 EP6] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP5 EP5] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP7 EP7] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP4 EP4] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP4 EP4] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP2 EP2] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP3 EP3] Memory pool end. avail mem=9.18 GB
+[2026-08-27 04:39:02 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:39:02 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:39:02 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP0 EP0] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.17 GB
+[2026-08-27 04:39:02 TP6 EP6] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP6 EP6] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP1 EP1] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP5 EP5] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP7 EP7] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP5 EP5] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP7 EP7] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:39:02 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP3 EP3] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.17 GB
+[2026-08-27 04:39:02 TP4 EP4] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP4 EP4] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:39:02 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP2 EP2] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+
  0%|          | 0/6 [00:00<?, ?it/s]
Capturing batches (bs=16 avail_mem=9.00 GB):   0%|          | 0/6 [00:00<?, ?it/s]
Capturing batches (bs=16 avail_mem=9.00 GB):  17%|█▋        | 1/6 [03:43<18:38, 223.66s/it]
Capturing batches (bs=12 avail_mem=8.84 GB):  17%|█▋        | 1/6 [03:43<18:38, 223.66s/it]
Capturing batches (bs=12 avail_mem=8.84 GB):  33%|███▎      | 2/6 [03:44<06:11, 92.84s/it] 
Capturing batches (bs=8 avail_mem=8.78 GB):  33%|███▎      | 2/6 [03:44<06:11, 92.84s/it] 
Capturing batches (bs=8 avail_mem=8.78 GB):  50%|█████     | 3/6 [03:49<02:36, 52.32s/it]
Capturing batches (bs=4 avail_mem=8.73 GB):  50%|█████     | 3/6 [03:49<02:36, 52.32s/it]
Capturing batches (bs=4 avail_mem=8.73 GB):  67%|██████▋   | 4/6 [03:49<01:03, 32.00s/it]
Capturing batches (bs=2 avail_mem=8.68 GB):  67%|██████▋   | 4/6 [03:49<01:03, 32.00s/it]
Capturing batches (bs=2 avail_mem=8.68 GB):  83%|████████▎ | 5/6 [03:50<00:20, 20.77s/it]
Capturing batches (bs=1 avail_mem=8.62 GB):  83%|████████▎ | 5/6 [03:50<00:20, 20.77s/it]
Capturing batches (bs=1 avail_mem=8.62 GB): 100%|██████████| 6/6 [03:52<00:00, 14.36s/it]
Capturing batches (bs=1 avail_mem=8.62 GB): 100%|██████████| 6/6 [03:52<00:00, 38.78s/it]
+[2026-08-27 04:42:56 TP4 EP4] Capture target decode CUDA graph end. elapsed=233.75 s, mem usage=0.60 GB, avail mem=8.52 GB.
+[2026-08-27 04:42:56 TP4 EP4] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP3 EP3] Capture target decode CUDA graph end. elapsed=233.75 s, mem usage=0.62 GB, avail mem=8.55 GB.
+[2026-08-27 04:42:56 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP0 EP0] Capture target decode CUDA graph end. elapsed=233.77 s, mem usage=0.60 GB, avail mem=8.56 GB.
+[2026-08-27 04:42:56 TP0 EP0] max_total_num_tokens=440256, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=181, context_len=1048576, available_gpu_mem=8.56 GB
+[2026-08-27 04:42:56 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP7 EP7] Capture target decode CUDA graph end. elapsed=233.77 s, mem usage=0.62 GB, avail mem=8.51 GB.
+[2026-08-27 04:42:56 TP7 EP7] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP6 EP6] Capture target decode CUDA graph end. elapsed=233.86 s, mem usage=0.56 GB, avail mem=8.59 GB.
+[2026-08-27 04:42:56 TP6 EP6] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP1 EP1] Capture target decode CUDA graph end. elapsed=233.86 s, mem usage=0.58 GB, avail mem=8.58 GB.
+[2026-08-27 04:42:56 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP2 EP2] Capture target decode CUDA graph end. elapsed=233.86 s, mem usage=0.56 GB, avail mem=8.59 GB.
+[2026-08-27 04:42:56 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP5 EP5] Capture target decode CUDA graph end. elapsed=233.90 s, mem usage=0.58 GB, avail mem=8.58 GB.
+[2026-08-27 04:42:56 TP5 EP5] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+W0827 04:42:57.550421  2569 memory_location.cpp:73] Failed to get NUMA node, addr: 0x599a0800, len: 23168: Operation not permitted [1]
+W0827 04:42:57.550521  2570 memory_location.cpp:73] Failed to get NUMA node, addr: 0x599cb380, len: 23168: Operation not permitted [1]
+W0827 04:42:57.550542  2571 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59e1b540, len: 23168: Operation not permitted [1]
+W0827 04:42:57.550628  2572 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59c8a2c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.550725  2573 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5be7a240, len: 185344: Operation not permitted [1]
+W0827 04:42:57.550769  2574 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5bea7680, len: 185344: Operation not permitted [1]
+W0827 04:42:57.550900  2575 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59cc48c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.551011  2576 memory_location.cpp:73] Failed to get NUMA node, addr: 0x45f39fc0, len: 46336: Operation not permitted [1]
+W0827 04:42:57.551070  2577 memory_location.cpp:73] Failed to get NUMA node, addr: 0x599bfbc0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.551105  2578 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59c69c80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.580384  2676 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59c988c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.580493  2678 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59a30c40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.580636  2677 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59d06340, len: 23168: Operation not permitted [1]
+W0827 04:42:57.580636  2679 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5bd91640, len: 185344: Operation not permitted [1]
+W0827 04:42:57.580685  2675 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59e30d80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.580811  2681 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59f53440, len: 23168: Operation not permitted [1]
+W0827 04:42:57.580950  2680 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5bdbea80, len: 185344: Operation not permitted [1]
+W0827 04:42:57.581368  2686 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59dca000, len: 23168: Operation not permitted [1]
+W0827 04:42:57.581421  2682 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5987fc00, len: 46336: Operation not permitted [1]
+W0827 04:42:57.581741  2684 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59e10480, len: 23168: Operation not permitted [1]
+W0827 04:42:57.597819  2845 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59224900, len: 23168: Operation not permitted [1]
+W0827 04:42:57.598558  2839 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5917e740, len: 23168: Operation not permitted [1]
+W0827 04:42:57.598562  2849 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58e6ce00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.598762  2838 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58e982c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.598834  2841 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5b32ec00, len: 185344: Operation not permitted [1]
+W0827 04:42:57.598897  2834 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5927b680, len: 23168: Operation not permitted [1]
+W0827 04:42:57.598922  2843 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5b35c040, len: 185344: Operation not permitted [1]
+W0827 04:42:57.598953  2836 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5939fc40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.599097  2850 memory_location.cpp:73] Failed to get NUMA node, addr: 0x592d3fc0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.599287  2847 memory_location.cpp:73] Failed to get NUMA node, addr: 0x453f4c40, len: 46336: Operation not permitted [1]
+W0827 04:42:57.612057  2918 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59f4a140, len: 23168: Operation not permitted [1]
+W0827 04:42:57.612175  2919 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59f098c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.612243  2920 memory_location.cpp:73] Failed to get NUMA node, addr: 0x46045dc0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.612710  2925 memory_location.cpp:73] Failed to get NUMA node, addr: 0x46106480, len: 46336: Operation not permitted [1]
+W0827 04:42:57.612715  2924 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59a7ab40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.612880  2922 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5bf25300, len: 185344: Operation not permitted [1]
+W0827 04:42:57.612898  2927 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59e8a9c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.612931  2921 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59fec840, len: 23168: Operation not permitted [1]
+W0827 04:42:57.612948  2923 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5bf52740, len: 185344: Operation not permitted [1]
+W0827 04:42:57.612996  2926 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59edf040, len: 23168: Operation not permitted [1]
+W0827 04:42:57.618498  2975 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a4cda80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.618582  2974 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a3878c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.618582  2973 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a3abc80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.618669  2976 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5c40da00, len: 185344: Operation not permitted [1]
+W0827 04:42:57.618669  2978 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59f66a40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.618741  2977 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5c43ae40, len: 185344: Operation not permitted [1]
+W0827 04:42:57.618854  2979 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59ce6c00, len: 46336: Operation not permitted [1]
+W0827 04:42:57.618898  2980 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a2ebe80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.619125  2982 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a37d0c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.632592  2972 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a371340, len: 23168: Operation not permitted [1]
+W0827 04:42:57.776260  3310 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58ee7080, len: 23168: Operation not permitted [1]
+W0827 04:42:57.776290  3311 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5854f480, len: 23168: Operation not permitted [1]
+W0827 04:42:57.776294  3312 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58edd900, len: 23168: Operation not permitted [1]
+W0827 04:42:57.776240  3309 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58f70e40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.776324  3313 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a63db40, len: 185344: Operation not permitted [1]
+W0827 04:42:57.776337  3314 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a66af80, len: 185344: Operation not permitted [1]
+W0827 04:42:57.776417  3315 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58e5ec40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.776444  3316 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58109800, len: 46336: Operation not permitted [1]
+W0827 04:42:57.776451  3317 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58ed2cc0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.776649  3318 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58316800, len: 23168: Operation not permitted [1]
+W0827 04:42:57.812515  3457 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58919a80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.812533  3458 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5892c840, len: 23168: Operation not permitted [1]
+W0827 04:42:57.812597  3460 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58063380, len: 23168: Operation not permitted [1]
+W0827 04:42:57.812603  3459 memory_location.cpp:73] Failed to get NUMA node, addr: 0x588ed280, len: 23168: Operation not permitted [1]
+W0827 04:42:57.812654  3461 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59fc0080, len: 185344: Operation not permitted [1]
+W0827 04:42:57.812734  3462 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59fed4c0, len: 185344: Operation not permitted [1]
+W0827 04:42:57.812816  3463 memory_location.cpp:73] Failed to get NUMA node, addr: 0x57ebf300, len: 23168: Operation not permitted [1]
+W0827 04:42:57.812877  3464 memory_location.cpp:73] Failed to get NUMA node, addr: 0x44a36bc0, len: 46336: Operation not permitted [1]
+W0827 04:42:57.812991  3466 memory_location.cpp:73] Failed to get NUMA node, addr: 0x44982040, len: 23168: Operation not permitted [1]
+W0827 04:42:57.812996  3465 memory_location.cpp:73] Failed to get NUMA node, addr: 0x588ba900, len: 23168: Operation not permitted [1]
+W0827 04:42:57.851508  3605 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5888adc0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.851550  3606 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58775700, len: 23168: Operation not permitted [1]
+W0827 04:42:57.851619  3607 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58aca800, len: 23168: Operation not permitted [1]
+W0827 04:42:57.851691  3608 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5886ea80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.851799  3609 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5aaacb40, len: 185344: Operation not permitted [1]
+W0827 04:42:57.851871  3611 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58ad7b00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.851959  3610 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5aad9f80, len: 185344: Operation not permitted [1]
+W0827 04:42:57.851996  3612 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58aa8840, len: 46336: Operation not permitted [1]
+W0827 04:42:57.852005  3614 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58ca01c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.852105  3613 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5879a9c0, len: 23168: Operation not permitted [1]
+[2026-08-27 04:43:01] INFO:     174.1.60.1:55082 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:43:01] INFO:     127.0.0.1:38530 - "GET /model_info HTTP/1.1" 200 OK
+[2026-08-27 04:43:02] INFO:     174.1.60.1:55098 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:43:03] INFO:     127.0.0.1:38536 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 04:43:04] INFO:     174.1.60.1:55110 - "GET /health HTTP/1.1" 200 OK
+[2026-08-27 04:43:25] INFO:     174.1.60.1:41170 - "GET /health HTTP/1.1" 200 OK
+[2026-08-27 04:48:11] INFO:     174.1.60.5:38682 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 04:43:00] INFO:     Started server process [1]
+[2026-08-27 04:43:00] INFO:     Waiting for application startup.
+[2026-08-27 04:43:00] Calling super().encode with {'add_special_tokens': False}
+[2026-08-27 04:43:00] Calling super().encode with {'add_special_tokens': False}
+[2026-08-27 04:43:00] INFO:     Application startup complete.
+[2026-08-27 04:43:00] INFO:     Uvicorn running on http://0.0.0.0:30000 (Press CTRL+C to quit)
+[2026-08-27 04:43:01] Start of pd disaggregation warmup ...
+[2026-08-27 04:43:03] Disaggregation warmup requests completed for all 1 DP ranks
+[2026-08-27 04:43:03] End of disaggregation warmup
+[2026-08-27 04:43:03] The server is fired up and ready to roll!
+I0827 04:48:13.997220  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:13.997509  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.013371  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.013677  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.042984  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.043259  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.046793  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.047015  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.057778  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.067556  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.076747  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.095343  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.099277  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.105573  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.105803  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.106056  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.109412  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.109645  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.109865  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.110114  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.110388  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.110597  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.110783  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.114174  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.114495  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.120798  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.136216  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.142400  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.142663  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.153999  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.154243  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.154534  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.154755  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.154990  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.158114  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.158325  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.170076  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.170320  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.170512  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.170779  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.185034  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.197149  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.197443  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.197665  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.197889  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.203205  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.203492  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.203737  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.203975  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.204202  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.215955  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.216298  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.216583  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.216789  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.222101  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.223455  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.232357  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.233336  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.233599  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.233934  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.248603  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.248853  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.249085  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.250525  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.252888  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.253137  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.253366  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.253639  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.256778  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.258342  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.259446  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.259649  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.259908  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.260114  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.260375  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.260576  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.262905  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.263399  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.263654  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.263896  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.264135  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.264392  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.264699  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.264971  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.276409  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.276680  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.280252  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.281527  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.289309  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.289914  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.290587  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.299579  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.299823  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.303650  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.303859  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.304064  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.307004  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.307231  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.307541  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.307801  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.308017  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.308123  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.308235  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.308485  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.308728  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.308988  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.309187  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.309360  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.309405  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.309656  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.309904  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.310132  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.310344  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.312103  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.312309  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.314198  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.315554  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.316985  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.320315  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.327411  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.337924  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.339138  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.344900  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.346967  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.348446  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.361662  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.361877  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.362105  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.365690  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.366845  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.373291  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.373983  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.374544  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.391561  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.391779  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.392053  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.392257  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.392505  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.392726  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.393024  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.393260  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.393772  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.395025  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.396075  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.396298  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.396525  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.396822  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.397025  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.397225  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.403301  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.404704  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.407682  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.417869  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.418104  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.418273  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.421886  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.422092  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.422293  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.422566  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.422824  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.424790  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.427910  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.429622  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.433352  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.433625  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.433885  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.434249  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.434464  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.436031  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.444931  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.445168  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.445376  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.445645  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.448356  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.448545  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.448810  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.449036  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.449236  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.450745  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.453661  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.455060  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.456887  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.457075  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.458276  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.463183  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.464545  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.473290  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.473594  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.473889  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.474606  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.478461  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.478672  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.478849  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.485463  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.485714  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.486359  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.486591  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.486837  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.486868  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.487639  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.492800  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.494125  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.496006  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.496238  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.496584  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.504025  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.504601  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.504830  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.505040  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.505227  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.505416  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.505596  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.505838  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.506070  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.506418  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.506714  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.506938  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.507185  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.507364  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.507618  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.507850  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.508181  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.509447  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.510848  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.510802  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.511059  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.511301  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.512179  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.513973  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.520185  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.521516  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.528406  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.528625  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.528868  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.532759  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.535007  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.535238  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.535440  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.535620  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.536624  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.538493  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.541709  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.541929  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.543128  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.543134  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.543255  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.543396  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.543586  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.543852  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.544047  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.544274  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.544514  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.544595  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.544750  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.544992  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.545264  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.545506  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.545732  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.549163  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.556982  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.557228  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.557428  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.558394  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.559183  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.563414  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.564719  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.564847  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.566068  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.568861  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.574138  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.575349  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.583217  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.583469  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.583667  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.584623  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.588591  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.588935  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.589197  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.590059  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.590425  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.591902  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.599382  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.599529  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.599663  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.599762  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.600073  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.600317  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.600823  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.605185  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.605659  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.605909  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.609126  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.610448  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.611255  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.614173  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.616300  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.616987  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.617319  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.617666  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.620188  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.620463  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.623186  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.623404  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.626320  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.629146  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.632516  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.632800  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.633054  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.634070  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.634683  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.637434  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.640316  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.640353  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.640614  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.641824  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.643349  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.646051  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.658253  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.662520  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.663751  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.663796  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.668447  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.674060  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.683264  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.683470  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.683725  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.684523  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.684583  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.684775  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.685092  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.685344  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.685595  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.685825  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.686066  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.694078  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.694272  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.694502  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.694731  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.696010  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.696235  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.696461  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.697254  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.702469  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.702711  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.702946  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.703189  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.712618  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.716657  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.717875  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.718667  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.718958  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.719236  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.722149  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.724561  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.724784  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.724995  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.730228  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.730405  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.730648  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.730953  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.731205  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.731441  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.736352  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.736553  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.736785  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.737946  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.739337  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.739604  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.743849  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.745306  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.747385  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.750119  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.755378  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.755673  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.755957  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.756151  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.756417  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.756649  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.756873  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.757055  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.757272  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.762418  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.762653  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.767700  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.767921  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.768116  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.769114  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.805779  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.806072  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.806393  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.816823  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.817051  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.817302  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.817519  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.822906  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.823174  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.823410  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.823618  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.828871  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.831584  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.834362  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.834635  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.845114  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.847841  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.853120  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.853320  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.853577  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.853799  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.856580  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.859321  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.859536  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.864822  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.875097  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.895438  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.895658  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.903518  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.903801  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.904052  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.904273  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.909722  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.911178  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.912268  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.912475  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.912720  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.912957  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.913183  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.913419  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.913677  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.913909  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.914156  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.914402  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.914625  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.917008  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.919989  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.920225  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.922858  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.923017  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.935199  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.935660  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.935937  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.939235  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.941046  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.943796  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.944716  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.946557  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.947388  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.949306  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.949542  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.949780  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.950085  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.950371  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.950598  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.950831  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.951049  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.951236  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.951502  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.955791  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.956029  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.964305  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.964589  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.964807  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.965030  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.965714  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.969712  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.971029  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.976545  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.978840  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.982290  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.993181  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.993423  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.993613  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.995042  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.995337  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.995555  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.996627  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.007792  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.008044  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.008275  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.009122  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.013960  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.014259  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.014477  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.014698  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.014875  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.015080  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.015322  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.015535  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.023283  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.024626  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.028729  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.028869  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.029064  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.029294  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.029582  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.030203  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.035056  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.047416  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.047739  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.047941  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.048739  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.054301  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.055744  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.056360  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.056833  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.057034  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.057294  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.057492  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.057713  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.057981  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.058207  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.058401  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.058617  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.069922  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.071324  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.075461  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.081681  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.093765  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.093961  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.094169  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.095338  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.101928  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.102181  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.102407  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.102629  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.102851  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.103041  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.103250  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.103484  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.103757  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.103947  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.104204  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.104463  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.104683  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.104951  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.105253  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.105489  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.105674  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.105890  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.106104  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.106318  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.107211  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.120297  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.124440  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.125627  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.131145  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.131403  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.131698  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.131965  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.132176  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.132416  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.132588  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.132663  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.132896  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.133141  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.133297  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.146152  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.146390  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.146620  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.146828  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.147426  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.156941  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.157151  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.178486  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.178771  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.179106  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.179323  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.208334  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.208683  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.208947  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.209216  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.209455  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.209815  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.210053  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.210210  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.210412  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.210608  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.210822  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.221958  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.222180  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.222395  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.223281  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.225317  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.226866  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.230994  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.237720  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.242835  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.243105  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.243345  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.243551  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.243834  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.244047  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.244271  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.250334  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.250538  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.250738  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.251941  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.257243  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.258617  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.273547  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.277629  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.278937  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.281635  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.281862  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.282078  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.282316  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.282523  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.282783  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.283066  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.283294  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.283550  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.283828  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.284119  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.284336  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.284598  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.285751  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.287382  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.298511  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.298730  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.298920  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.299866  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.301535  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.302943  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.305897  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.306208  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.306404  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.306623  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.306824  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.307075  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.307155  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.307289  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.307511  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.310241  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.310458  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.310705  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.311606  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.313308  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.320852  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.321092  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.321306  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.321498  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.322360  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.323149  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.323376  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.344715  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.362433  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.365234  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.365482  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.365670  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.365831  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.366075  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.366329  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.366532  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.369321  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.379413  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.379633  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.379848  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.380110  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.380908  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.388870  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.389055  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.393224  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.396981  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.397784  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.398291  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.404456  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.405587  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.405781  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.405988  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.416702  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.416770  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.416965  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.417004  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.417227  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.418077  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.419777  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.425088  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.425359  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.425642  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.425822  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.428227  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.428457  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.428719  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.429623  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.431046  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.433805  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.434010  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.441807  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.442006  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.442276  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.447495  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.447705  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.447944  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.453174  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.453361  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.456089  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.456282  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.457403  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.487757  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.487957  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.488181  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.488385  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.488687  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.489209  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.497211  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.497412  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.498611  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.506928  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.507129  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.508323  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.533521  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.533809  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.533984  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.534192  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+[2026-08-27 04:48:16] INFO:     174.1.60.5:49356 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 04:48:16] INFO:     174.1.60.5:49370 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 04:57:13] INFO:     174.1.60.5:34722 - "POST /generate HTTP/1.1" 200 OK
+I0827 04:48:15.534411  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.534649  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.534850  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.535123  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.535336  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.538051  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.553102  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.555838  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.556046  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.556244  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.556519  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.556715  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.556900  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.559612  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.567348  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.570070  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.570282  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.570564  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.571498  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+[2026-08-27 04:57:14] INFO:     174.1.60.5:34732 - "GET /server_info HTTP/1.1" 200 OK
+I0827 04:57:14.461794  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.462142  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.463397  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.464907  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.466337  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.472958  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.474522  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.477468  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.477753  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.479032  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.480643  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.498212  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.498531  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.498838  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.499125  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.499409  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.499795  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.505314  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.517073  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.518641  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.527722  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.529148  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.534837  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+[2026-08-27 04:57:14] INFO:     174.1.60.5:34748 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:00:07] INFO:     174.1.60.5:40816 - "POST /generate HTTP/1.1" 200 OK
+I0827 05:04:21.299574  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.301399  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.307986  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.310060  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.311648  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.317925  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.321113  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.321422  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.322642  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.325122  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.326697  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.332526  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.332682  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.334165  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.340161  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.341799  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.343374  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.345765  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.347450  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.349627  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.350939  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.352005  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.355481  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+[2026-08-27 05:04:22] INFO:     174.1.60.5:33422 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:04:22] INFO:     174.1.60.5:33424 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:04:42] INFO:     174.1.60.5:58198 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:05:38] INFO:     174.1.60.5:48352 - "POST /generate HTTP/1.1" 200 OK
+I0827 05:04:21.355765  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.357254  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.357306  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.358845  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.363112  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.363651  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.368119  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.373224  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.374758  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.379626  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.379845  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.380074  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.381069  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.382846  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.384393  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.386190  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.387002  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.387748  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.388473  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.388860  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.399398  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.399649  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.400830  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.402611  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.402901  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.404011  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.405441  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.408586  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.409112  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.414987  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.416474  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.418708  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.420038  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.421981  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.424405  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.437287  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.437530  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.438683  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.443255  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+[2026-08-27 05:04:22 TP0 EP0] Decode batch, #running-req: 1, #full token: 0, full token usage: 0.00, mamba num: 0, mamba usage: 0.00, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 0.03, #queue-req: 0
+I0827 05:04:41.398729  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.399127  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.400347  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.401475  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.401768  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.403021  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.413998  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.415591  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.421803  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.576606  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.576977  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.578258  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+[2026-08-27 05:07:16] INFO:     174.1.60.5:48172 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:07:33] INFO:     174.1.60.5:48616 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:07:40] INFO:     174.1.60.5:48620 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:07:48] INFO:     174.1.60.5:33014 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:07:56] INFO:     174.1.60.5:48046 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:08:03] INFO:     174.1.60.5:43662 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:08:11] INFO:     174.1.60.5:50736 - "POST /generate HTTP/1.1" 200 OK
+I0827 05:05:39.600133  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.600483  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.600749  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.601066  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.601655  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.607283  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.615131  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.620574  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.620859  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.621142  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.622053  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.626708  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.148217  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.148566  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.149849  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.155894  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.657902  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.659292  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.665387  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.682516  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.682862  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.683116  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.683411  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.683794  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.684041  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.690148  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.690267  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.691737  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.698587  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.699856  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.700106  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.701548  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.704351  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.705502  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.705786  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.706010  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.706919  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.708040  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.713306  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.714720  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+[2026-08-27 05:08:18] INFO:     174.1.60.5:50748 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:08:26] INFO:     174.1.60.5:48502 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:08:26] INFO:     174.1.60.5:48506 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:11:44] INFO:     174.1.60.5:54256 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:11:46] INFO:     174.1.60.5:54260 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:11:46] INFO:     174.1.60.5:54274 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:14:43] INFO:     174.1.60.5:52586 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:14:43] INFO:     174.1.60.5:52600 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:14:43] INFO:     174.1.60.5:52616 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:17:40] INFO:     174.1.60.5:47526 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:17:47] INFO:     174.1.60.5:33870 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:17:55] INFO:     174.1.60.5:33666 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:18:02] INFO:     174.1.60.5:59970 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:18:10] INFO:     174.1.60.5:59984 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:18:18] INFO:     174.1.60.5:35200 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:18:25] INFO:     174.1.60.5:47324 - "POST /generate HTTP/1.1" 200 OK
+I0827 05:08:18.720657  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.722337  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.726626  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+[2026-08-27 05:11:46 TP0 EP0] Decode batch, #running-req: 1, #full token: 0, full token usage: 0.00, mamba num: 0, mamba usage: 0.00, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 0.09, #queue-req: 0
+I0827 05:14:42.491809  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.493278  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.499468  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.158722  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.159102  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.160380  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.167616  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.169173  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+[2026-08-27 05:18:33] INFO:     174.1.60.5:60276 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:18:40] INFO:     174.1.60.5:60282 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:18:40] INFO:     174.1.60.5:60286 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:21:37] INFO:     174.1.60.5:39028 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:37] INFO:     174.1.60.5:39042 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:37] INFO:     174.1.60.5:39052 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:37] INFO:     174.1.60.5:39066 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:37] INFO:     174.1.60.5:39070 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:37] INFO:     174.1.60.5:39084 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:37] INFO:     174.1.60.5:39092 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:37] INFO:     174.1.60.5:39102 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:51] INFO:     174.1.60.5:33930 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:53] INFO:     174.1.60.5:33942 - "POST /generate HTTP/1.1" 200 OK
+I0827 05:18:33.173859  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.185045  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.185313  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.186507  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.192727  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:47.564407  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:47.575759  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:47.585628  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:47.597601  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:47.599035  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:48.554785  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:48.565164  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:49.333541  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:49.343402  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:49.352372  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:49.361497  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:50.265654  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:50.274874  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:50.283715  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.110808  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.124187  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.162127  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.173472  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.005925  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.016026  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.024489  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.033867  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.856267  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.867897  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:53.712086  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:53.724234  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:54.551172  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:54.564998  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:55.326069  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:55.338083  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+[2026-08-27 05:21:55] INFO:     174.1.60.5:33946 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:57] INFO:     174.1.60.5:33950 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:57] INFO:     174.1.60.5:33960 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:57] INFO:     174.1.60.5:33968 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:59] INFO:     174.1.60.5:33980 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:00] INFO:     174.1.60.5:33990 - "POST /generate HTTP/1.1" 200 OK
+I0827 05:21:56.186933  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:56.188442  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:56.200929  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:57.056439  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:57.056681  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:57.058399  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:57.654704  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:05.685521  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:05.687002  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:06.720834  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:07.458693  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:08.386162  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:08.387703  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+[2026-08-27 05:22:11] INFO:     174.1.60.5:51774 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:13] INFO:     174.1.60.5:51778 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:15] INFO:     174.1.60.5:51794 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:17] INFO:     174.1.60.5:51800 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:18] INFO:     174.1.60.5:51812 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:18] INFO:     174.1.60.5:51824 - "POST /generate HTTP/1.1" 200 OK
+I0827 05:22:08.395984  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:10.976925  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:16.233013  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:17.990159  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:17.991626  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:17.999158  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:18.227823  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:18.250644  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:18.360224  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+[2026-08-27 05:22:20] INFO:     174.1.60.5:51834 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:21] INFO:     174.1.60.5:46458 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:31] INFO:     174.1.60.5:36970 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:34] INFO:     174.1.60.5:36980 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:35] INFO:     174.1.60.5:36990 - "POST /generate HTTP/1.1" 200 OK
+I0827 05:22:28.065888  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:36.003260  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:36.977586  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+[2026-08-27 05:22:37] INFO:     174.1.60.5:36996 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:38] INFO:     174.1.60.5:37004 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:39] INFO:     174.1.60.5:37012 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:40] INFO:     174.1.60.5:37020 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:42] INFO:     174.1.60.5:36076 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:52] INFO:     174.1.60.5:50774 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:54] INFO:     174.1.60.5:50786 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:56] INFO:     174.1.60.5:50788 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:58] INFO:     174.1.60.5:50790 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:59] INFO:     174.1.60.5:50806 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:23:00] INFO:     174.1.60.5:50818 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:23:01] INFO:     174.1.60.5:49800 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:23:03] INFO:     174.1.60.5:49808 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:23:20] INFO:     174.1.60.5:35588 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:23:20] INFO:     174.1.60.5:35590 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:32:54] INFO:     174.1.60.1:40486 - "GET /health HTTP/1.1" 200 OK
+[2026-08-27 05:35:46] INFO:     174.1.60.5:41898 - "POST /generate HTTP/1.1" 200 OK
+I0827 05:22:38.816133  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:38.914274  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:39.463928  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+[2026-08-27 05:36:14] INFO:     174.1.60.5:52884 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:36:14] INFO:     174.1.60.5:52888 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:35:55 TP0 EP0] Decode batch, #running-req: 1, #full token: 16896, full token usage: 0.04, mamba num: 1, mamba usage: 0.01, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 0.03, #queue-req: 0
+[2026-08-27 05:35:56 TP0 EP0] Decode batch, #running-req: 1, #full token: 16960, full token usage: 0.04, mamba num: 1, mamba usage: 0.01, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 25.04, #queue-req: 0
+[2026-08-27 05:35:58 TP0 EP0] Decode batch, #running-req: 1, #full token: 16960, full token usage: 0.04, mamba num: 1, mamba usage: 0.01, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 25.06, #queue-req: 0
+[2026-08-27 05:36:00 TP0 EP0] Decode batch, #running-req: 1, #full token: 17024, full token usage: 0.04, mamba num: 1, mamba usage: 0.01, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 25.08, #queue-req: 0
+[2026-08-27 05:36:01 TP0 EP0] Decode batch, #running-req: 1, #full token: 17088, full token usage: 0.04, mamba num: 1, mamba usage: 0.01, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 25.07, #queue-req: 0
+[2026-08-27 05:36:03 TP0 EP0] Decode batch, #running-req: 1, #full token: 17088, full token usage: 0.04, mamba num: 1, mamba usage: 0.01, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 25.83, #queue-req: 0
+[2026-08-27 05:36:04 TP0 EP0] Decode batch, #running-req: 1, #full token: 17152, full token usage: 0.04, mamba num: 1, mamba usage: 0.01, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 26.19, #queue-req: 0
+[2026-08-27 05:36:06 TP0 EP0] Decode batch, #running-req: 1, #full token: 17152, full token usage: 0.04, mamba num: 1, mamba usage: 0.01, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 26.20, #queue-req: 0
+[2026-08-27 05:36:07 TP0 EP0] Decode batch, #running-req: 1, #full token: 17216, full token usage: 0.04, mamba num: 1, mamba usage: 0.01, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 26.31, #queue-req: 0
+[2026-08-27 05:36:09 TP0 EP0] Decode batch, #running-req: 1, #full token: 17280, full token usage: 0.04, mamba num: 1, mamba usage: 0.01, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 26.34, #queue-req: 0
+[2026-08-27 05:36:10 TP0 EP0] Decode batch, #running-req: 1, #full token: 17280, full token usage: 0.04, mamba num: 1, mamba usage: 0.01, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 26.41, #queue-req: 0
+[2026-08-27 05:36:12 TP0 EP0] Decode batch, #running-req: 1, #full token: 17344, full token usage: 0.04, mamba num: 1, mamba usage: 0.01, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 26.42, #queue-req: 0
+[2026-08-27 05:36:13 TP0 EP0] Decode batch, #running-req: 1, #full token: 17408, full token usage: 0.04, mamba num: 1, mamba usage: 0.01, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 26.41, #queue-req: 0
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-20260827-1325/logs/manual_d_node1.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-20260827-1325/logs/manual_d_node1.log
new file mode 100644
index 0000000..4ae8fab
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-20260827-1325/logs/manual_d_node1.log
@@ -0,0 +1,1458 @@
+/usr/local/lib/python3.12/dist-packages/torchao/quantization/quant_api.py:1731: SyntaxWarning: invalid escape sequence '\.'
+  """Configuration class for applying different quantization configs to modules or parameters based on their fully qualified names (FQNs).
+/sgl-workspace/sglang/python/sglang/launch_server.py:56: UserWarning: 'python -m sglang.launch_server' is still supported, but 'sglang serve' is the recommended entrypoint.
+  Example: sglang serve --model-path <model> [options]
+  warnings.warn(
+[2026-08-27 04:35:58] KV cache is forced as chunk cache for decode server
+[2026-08-27 04:36:02] Breakable CUDA graph is incompatible with MLA attention; disabling prefill CUDA graph.
+[2026-08-27 04:36:02] Attention backend not specified. Use triton backend by default.
+[2026-08-27 04:36:05] server_args=ServerArgs(model_path='/data/hf_models/Kimi-K3', tokenizer_path='/data/hf_models/Kimi-K3', tokenizer_mode='auto', tokenizer_backend='huggingface', tokenizer_worker_num=1, detokenizer_worker_num=1, skip_tokenizer_init=False, load_format='auto', model_loader_extra_config='{}', trust_remote_code=True, context_length=None, is_embedding=False, enable_multimodal=None, revision=None, model_impl='auto', model_config_parser='auto', json_model_override_args='{}', dtype='auto', quantization=None, quantization_param_path=None, kv_cache_dtype='auto', enable_fp32_lm_head=False, modelopt_quant=None, modelopt_checkpoint_restore_path=None, modelopt_checkpoint_save_path=None, modelopt_export_path=None, quantize_and_serve=False, rl_quant_profile=None, enable_tf32_matmul=False, mem_fraction_static=0.88, max_running_requests=None, max_queued_requests=None, max_total_tokens=None, chunked_prefill_size=8192, enable_dynamic_chunking=False, max_prefill_tokens=16384, prefill_max_requests=None, schedule_policy='fcfs', enable_priority_scheduling=False, disable_priority_preemption=False, default_priority_value=None, abort_on_priority_when_disabled=False, schedule_low_priority_values_first=False, priority_scheduling_preemption_threshold=10, retraction_policy='length', schedule_conservativeness=1.0, page_size=64, swa_full_tokens_ratio=0.8, disable_hybrid_swa_memory=False, radix_eviction_policy='lru', prefill_only_disable_kv_cache=False, disable_radix_cache=True, enable_page_major_kv_layout=False, enable_unified_memory=False, disable_chunked_prefix_cache=False, disable_overlap_schedule=False, num_continuous_decode_steps=1, scheduler_recv_interval=1, enable_mixed_chunk=False, nccl_port=None, dist_timeout=3600, dist_init_addr='174.1.60.5:20000', nnodes=4, node_rank=1, tp_size=32, dcp_size=1, pp_size=1, pp_max_micro_batch_size=None, pp_async_batch_depth=0, dp_size=1, load_balance_method='round_robin', attn_cp_size=1, moe_dp_size=1, dcp_comm_backend='ag_rs', dcp_replicate_q_proj=None, dwdp_size=1, enable_prefill_cp=False, cp_strategy=None, enable_dsa_cache_layer_split=False, enable_dsa_prefill_context_parallel=False, dsa_prefill_cp_mode='round-robin-split', enable_prefill_context_parallel=False, prefill_cp_mode='in-seq-split', enable_cp_decode_attn_tp=False, enable_dp_attention=False, enable_dp_attention_local_control_broadcast=False, enable_dp_lm_head=False, enable_attn_tp_input_scattered=False, disable_attn_tp_gather=False, enable_p2p_check=False, device='cuda', base_gpu_id=0, gpu_id_step=1, random_seed=434112381, watchdog_timeout=300, soft_watchdog_timeout=None, sleep_on_idle=False, use_ray=False, custom_sigquit_handler=None, numa_node=None, gc_threshold=None, host='0.0.0.0', port=30000, fastapi_root_path='', smg_grpc_mode=False, grpc_mode=False, grpc_port=None, sidecar=None, sidecar_args=None, skip_server_warmup=False, warmups=None, enable_http2=False, ssl_keyfile=None, ssl_certfile=None, ssl_ca_certs=None, ssl_keyfile_password=None, enable_ssl_refresh=False, api_key=None, admin_api_key=None, served_model_name='kimi-k3', weight_version='default', chat_template=None, hf_chat_template_name=None, completion_template=None, file_storage_path='sglang_storage', enable_cache_report=False, reasoning_parser=None, default_chat_template_kwargs=None, strip_thinking_cache=False, enable_strict_thinking=False, tool_call_parser=None, tool_server=None, sampling_defaults='model', asr_max_buffer_seconds=60, asr_max_concurrent_sessions=32, preferred_sampling_params=None, allow_auto_truncate=False, stream_interval=1, batch_notify_size=16, stream_response_default_include_usage=False, incremental_streaming_output=False, enable_streaming_session=False, enable_session_radix_cache=False, log_level='info', log_level_http=None, log_requests=False, log_requests_level=2, log_requests_format='text', log_requests_target=None, uvicorn_access_log_exclude_prefixes=[], crash_dump_folder=None, show_time_cost=False, enable_metrics=False, smg_http_sidecar_port=None, enable_mfu_metrics=False, enable_metrics_for_all_schedulers=False, load_snapshot_publish_interval=15, tokenizer_metrics_custom_labels_header='x-custom-labels', tokenizer_metrics_allowed_custom_labels=None, extra_metric_labels=None, bucket_time_to_first_token=None, bucket_inter_token_latency=None, bucket_e2e_request_latency=None, prompt_tokens_buckets=None, generation_tokens_buckets=None, gc_warning_threshold_secs=0.0, decode_log_interval=40, enable_request_time_stats_logging=False, kv_events_config=None, enable_forward_pass_metrics=False, forward_pass_metrics_worker_id='', forward_pass_metrics_ipc_name=None, enable_trace=False, trace_modules='request', otlp_traces_endpoint='localhost:4317', export_metrics_to_file=False, export_metrics_to_file_dir=None, stat_loggers=None, constrained_json_whitespace_pattern=None, constrained_json_disable_any_whitespace=False, attention_backend='triton', decode_attention_backend=None, prefill_attention_backend=None, sampling_backend='flashinfer', grammar_backend='xgrammar', radix_cache_backend=None, mm_attention_backend=None, fp8_gemm_runner_backend='auto', fp4_gemm_runner_backend='auto', bf16_gemm_backend='auto', dsa_prefill_backend=None, dsa_decode_backend=None, dsa_paged_mqa_logits_backend='auto', dsa_topk_backend='sgl-kernel', disable_flashinfer_autotune=False, flashinfer_autotune_skip_ops=None, mamba_backend='triton', cuda_graph_config=CudaGraphConfig(decode=PhaseConfig(backend='full', max_bs=16, bs=[1, 2, 4, 8, 12, 16], tc_compiler='eager', full_prefill_max_req=None), prefill=PhaseConfig(backend='disabled', max_bs=2048, bs=[4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], tc_compiler='eager', full_prefill_max_req=None)), cuda_graph_backend_decode=None, cuda_graph_backend_prefill=None, cuda_graph_max_bs_decode=16, cuda_graph_max_bs_prefill=None, cuda_graph_bs_decode=None, cuda_graph_bs_prefill=None, cuda_graph_tc_compiler=None, disable_prefill_cuda_graph=False, disable_decode_cuda_graph=False, disable_cuda_graph=False, disable_cuda_graph_padding=False, enable_profile_cuda_graph=False, enable_cudagraph_gc=False, debug_cuda_graph=False, enable_layerwise_nvtx_marker=False, enable_nccl_nvls=False, enable_symm_mem=False, triton_attention_reduce_in_fp32=False, triton_attention_num_kv_splits=8, triton_attention_split_tile_size=None, flashinfer_mla_disable_ragged=False, enable_fused_qk_norm_rope=False, enable_precise_embedding_interpolation=False, enable_fused_moe_sum_all_reduce=False, enable_deepseek_v4_fp4_indexer=False, disable_custom_all_reduce=False, enable_mscclpp=False, enable_torch_symm_mem=False, enable_scattered_sconv=False, pre_warm_nccl=False, enable_quant_communications=False, enable_flashinfer_allreduce_fusion=False, enforce_disable_flashinfer_allreduce_fusion=False, flashinfer_allreduce_fusion_backend=None, enable_aiter_allreduce_fusion=False, enable_torch_compile=False, enable_torch_compile_debug_mode=False, torch_compile_max_bs=32, torchao_config='', speculative_algorithm=None, speculative_draft_model_path=None, speculative_draft_model_revision=None, speculative_draft_load_format=None, speculative_num_steps=None, speculative_eagle_topk=None, speculative_num_draft_tokens=None, speculative_dflash_block_size=None, speculative_dspark_block_size=None, speculative_dspark_sps_table_path=None, speculative_dspark_confidence_sts_path=None, speculative_dspark_align_verify_tokens_to_graph_tier=False, speculative_accept_threshold_single=1.0, speculative_accept_threshold_acc=1.0, speculative_use_rejection_sampling=False, speculative_token_map=None, speculative_attention_mode='prefill', speculative_draft_attention_backend=None, speculative_draft_window_size=None, speculative_moe_runner_backend='marlin', speculative_moe_a2a_backend=None, speculative_draft_model_quantization=None, speculative_skip_dp_mlp_sync=False, enable_multi_layer_eagle=False, speculative_adaptive=False, speculative_adaptive_config=None, decoupled_spec_bind_endpoint=None, decoupled_spec_connect_endpoints=None, decoupled_spec_rank=None, decoupled_spec_role='null', spec_trace_dir=None, speculative_ngram_min_bfs_breadth=1, speculative_ngram_max_bfs_breadth=10, speculative_ngram_match_type='BFS', speculative_ngram_max_trie_depth=18, speculative_ngram_capacity=10000000, speculative_ngram_external_corpus_path=None, speculative_ngram_external_sam_budget=0, speculative_ngram_external_corpus_max_tokens=10000000, ep_size=32, moe_a2a_backend='none', moe_runner_backend='marlin', flashinfer_mxfp4_moe_precision='default', deepep_mode='auto', fuseep_mode=2, deepep_dispatcher_output_dtype='auto', ep_num_redundant_experts=0, ep_dispatch_algorithm=None, init_expert_location='trivial', enable_eplb=False, eplb_algorithm='auto', eplb_rebalance_num_iterations=1000, eplb_rebalance_layers_per_chunk=None, eplb_min_rebalancing_utilization_threshold=1.0, expert_distribution_recorder_mode=None, expert_distribution_recorder_buffer_size=1000, enable_expert_distribution_metrics=False, deepep_config=None, moe_dense_tp_size=None, elastic_ep_backend=None, enable_elastic_expert_backup=False, mooncake_ib_device=None, enable_waterfill=False, ep_join_mode=None, ep_join_rank_offset=0, elastic_ep_initial_size=None, max_ep_size=None, elastic_ep_scale_timeout=600, elastic_ep_rejoin=False, disable_flashinfer_cutlass_moe_fp4_allgather=False, disable_shared_experts_fusion=False, enforce_shared_experts_fusion=False, max_mamba_cache_size=None, mamba_ssm_dtype=None, mamba_max_states_per_path=-1, enable_mamba_cache_stochastic_rounding=False, mamba_cache_philox_rounds=0, mamba_full_memory_ratio=0.21, mamba_radix_cache_strategy='extra_buffer_lazy', uses_mamba_radix_cache=False, mamba_track_interval=256, enable_int8_mamba_checkpoint=False, int8_mamba_ckpt_size=None, linear_attn_backend='triton', linear_attn_decode_backend=None, linear_attn_prefill_backend=None, linear_attn_verify_backend=None, enable_linear_replayssm=False, linear_replayssm_cache_len=16, enable_linear_replayssm_spec=False, enable_hierarchical_cache=False, hicache_ratio=2.0, hicache_size=0, hicache_write_policy='write_through', hicache_io_backend='kernel', hicache_mem_layout='page_first', hicache_storage_backend=None, hicache_storage_prefetch_policy='timeout', hicache_storage_backend_extra_config=None, enable_hisparse=False, hisparse_config=None, enable_broadcast_mm_inputs_process=False, enable_prefix_mm_cache=False, mm_enable_dp_encoder=False, mm_process_config={}, mm_processor_worker_num=0, mm_io_worker_num=0, limit_mm_data_per_request=None, enable_mm_global_cache=False, disable_fast_image_processor=False, mm_feature_transport='cpu', keep_mm_feature_on_device=False, enable_lora=None, enable_lora_overlap_loading=None, max_lora_rank=None, lora_target_modules=None, lora_paths=None, max_loaded_loras=None, max_loras_per_batch=8, lora_eviction_policy='lru', lora_backend='csgmv', max_lora_chunk_size=16, experts_shared_outer_loras=None, lora_use_virtual_experts=False, lora_strict_loading=False, lora_drain_wait_threshold=0.0, enable_two_batch_overlap=False, enable_single_batch_overlap=False, tbo_token_distribution_threshold=0.48, cpu_offload_gb=0, offload_group_size=-1, offload_num_in_group=1, offload_prefetch_step=1, offload_mode='cpu', enable_lmcache=False, lmcache_config_file=None, enable_flexkv=False, flexkv_config_file=None, kt_weight_path=None, kt_method='AMXINT4', kt_cpuinfer=None, kt_threadpool_count=2, kt_num_gpu_experts=None, kt_max_deferred_experts_per_token=None, dllm_algorithm=None, dllm_algorithm_config=None, dllm_fdfo=True, disaggregation_mode='decode', disaggregation_transfer_backend='mooncake', disaggregation_bootstrap_port=28800, disaggregation_ib_device='mlx5_0,mlx5_1,mlx5_2,mlx5_3', disaggregation_decode_enable_radix_cache=False, disaggregation_decode_enable_offload_kvcache=False, num_reserved_decode_tokens=512, disaggregation_decode_extra_slots=0, disaggregation_decode_polling_interval=1, optimistic_prefill_attempts=0, encoder_only=False, language_only=False, encoder_transfer_backend='zmq_to_tokenizer', encoder_urls=[], encoder_bootstrap_port=8997, encoder_register_urls=[], enable_adaptive_dispatch_to_encoder=False, enable_pdmux=False, pdmux_config_path=None, sm_group_num=8, custom_weight_loader=[], weight_loader_disable_mmap=False, weight_loader_prefetch_checkpoints=False, weight_loader_prefetch_num_threads=4, weight_loader_drop_cache_after_load=False, remote_instance_weight_loader_seed_instance_ip=None, remote_instance_weight_loader_seed_instance_service_port=None, remote_instance_weight_loader_send_weights_group_ports=None, remote_instance_weight_loader_backend='nccl', remote_instance_weight_loader_start_seed_via_transfer_engine=False, engine_info_bootstrap_port=6789, modelexpress_config=None, download_dir=None, model_checksum=None, delete_ckpt_after_loading=False, decrypted_config_file=None, decrypted_draft_config_file=None, checkpoint_engine_wait_weights_before_ready=False, enable_prefill_delayer=False, prefill_delayer_max_delay_passes=30, prefill_delayer_token_usage_low_watermark=None, prefill_delayer_forward_passes_buckets=None, prefill_delayer_wait_seconds_buckets=None, prefill_delayer_queue_min_ratio=None, prefill_delayer_max_delay_ms=None, min_free_slots_delay=None, enable_deterministic_inference=False, rl_on_policy_target=None, kv_canary='none', kv_canary_real_data='none', kv_canary_sweep_interval=0, enable_dynamic_batch_tokenizer=False, dynamic_batch_tokenizer_batch_size=32, dynamic_batch_tokenizer_batch_timeout=0.002, enable_tokenizer_batch_encode=False, disable_tokenizer_batch_decode=False, debug_tensor_dump_output_folder=None, debug_tensor_dump_layers=None, debug_tensor_dump_input_file=None, enable_memory_saver=False, enable_weights_cpu_backup=False, enable_draft_weights_cpu_backup=False, enable_custom_logit_processor=False, enable_return_hidden_states=False, enable_return_routed_experts=False, enable_return_indexer_topk=False, disable_outlines_disk_cache=False, enable_mis=False, weight_cache_mode='off', weight_cache_socket=None, weight_cache_timeout=1800, forward_hooks=None, msprobe_dump_config=None)
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:22 TP15 EP15] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:22 TP15 EP15] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:23 TP9 EP9] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:23 TP9 EP9] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:23 TP12 EP12] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:23 TP12 EP12] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:23 TP13 EP13] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:23 TP13 EP13] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:23 TP14 EP14] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:23 TP14 EP14] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:23 TP8 EP8] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:23 TP8 EP8] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:23 TP10 EP10] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:23 TP10 EP10] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:23 TP11 EP11] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:23 TP11 EP11] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:24 TP15 EP15] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:24 TP9 EP9] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:24 TP13 EP13] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:24 TP12 EP12] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:24 TP14 EP14] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:25 TP10 EP10] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:25 TP8 EP8] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:25 TP11 EP11] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:26.087226   242 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:26.087333   242 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:26.087350   242 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.6 port: 12001
+I0827 04:36:26.087409   242 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.6:16298
+I0827 04:36:26.087569   242 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:26.095995   236 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:26.096091   236 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:26.096107   236 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.6 port: 12001
+I0827 04:36:26.096150   236 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.6:15958
+I0827 04:36:26.096269   236 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:26.096457   240 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:26.096560   240 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:26.096577   240 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.6 port: 12001
+I0827 04:36:26.096626   240 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.6:16447
+I0827 04:36:26.096773   240 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:26.096876   241 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:26.096975   241 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:26.096992   241 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.6 port: 12001
+I0827 04:36:26.097035   241 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.6:16035
+I0827 04:36:26.097170   241 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:26.103180   241 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:26.107820   240 topology.cpp:127] Device mlx5_0 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:26.109326   239 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:26.109416   239 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:26.109428   239 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.6 port: 12001
+I0827 04:36:26.109464   239 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.6:16843
+I0827 04:36:26.109562   239 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:26.114297   242 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:26.114653   239 topology.cpp:127] Device mlx5_0 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:26.115540   237 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:26.115630   237 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:26.115645   237 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.6 port: 12001
+I0827 04:36:26.115686   237 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.6:15544
+I0827 04:36:26.115813   237 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:26.118220   240 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:26.121481   242 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:26.122418   237 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:26.126497   236 topology.cpp:127] Device mlx5_0 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:26.126574   235 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:26.126684   235 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:26.126703   235 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.6 port: 12001
+I0827 04:36:26.126758   235 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.6:15006
+I0827 04:36:26.126933   235 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:26.127116   240 topology.cpp:127] Device mlx5_2 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:26.127162   238 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:26.127270   238 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:26.127285   238 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.6 port: 12001
+I0827 04:36:26.127342   238 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.6:16140
+I0827 04:36:26.127489   238 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:26.128777   237 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:26.129559   242 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:26.130432   241 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:26.133821   238 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:26.134943   235 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:26.135799   240 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:26.136080   237 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:26.138710   241 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:26.139173   242 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:26.139978   240 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:26.140033   240 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:26.140090   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.140504   235 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:26.142936   242 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:26.142961   242 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:26.143014   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.144655   237 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:26.146834   235 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:26.147805   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:26.147867   241 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:26.148357   239 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:26.148370   237 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:26.148394   237 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:26.148444   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.149503   240 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:06
+I0827 04:36:26.149536   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.150290   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:26.150381   236 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:26.151773   242 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:06
+I0827 04:36:26.151803   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.152055   241 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:26.152087   241 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:26.152148   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.153594   235 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:26.156044   235 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:26.156075   235 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:26.156124   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.157910   238 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:26.159448   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:26.159547   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:26.160912   242 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:06
+I0827 04:36:26.160940   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.161144   241 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:06
+I0827 04:36:26.161176   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.162882   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:26.164348   235 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:06
+I0827 04:36:26.164378   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.165159   238 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:26.168772   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:26.170400   241 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:06
+I0827 04:36:26.170428   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.172459   238 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:26.174376   239 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:26.175243   238 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:26.175272   238 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:26.175325   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.178375   236 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:26.178763   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:26.178785   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:26.180411   237 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:06
+I0827 04:36:26.180462   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.180604   240 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:06
+I0827 04:36:26.180672   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.181643   239 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:26.182818   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:26.184404   238 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:06
+I0827 04:36:26.184437   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.184495   239 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:26.184537   239 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:26.184595   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.185601   236 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:26.186753   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:26.188311   242 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:06
+I0827 04:36:26.188344   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.188390   236 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:26.188416   236 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:26.188470   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.190752   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:26.192051   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:26.192406   235 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:06
+I0827 04:36:26.192444   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.193856   240 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:06
+I0827 04:36:26.193897   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.194197   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:26.195964   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:26.195997   238 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:06
+I0827 04:36:26.196039   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.196040   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:26.197574   242 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:06
+I0827 04:36:26.197599   236 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:06
+I0827 04:36:26.197629   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.197691   242 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:26 TP15 EP15] Init torch distributed begin.
+I0827 04:36:26.198544   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:26.200238   241 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:06
+I0827 04:36:26.200268   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.204000   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:26.205737   238 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:06
+I0827 04:36:26.205766   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.210784   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:26.212373   237 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:06
+I0827 04:36:26.212417   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.218751   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:26.220386   235 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:06
+I0827 04:36:26.220422   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.222854   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:26.223088   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:26.223217   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:26.223345   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:26.224500   241 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:06
+I0827 04:36:26.224536   240 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:06
+I0827 04:36:26.224612   241 transfer_engine_impl.cpp:388] installTransport, type=rdma
+I0827 04:36:26.224682   240 transfer_engine_impl.cpp:388] installTransport, type=rdma
+I0827 04:36:26.224973   236 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:06
+I0827 04:36:26.225018   236 rdma_context.cpp:205] Using SIEVE endpoint store
+[2026-08-27 04:36:26 TP14 EP14] Init torch distributed begin.
+[2026-08-27 04:36:26 TP13 EP13] Init torch distributed begin.
+I0827 04:36:26.225459   239 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:06
+I0827 04:36:26.225523   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.230858   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:26.232468   238 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:06
+I0827 04:36:26.232614   238 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:26 TP11 EP11] Init torch distributed begin.
+I0827 04:36:26.233065   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:26.234685   236 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:06
+I0827 04:36:26.234727   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.238961   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:26.240710   237 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:06
+I0827 04:36:26.240758   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.242511   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:26.244089   236 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:06
+I0827 04:36:26.244236   236 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:26 TP9 EP9] Init torch distributed begin.
+I0827 04:36:26.247059   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:26.248764   235 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:06
+I0827 04:36:26.248920   235 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:26 TP8 EP8] Init torch distributed begin.
+I0827 04:36:26.250787   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:26.252344   239 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:06
+I0827 04:36:26.252393   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.266752   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:26.268360   237 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:06
+I0827 04:36:26.268492   237 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:26 TP10 EP10] Init torch distributed begin.
+I0827 04:36:26.278649   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:26.280236   239 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:06
+I0827 04:36:26.280287   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.306850   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:26.308480   239 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:06
+I0827 04:36:26.308605   239 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:26 TP12 EP12] Init torch distributed begin.
+[2026-08-27 04:36:48 TP14 EP14] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP15 EP15] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP8 EP8] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP11 EP11] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP9 EP9] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP13 EP13] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP10 EP10] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP12 EP12] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP15 EP15] Init torch distributed ends. elapsed=22.03 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP14 EP14] Init torch distributed ends. elapsed=22.01 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP13 EP13] Init torch distributed ends. elapsed=22.01 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP12 EP12] Init torch distributed ends. elapsed=21.92 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP11 EP11] Init torch distributed ends. elapsed=22.00 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP9 EP9] Init torch distributed ends. elapsed=21.99 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP10 EP10] Init torch distributed ends. elapsed=21.96 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP8 EP8] Init torch distributed ends. elapsed=21.98 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP14 EP14] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP15 EP15] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP9 EP9] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP13 EP13] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP8 EP8] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP10 EP10] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP12 EP12] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP14 EP14] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP15 EP15] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP9 EP9] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP10 EP10] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP13 EP13] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP8 EP8] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP12 EP12] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP11 EP11] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP11 EP11] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP13 EP13] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP14 EP14] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP15 EP15] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP12 EP12] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP10 EP10] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP8 EP8] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP9 EP9] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP13 EP13] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP12 EP12] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP14 EP14] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP15 EP15] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP10 EP10] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP8 EP8] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP9 EP9] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP11 EP11] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP11 EP11] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:50 TP8 EP8] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP9 EP9] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP11 EP11] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP12 EP12] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP13 EP13] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP14 EP14] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP10 EP10] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP15 EP15] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:38:57 TP14 EP14] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:57 TP14 EP14] Load weight end. elapsed=129.27 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP9 EP9] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP9 EP9] Load weight end. elapsed=129.52 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP10 EP10] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP10 EP10] Load weight end. elapsed=129.56 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP8 EP8] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP8 EP8] Load weight end. elapsed=129.57 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP13 EP13] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP13 EP13] Load weight end. elapsed=129.58 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP11 EP11] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP11 EP11] Load weight end. elapsed=129.64 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP15 EP15] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP15 EP15] Load weight end. elapsed=129.77 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP12 EP12] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP12 EP12] Load weight end. elapsed=129.90 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:39:01 TP14 EP14] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP14 EP14] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP11 EP11] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP11 EP11] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP8 EP8] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP8 EP8] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP13 EP13] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP13 EP13] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP9 EP9] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP9 EP9] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP15 EP15] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP15 EP15] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP12 EP12] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP12 EP12] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP10 EP10] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP10 EP10] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:02 TP14 EP14] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP15 EP15] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP12 EP12] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP13 EP13] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP10 EP10] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP9 EP9] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP11 EP11] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP8 EP8] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP15 EP15] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP12 EP12] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP10 EP10] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP8 EP8] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP11 EP11] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP13 EP13] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP9 EP9] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP14 EP14] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP9 EP9] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP12 EP12] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP9 EP9] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP10 EP10] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP12 EP12] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP13 EP13] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP15 EP15] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP10 EP10] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP11 EP11] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP13 EP13] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP14 EP14] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP15 EP15] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP11 EP11] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP8 EP8] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP14 EP14] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP8 EP8] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP14 EP14] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP14 EP14] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP9 EP9] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP9 EP9] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP12 EP12] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP12 EP12] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:39:02 TP10 EP10] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP10 EP10] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP8 EP8] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP8 EP8] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:39:02 TP11 EP11] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP11 EP11] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:39:02 TP13 EP13] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP13 EP13] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP15 EP15] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP15 EP15] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:42:56 TP8 EP8] Capture target decode CUDA graph end. elapsed=233.78 s, mem usage=0.60 GB, avail mem=8.52 GB.
+[2026-08-27 04:42:56 TP13 EP13] Capture target decode CUDA graph end. elapsed=233.78 s, mem usage=0.58 GB, avail mem=8.58 GB.
+[2026-08-27 04:42:56 TP8 EP8] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP13 EP13] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP14 EP14] Capture target decode CUDA graph end. elapsed=233.81 s, mem usage=0.56 GB, avail mem=8.59 GB.
+[2026-08-27 04:42:56 TP10 EP10] Capture target decode CUDA graph end. elapsed=233.79 s, mem usage=0.56 GB, avail mem=8.59 GB.
+[2026-08-27 04:42:56 TP14 EP14] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP12 EP12] Capture target decode CUDA graph end. elapsed=233.80 s, mem usage=0.60 GB, avail mem=8.52 GB.
+[2026-08-27 04:42:56 TP10 EP10] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP12 EP12] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP11 EP11] Capture target decode CUDA graph end. elapsed=233.80 s, mem usage=0.62 GB, avail mem=8.51 GB.
+[2026-08-27 04:42:56 TP11 EP11] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP9 EP9] Capture target decode CUDA graph end. elapsed=233.91 s, mem usage=0.58 GB, avail mem=8.58 GB.
+[2026-08-27 04:42:56 TP9 EP9] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP15 EP15] Capture target decode CUDA graph end. elapsed=233.94 s, mem usage=0.62 GB, avail mem=8.51 GB.
+[2026-08-27 04:42:56 TP15 EP15] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+W0827 04:42:57.337543  2432 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5972aec0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.337602  2433 memory_location.cpp:73] Failed to get NUMA node, addr: 0x594a1440, len: 23168: Operation not permitted [1]
+W0827 04:42:57.337656  2434 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59f51680, len: 23168: Operation not permitted [1]
+W0827 04:42:57.337674  2435 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59ebed40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.337788  2436 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5b797200, len: 185344: Operation not permitted [1]
+W0827 04:42:57.337911  2437 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5b7c4640, len: 185344: Operation not permitted [1]
+W0827 04:42:57.337891  2438 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59639800, len: 23168: Operation not permitted [1]
+W0827 04:42:57.337962  2439 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59435700, len: 23168: Operation not permitted [1]
+W0827 04:42:57.338008  2440 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59660c80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.338057  2441 memory_location.cpp:73] Failed to get NUMA node, addr: 0x594c9400, len: 46336: Operation not permitted [1]
+W0827 04:42:57.338071  2442 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59bb1f00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.338107  2443 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59f5fc80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.338161  2445 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5953b700, len: 23168: Operation not permitted [1]
+W0827 04:42:57.338165  2444 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59812080, len: 23168: Operation not permitted [1]
+W0827 04:42:57.338204  2446 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5baddac0, len: 185344: Operation not permitted [1]
+W0827 04:42:57.338325  2448 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59994940, len: 23168: Operation not permitted [1]
+W0827 04:42:57.338340  2447 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5bb0af00, len: 185344: Operation not permitted [1]
+W0827 04:42:57.338538  2449 memory_location.cpp:73] Failed to get NUMA node, addr: 0x45ba2780, len: 46336: Operation not permitted [1]
+W0827 04:42:57.338596  2450 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59ac5f00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.338649  2451 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59984000, len: 23168: Operation not permitted [1]
+W0827 04:42:57.381100  2729 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59997640, len: 23168: Operation not permitted [1]
+W0827 04:42:57.381177  2728 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59a855c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.381280  2730 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59953e40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.381347  2732 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5b87eac0, len: 185344: Operation not permitted [1]
+W0827 04:42:57.381466  2731 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59971000, len: 23168: Operation not permitted [1]
+W0827 04:42:57.381542  2733 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5b8abf00, len: 185344: Operation not permitted [1]
+W0827 04:42:57.381620  2734 memory_location.cpp:73] Failed to get NUMA node, addr: 0x599b8540, len: 23168: Operation not permitted [1]
+W0827 04:42:57.381628  2735 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5927e3c0, len: 46336: Operation not permitted [1]
+W0827 04:42:57.381863  2736 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59900b40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.381878  2737 memory_location.cpp:73] Failed to get NUMA node, addr: 0x598b4f80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.421048  2876 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58bb5680, len: 23168: Operation not permitted [1]
+W0827 04:42:57.421196  2878 memory_location.cpp:73] Failed to get NUMA node, addr: 0x592f7a00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.421193  2877 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59321cc0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.421262  2881 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ab93340, len: 185344: Operation not permitted [1]
+W0827 04:42:57.421321  2879 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5929c2c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.421382  2880 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ab65f00, len: 185344: Operation not permitted [1]
+W0827 04:42:57.421423  2882 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58aa4dc0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.421440  2883 memory_location.cpp:73] Failed to get NUMA node, addr: 0x585179c0, len: 46336: Operation not permitted [1]
+W0827 04:42:57.421434  2884 memory_location.cpp:73] Failed to get NUMA node, addr: 0x588b0240, len: 23168: Operation not permitted [1]
+W0827 04:42:57.421594  2885 memory_location.cpp:73] Failed to get NUMA node, addr: 0x588d0800, len: 23168: Operation not permitted [1]
+W0827 04:42:57.423333  2900 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58bc9040, len: 23168: Operation not permitted [1]
+W0827 04:42:57.423418  2903 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58a57e40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.423974  2913 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ae07a80, len: 185344: Operation not permitted [1]
+W0827 04:42:57.424072  2911 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5adda640, len: 185344: Operation not permitted [1]
+W0827 04:42:57.424093  2908 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58f8c900, len: 23168: Operation not permitted [1]
+W0827 04:42:57.424189  2924 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58d795c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.424269  2906 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58e657c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.424296  2915 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58c46ac0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.424311  2922 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58e39680, len: 23168: Operation not permitted [1]
+W0827 04:42:57.424809  2930 memory_location.cpp:73] Failed to get NUMA node, addr: 0x591a7f40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.424866  2932 memory_location.cpp:73] Failed to get NUMA node, addr: 0x593dee00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.425096  2934 memory_location.cpp:73] Failed to get NUMA node, addr: 0x596af5c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.425186  2935 memory_location.cpp:73] Failed to get NUMA node, addr: 0x595b2ac0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.425271  2938 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5b4f5f80, len: 185344: Operation not permitted [1]
+W0827 04:42:57.425292  2937 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5b4c8b40, len: 185344: Operation not permitted [1]
+W0827 04:42:57.425374  2939 memory_location.cpp:73] Failed to get NUMA node, addr: 0x594e07c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.425433  2942 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58ec6dc0, len: 46336: Operation not permitted [1]
+W0827 04:42:57.425568  2944 memory_location.cpp:73] Failed to get NUMA node, addr: 0x596a8180, len: 23168: Operation not permitted [1]
+W0827 04:42:57.425750  2919 memory_location.cpp:73] Failed to get NUMA node, addr: 0x45850900, len: 46336: Operation not permitted [1]
+W0827 04:42:57.425791  2946 memory_location.cpp:73] Failed to get NUMA node, addr: 0x592c6f00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.427150  2998 memory_location.cpp:73] Failed to get NUMA node, addr: 0x599c15c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.427410  2989 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59750440, len: 23168: Operation not permitted [1]
+W0827 04:42:57.427572  3010 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5bad0cc0, len: 185344: Operation not permitted [1]
+W0827 04:42:57.427641  3014 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a39cd80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.427647  3024 memory_location.cpp:73] Failed to get NUMA node, addr: 0x464ff140, len: 23168: Operation not permitted [1]
+W0827 04:42:57.427655  3002 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a377380, len: 23168: Operation not permitted [1]
+W0827 04:42:57.427663  2993 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59876080, len: 23168: Operation not permitted [1]
+W0827 04:42:57.427681  3007 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5baa3880, len: 185344: Operation not permitted [1]
+W0827 04:42:57.427703  3025 memory_location.cpp:73] Failed to get NUMA node, addr: 0x463cd480, len: 23168: Operation not permitted [1]
+W0827 04:42:57.427706  3018 memory_location.cpp:73] Failed to get NUMA node, addr: 0x464e6b00, len: 46336: Operation not permitted [1]
+W0827 04:42:57.586287  3470 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a4fe500, len: 23168: Operation not permitted [1]
+W0827 04:42:57.586316  3471 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a8129c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.586287  3468 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a530a00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.586329  3469 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a72a280, len: 23168: Operation not permitted [1]
+W0827 04:42:57.586364  3472 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5c756140, len: 185344: Operation not permitted [1]
+W0827 04:42:57.586422  3474 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a287d00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.586429  3473 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5c783580, len: 185344: Operation not permitted [1]
+W0827 04:42:57.586477  3476 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a731080, len: 23168: Operation not permitted [1]
+W0827 04:42:57.586475  3475 memory_location.cpp:73] Failed to get NUMA node, addr: 0x46a69900, len: 46336: Operation not permitted [1]
+W0827 04:42:57.586767  3477 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a2cd580, len: 23168: Operation not permitted [1]
+[2026-08-27 04:42:59] Dummy health check server started in background thread at 0.0.0.0:30000
+I0827 04:48:13.995834  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:13.996141  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.010807  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.011080  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.031366  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.039373  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.042748  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.043025  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.057039  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.060467  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.074244  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.093904  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.097144  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.097398  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.100539  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.103674  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.106997  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.107286  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.107595  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.107802  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.108084  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.108319  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.111485  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.114651  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.114880  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.118037  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.127089  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.130350  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.136305  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.145159  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.145349  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.145601  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.148783  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.161775  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.162027  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.165246  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.165525  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.165712  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.171312  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.182595  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.182830  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.183091  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.183334  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.197396  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.202782  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.202994  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.211761  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.212010  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.212227  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.212466  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.218309  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.218585  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.218844  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.219072  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.219323  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.221719  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.223042  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.230329  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.230581  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.233455  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.235435  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.239619  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.239846  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.243142  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.248512  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.248768  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.249244  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.249539  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.249773  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.257058  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.259995  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.260232  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.260426  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.260643  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.263525  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.263789  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.264045  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.270218  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.272343  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.272601  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.272828  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.273056  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.273649  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.273857  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.274083  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.274338  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.274350  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.274650  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.274857  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.280273  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.281651  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.283277  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.288719  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.291597  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.291850  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.294591  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.297427  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.297646  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.297865  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.298131  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.298350  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.298571  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.300019  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.300335  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.301384  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.303872  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.304067  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.306974  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.307229  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.307489  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.308355  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.311895  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.312120  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.312359  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.312578  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.313377  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.317476  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.317679  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.319007  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.323084  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.323379  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.323665  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.333745  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.333979  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.334159  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.334322  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.337862  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.351508  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.351702  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.351871  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.352072  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.352252  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.355370  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.364216  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.367312  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.367552  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.367769  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.368913  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.375226  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.375468  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.375686  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.375947  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.376187  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.376428  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.376683  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.376904  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.377125  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.380695  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.382211  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.384042  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.384289  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.384522  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.390187  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.390600  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.391651  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.392228  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.398043  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.404222  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.404719  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.404901  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.405130  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.405673  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.405961  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.406175  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.406375  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.406561  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.406761  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.406991  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.407531  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.411696  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.411933  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.412912  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.416914  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.417143  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.417341  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.418166  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.418262  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.423719  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.425009  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.431804  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.432065  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.432313  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.432547  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.432744  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.432958  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.433151  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.433398  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.433662  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.433943  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.435259  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.439250  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.440708  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.443462  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.444864  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.446599  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.453094  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.457774  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.458002  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.458236  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.461966  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.463891  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.464155  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.465241  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.467888  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.468128  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.472088  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.473263  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.474021  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.474298  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.475373  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.479022  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.479259  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.479537  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.479795  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.480019  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.480264  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.480456  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.480696  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.480752  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.480891  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.481091  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.481326  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.481606  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.481833  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.482064  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.482175  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.482331  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.482553  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.482764  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.491822  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.493273  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.497345  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.498744  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.501601  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.502969  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.502988  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.509712  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.516039  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.516237  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.516439  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.519500  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.519733  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.519950  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.520223  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.520452  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.520711  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.520921  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.521140  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.521364  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.521629  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.521858  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.522101  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.522328  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.522553  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.522855  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.523010  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.523072  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.523293  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.523304  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.524191  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.524235  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.528468  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.535353  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.535600  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.535769  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.535825  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.536702  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.537087  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.542068  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.543424  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.545435  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.546203  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.546453  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.546715  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.546816  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.546933  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.547155  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.547389  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.547611  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.547857  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.548101  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.551676  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.553035  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.553618  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.554762  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.555172  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.556280  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.558871  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.565316  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.568676  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.569031  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.569290  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.569475  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.570151  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.572939  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.574455  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.575088  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.575302  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.576429  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.578527  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.578720  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.580932  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.582515  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.583039  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.583253  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.583478  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.583778  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.583832  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.584945  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.585202  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.586391  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.591900  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.593173  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.593271  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.593396  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.593664  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.596908  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.601369  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.602131  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.602725  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.604811  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.605091  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.606220  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.608003  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.610472  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.610689  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.610699  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.610924  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.611121  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.611315  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.611521  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.611792  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.612001  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.612202  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.612247  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.612334  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.612481  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.613652  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.615125  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.619858  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.620216  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.621151  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.624950  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.625170  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.625332  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.625414  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.625568  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.625614  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.625829  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.626011  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.626319  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.631613  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.632304  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.632570  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.632856  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.633095  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.633306  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.633574  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.634308  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.637017  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.637279  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.637583  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.638610  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.639593  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.640806  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.640900  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.641085  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.643601  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.644850  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.645069  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.646034  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.646308  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.646575  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.647399  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.648111  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.649897  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.651033  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.651146  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.653847  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.655079  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.659225  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.662467  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.663832  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.664858  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.665107  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.665309  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.666141  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.669166  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.673557  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.673764  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.673919  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.674009  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.674209  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.674503  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.674647  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.674710  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.674896  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.674955  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.675308  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.675336  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.675500  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.675552  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.675832  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.676005  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.676035  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.676223  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.676425  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.676607  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.680454  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.681128  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.685020  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.685290  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.686386  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.686959  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.694617  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.694845  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.695986  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.697674  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.698963  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.699177  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.699368  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.701349  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.701767  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.702576  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.702847  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.703204  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.706663  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.706753  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.706854  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.707039  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.707223  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.707465  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.707770  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.708069  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.708083  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.708248  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.708472  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.708722  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.708920  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.709123  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.710752  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.710954  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.712257  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.719622  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.719846  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.720033  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.720993  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.725260  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.725533  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.725705  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.725924  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.726168  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.726495  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.726595  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.727191  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.728030  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.728752  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.732066  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.734992  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.740563  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.746973  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.750195  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.750399  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.750788  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.752028  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.753922  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.756233  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.756840  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.759518  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.760934  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.766074  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.767369  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.769282  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.769475  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.769667  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.770648  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.774008  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.774199  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.774485  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.774672  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.774865  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.775053  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.775226  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.777894  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.778448  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.778666  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.779709  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.780627  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.781956  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.783814  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.785252  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.785427  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.785633  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.786656  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.787170  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.792593  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.797327  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.797559  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.797796  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.798789  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.801191  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.804433  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.804607  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.804807  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.805258  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.805953  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.806377  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.810664  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.811244  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.812598  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.824426  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.824638  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.824924  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.825793  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.831593  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.833058  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.833375  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.833634  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.833873  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.834122  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.834323  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.834601  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.834823  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.835043  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.835248  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.835427  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.835631  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.835824  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.836011  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.838608  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.838699  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.854580  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.858219  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.859412  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.862950  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.864025  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.865686  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.865891  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.866111  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.866333  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.866657  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.866854  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.867084  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.867278  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.867471  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.867661  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.872786  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.876446  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.876659  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.876896  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.877848  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.880227  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.880436  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.883129  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.883190  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.883363  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.883567  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.884621  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.884654  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.888731  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.926294  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.926565  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.926801  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.926995  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.927213  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.927462  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.927661  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.927874  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.937942  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.938158  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.938393  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.939355  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.949986  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.950210  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.950419  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.951480  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.985823  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.986092  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.986321  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.986529  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.986739  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.986924  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.987135  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.987387  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.987639  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.987892  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.988083  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.988299  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.988473  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.988667  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.012990  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.013230  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.013480  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.013671  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.013873  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.014089  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.014310  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.014530  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.014705  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.014922  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.029297  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.029497  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.029675  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.029886  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.030081  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.030721  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.075974  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.076195  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.076392  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.076560  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.076789  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.076989  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.077189  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.077405  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.086928  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.087107  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.087375  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.088318  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.098666  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.098937  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.099119  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.100116  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.134526  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.134721  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.134938  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.135179  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.135361  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.135597  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.135814  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.136029  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.136237  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.136453  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.136703  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.136907  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.137085  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.139734  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.164215  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.166940  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.167219  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.167455  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.167656  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.167850  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.168042  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.168241  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.168471  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.168686  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.168931  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.169119  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.183355  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.183578  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.183832  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.184046  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.184273  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.184649  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.231909  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.232179  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.232437  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.232667  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.232929  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.233191  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.233395  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.236016  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.241250  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.243832  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.244118  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.246883  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.252156  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.254787  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.257608  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.265419  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.273245  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.283512  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.283768  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.286342  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.291661  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.291878  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.292133  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.292326  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.292569  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.292784  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.293030  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.298199  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.298434  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.298682  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.311589  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.311841  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.319455  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.319670  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.319875  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.320118  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.320379  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.323133  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.323354  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.323570  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.331285  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.336499  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.336781  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.336977  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.337196  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.337783  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.375766  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.375975  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.376159  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.376356  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.376547  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.376745  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.376950  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.379487  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.384498  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.387058  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.387269  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.389864  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.394851  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.397480  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.397692  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.407253  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.424064  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.424232  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.424441  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.424628  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.432085  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.432324  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.432541  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.432767  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.432996  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.433199  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.433389  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.438321  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.438519  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.438731  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.450695  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.450888  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.458081  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.458277  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.458539  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.458791  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.458984  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.461611  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.461808  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.462006  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.469441  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.474292  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.474494  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.474697  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.474973  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.475522  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.457798  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.458177  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.459460  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.463181  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.464798  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.469339  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.469461  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.470870  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.475363  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.479321  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.480788  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.481803  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.483294  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.485772  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.494769  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.496369  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.500792  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.299279  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.299625  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.301414  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.309536  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.311267  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.312475  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.314007  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.319712  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.324035  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.325511  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.330739  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.330972  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.331518  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.332414  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.337031  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.339011  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.339269  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.340338  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.342085  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.343655  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.344139  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.345620  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.348776  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.349041  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.349591  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.350200  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.360395  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.360633  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.361879  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.362035  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.362210  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.363549  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.366465  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.370512  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.371948  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.373085  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.373347  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.374653  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.376161  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.378137  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.378718  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.378968  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.379657  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.384268  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.385306  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.385583  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.386750  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.386739  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.386987  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.388157  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.394881  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.398934  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.400285  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.402129  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.402400  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.403517  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.406072  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.406335  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.406554  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.406766  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.407377  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.416814  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.417057  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.418246  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.426201  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.426426  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.426638  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.426827  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.427093  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.427573  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.430253  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.430498  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.431689  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.391251  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.393167  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.399513  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.593394  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.593744  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.595059  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.599748  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.607136  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.608918  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.613543  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:40.735651  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:40.737267  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:40.743311  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:40.749338  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:40.750874  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:48.322273  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:48.323827  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:48.330022  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:55.890759  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:55.891063  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:55.892275  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:55.900822  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:55.902331  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:55.907302  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:55.915869  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:55.917403  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.662242  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.663878  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.676903  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.677158  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.677393  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.678658  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.690680  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.691715  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.692193  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.693166  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.699034  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.706213  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.707717  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.708302  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.710057  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.710531  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.725970  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.726244  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.727491  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.460754  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.462435  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.468483  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.477201  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.477489  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.478799  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.514531  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.514875  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.516037  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.170460  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.172108  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.178090  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.187213  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.187461  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.188678  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.201197  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.201504  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.202839  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.215639  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.215888  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.217134  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:47.594174  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:47.607700  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:48.578758  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:48.588865  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:48.597946  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:49.367239  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:49.380311  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:50.292033  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:50.302100  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:50.303560  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:50.312023  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:50.320772  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.137686  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.152026  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.162048  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.026280  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.032133  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.038511  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.041432  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.047593  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.050129  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.058367  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.826843  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.838438  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.847219  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:53.739216  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:54.582928  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:55.353269  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:55.481009  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:56.209721  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:57.074601  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:57.089560  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:57.620945  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:57.639796  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:57.643188  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:57.764745  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:05.707906  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:05.720738  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:06.742728  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:06.752833  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:07.478389  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:08.273787  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:09.240689  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:09.253865  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:10.164781  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:10.170979  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:11.012079  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:11.870416  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:18.023379  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:18.250713  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:18.926174  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:30.803402  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:30.804934  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:32.463899  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:52.158495  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:56.508555  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:56.518028  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:58.313472  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-20260827-1325/logs/manual_d_node2.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-20260827-1325/logs/manual_d_node2.log
new file mode 100644
index 0000000..cf2064d
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-20260827-1325/logs/manual_d_node2.log
@@ -0,0 +1,1470 @@
+/usr/local/lib/python3.12/dist-packages/torchao/quantization/quant_api.py:1731: SyntaxWarning: invalid escape sequence '\.'
+  """Configuration class for applying different quantization configs to modules or parameters based on their fully qualified names (FQNs).
+/sgl-workspace/sglang/python/sglang/launch_server.py:56: UserWarning: 'python -m sglang.launch_server' is still supported, but 'sglang serve' is the recommended entrypoint.
+  Example: sglang serve --model-path <model> [options]
+  warnings.warn(
+[2026-08-27 04:35:59] KV cache is forced as chunk cache for decode server
+[2026-08-27 04:36:03] Breakable CUDA graph is incompatible with MLA attention; disabling prefill CUDA graph.
+[2026-08-27 04:36:03] Attention backend not specified. Use triton backend by default.
+[2026-08-27 04:36:06] server_args=ServerArgs(model_path='/data/hf_models/Kimi-K3', tokenizer_path='/data/hf_models/Kimi-K3', tokenizer_mode='auto', tokenizer_backend='huggingface', tokenizer_worker_num=1, detokenizer_worker_num=1, skip_tokenizer_init=False, load_format='auto', model_loader_extra_config='{}', trust_remote_code=True, context_length=None, is_embedding=False, enable_multimodal=None, revision=None, model_impl='auto', model_config_parser='auto', json_model_override_args='{}', dtype='auto', quantization=None, quantization_param_path=None, kv_cache_dtype='auto', enable_fp32_lm_head=False, modelopt_quant=None, modelopt_checkpoint_restore_path=None, modelopt_checkpoint_save_path=None, modelopt_export_path=None, quantize_and_serve=False, rl_quant_profile=None, enable_tf32_matmul=False, mem_fraction_static=0.88, max_running_requests=None, max_queued_requests=None, max_total_tokens=None, chunked_prefill_size=8192, enable_dynamic_chunking=False, max_prefill_tokens=16384, prefill_max_requests=None, schedule_policy='fcfs', enable_priority_scheduling=False, disable_priority_preemption=False, default_priority_value=None, abort_on_priority_when_disabled=False, schedule_low_priority_values_first=False, priority_scheduling_preemption_threshold=10, retraction_policy='length', schedule_conservativeness=1.0, page_size=64, swa_full_tokens_ratio=0.8, disable_hybrid_swa_memory=False, radix_eviction_policy='lru', prefill_only_disable_kv_cache=False, disable_radix_cache=True, enable_page_major_kv_layout=False, enable_unified_memory=False, disable_chunked_prefix_cache=False, disable_overlap_schedule=False, num_continuous_decode_steps=1, scheduler_recv_interval=1, enable_mixed_chunk=False, nccl_port=None, dist_timeout=3600, dist_init_addr='174.1.60.5:20000', nnodes=4, node_rank=2, tp_size=32, dcp_size=1, pp_size=1, pp_max_micro_batch_size=None, pp_async_batch_depth=0, dp_size=1, load_balance_method='round_robin', attn_cp_size=1, moe_dp_size=1, dcp_comm_backend='ag_rs', dcp_replicate_q_proj=None, dwdp_size=1, enable_prefill_cp=False, cp_strategy=None, enable_dsa_cache_layer_split=False, enable_dsa_prefill_context_parallel=False, dsa_prefill_cp_mode='round-robin-split', enable_prefill_context_parallel=False, prefill_cp_mode='in-seq-split', enable_cp_decode_attn_tp=False, enable_dp_attention=False, enable_dp_attention_local_control_broadcast=False, enable_dp_lm_head=False, enable_attn_tp_input_scattered=False, disable_attn_tp_gather=False, enable_p2p_check=False, device='cuda', base_gpu_id=0, gpu_id_step=1, random_seed=212084647, watchdog_timeout=300, soft_watchdog_timeout=None, sleep_on_idle=False, use_ray=False, custom_sigquit_handler=None, numa_node=None, gc_threshold=None, host='0.0.0.0', port=30000, fastapi_root_path='', smg_grpc_mode=False, grpc_mode=False, grpc_port=None, sidecar=None, sidecar_args=None, skip_server_warmup=False, warmups=None, enable_http2=False, ssl_keyfile=None, ssl_certfile=None, ssl_ca_certs=None, ssl_keyfile_password=None, enable_ssl_refresh=False, api_key=None, admin_api_key=None, served_model_name='kimi-k3', weight_version='default', chat_template=None, hf_chat_template_name=None, completion_template=None, file_storage_path='sglang_storage', enable_cache_report=False, reasoning_parser=None, default_chat_template_kwargs=None, strip_thinking_cache=False, enable_strict_thinking=False, tool_call_parser=None, tool_server=None, sampling_defaults='model', asr_max_buffer_seconds=60, asr_max_concurrent_sessions=32, preferred_sampling_params=None, allow_auto_truncate=False, stream_interval=1, batch_notify_size=16, stream_response_default_include_usage=False, incremental_streaming_output=False, enable_streaming_session=False, enable_session_radix_cache=False, log_level='info', log_level_http=None, log_requests=False, log_requests_level=2, log_requests_format='text', log_requests_target=None, uvicorn_access_log_exclude_prefixes=[], crash_dump_folder=None, show_time_cost=False, enable_metrics=False, smg_http_sidecar_port=None, enable_mfu_metrics=False, enable_metrics_for_all_schedulers=False, load_snapshot_publish_interval=15, tokenizer_metrics_custom_labels_header='x-custom-labels', tokenizer_metrics_allowed_custom_labels=None, extra_metric_labels=None, bucket_time_to_first_token=None, bucket_inter_token_latency=None, bucket_e2e_request_latency=None, prompt_tokens_buckets=None, generation_tokens_buckets=None, gc_warning_threshold_secs=0.0, decode_log_interval=40, enable_request_time_stats_logging=False, kv_events_config=None, enable_forward_pass_metrics=False, forward_pass_metrics_worker_id='', forward_pass_metrics_ipc_name=None, enable_trace=False, trace_modules='request', otlp_traces_endpoint='localhost:4317', export_metrics_to_file=False, export_metrics_to_file_dir=None, stat_loggers=None, constrained_json_whitespace_pattern=None, constrained_json_disable_any_whitespace=False, attention_backend='triton', decode_attention_backend=None, prefill_attention_backend=None, sampling_backend='flashinfer', grammar_backend='xgrammar', radix_cache_backend=None, mm_attention_backend=None, fp8_gemm_runner_backend='auto', fp4_gemm_runner_backend='auto', bf16_gemm_backend='auto', dsa_prefill_backend=None, dsa_decode_backend=None, dsa_paged_mqa_logits_backend='auto', dsa_topk_backend='sgl-kernel', disable_flashinfer_autotune=False, flashinfer_autotune_skip_ops=None, mamba_backend='triton', cuda_graph_config=CudaGraphConfig(decode=PhaseConfig(backend='full', max_bs=16, bs=[1, 2, 4, 8, 12, 16], tc_compiler='eager', full_prefill_max_req=None), prefill=PhaseConfig(backend='disabled', max_bs=2048, bs=[4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], tc_compiler='eager', full_prefill_max_req=None)), cuda_graph_backend_decode=None, cuda_graph_backend_prefill=None, cuda_graph_max_bs_decode=16, cuda_graph_max_bs_prefill=None, cuda_graph_bs_decode=None, cuda_graph_bs_prefill=None, cuda_graph_tc_compiler=None, disable_prefill_cuda_graph=False, disable_decode_cuda_graph=False, disable_cuda_graph=False, disable_cuda_graph_padding=False, enable_profile_cuda_graph=False, enable_cudagraph_gc=False, debug_cuda_graph=False, enable_layerwise_nvtx_marker=False, enable_nccl_nvls=False, enable_symm_mem=False, triton_attention_reduce_in_fp32=False, triton_attention_num_kv_splits=8, triton_attention_split_tile_size=None, flashinfer_mla_disable_ragged=False, enable_fused_qk_norm_rope=False, enable_precise_embedding_interpolation=False, enable_fused_moe_sum_all_reduce=False, enable_deepseek_v4_fp4_indexer=False, disable_custom_all_reduce=False, enable_mscclpp=False, enable_torch_symm_mem=False, enable_scattered_sconv=False, pre_warm_nccl=False, enable_quant_communications=False, enable_flashinfer_allreduce_fusion=False, enforce_disable_flashinfer_allreduce_fusion=False, flashinfer_allreduce_fusion_backend=None, enable_aiter_allreduce_fusion=False, enable_torch_compile=False, enable_torch_compile_debug_mode=False, torch_compile_max_bs=32, torchao_config='', speculative_algorithm=None, speculative_draft_model_path=None, speculative_draft_model_revision=None, speculative_draft_load_format=None, speculative_num_steps=None, speculative_eagle_topk=None, speculative_num_draft_tokens=None, speculative_dflash_block_size=None, speculative_dspark_block_size=None, speculative_dspark_sps_table_path=None, speculative_dspark_confidence_sts_path=None, speculative_dspark_align_verify_tokens_to_graph_tier=False, speculative_accept_threshold_single=1.0, speculative_accept_threshold_acc=1.0, speculative_use_rejection_sampling=False, speculative_token_map=None, speculative_attention_mode='prefill', speculative_draft_attention_backend=None, speculative_draft_window_size=None, speculative_moe_runner_backend='marlin', speculative_moe_a2a_backend=None, speculative_draft_model_quantization=None, speculative_skip_dp_mlp_sync=False, enable_multi_layer_eagle=False, speculative_adaptive=False, speculative_adaptive_config=None, decoupled_spec_bind_endpoint=None, decoupled_spec_connect_endpoints=None, decoupled_spec_rank=None, decoupled_spec_role='null', spec_trace_dir=None, speculative_ngram_min_bfs_breadth=1, speculative_ngram_max_bfs_breadth=10, speculative_ngram_match_type='BFS', speculative_ngram_max_trie_depth=18, speculative_ngram_capacity=10000000, speculative_ngram_external_corpus_path=None, speculative_ngram_external_sam_budget=0, speculative_ngram_external_corpus_max_tokens=10000000, ep_size=32, moe_a2a_backend='none', moe_runner_backend='marlin', flashinfer_mxfp4_moe_precision='default', deepep_mode='auto', fuseep_mode=2, deepep_dispatcher_output_dtype='auto', ep_num_redundant_experts=0, ep_dispatch_algorithm=None, init_expert_location='trivial', enable_eplb=False, eplb_algorithm='auto', eplb_rebalance_num_iterations=1000, eplb_rebalance_layers_per_chunk=None, eplb_min_rebalancing_utilization_threshold=1.0, expert_distribution_recorder_mode=None, expert_distribution_recorder_buffer_size=1000, enable_expert_distribution_metrics=False, deepep_config=None, moe_dense_tp_size=None, elastic_ep_backend=None, enable_elastic_expert_backup=False, mooncake_ib_device=None, enable_waterfill=False, ep_join_mode=None, ep_join_rank_offset=0, elastic_ep_initial_size=None, max_ep_size=None, elastic_ep_scale_timeout=600, elastic_ep_rejoin=False, disable_flashinfer_cutlass_moe_fp4_allgather=False, disable_shared_experts_fusion=False, enforce_shared_experts_fusion=False, max_mamba_cache_size=None, mamba_ssm_dtype=None, mamba_max_states_per_path=-1, enable_mamba_cache_stochastic_rounding=False, mamba_cache_philox_rounds=0, mamba_full_memory_ratio=0.21, mamba_radix_cache_strategy='extra_buffer_lazy', uses_mamba_radix_cache=False, mamba_track_interval=256, enable_int8_mamba_checkpoint=False, int8_mamba_ckpt_size=None, linear_attn_backend='triton', linear_attn_decode_backend=None, linear_attn_prefill_backend=None, linear_attn_verify_backend=None, enable_linear_replayssm=False, linear_replayssm_cache_len=16, enable_linear_replayssm_spec=False, enable_hierarchical_cache=False, hicache_ratio=2.0, hicache_size=0, hicache_write_policy='write_through', hicache_io_backend='kernel', hicache_mem_layout='page_first', hicache_storage_backend=None, hicache_storage_prefetch_policy='timeout', hicache_storage_backend_extra_config=None, enable_hisparse=False, hisparse_config=None, enable_broadcast_mm_inputs_process=False, enable_prefix_mm_cache=False, mm_enable_dp_encoder=False, mm_process_config={}, mm_processor_worker_num=0, mm_io_worker_num=0, limit_mm_data_per_request=None, enable_mm_global_cache=False, disable_fast_image_processor=False, mm_feature_transport='cpu', keep_mm_feature_on_device=False, enable_lora=None, enable_lora_overlap_loading=None, max_lora_rank=None, lora_target_modules=None, lora_paths=None, max_loaded_loras=None, max_loras_per_batch=8, lora_eviction_policy='lru', lora_backend='csgmv', max_lora_chunk_size=16, experts_shared_outer_loras=None, lora_use_virtual_experts=False, lora_strict_loading=False, lora_drain_wait_threshold=0.0, enable_two_batch_overlap=False, enable_single_batch_overlap=False, tbo_token_distribution_threshold=0.48, cpu_offload_gb=0, offload_group_size=-1, offload_num_in_group=1, offload_prefetch_step=1, offload_mode='cpu', enable_lmcache=False, lmcache_config_file=None, enable_flexkv=False, flexkv_config_file=None, kt_weight_path=None, kt_method='AMXINT4', kt_cpuinfer=None, kt_threadpool_count=2, kt_num_gpu_experts=None, kt_max_deferred_experts_per_token=None, dllm_algorithm=None, dllm_algorithm_config=None, dllm_fdfo=True, disaggregation_mode='decode', disaggregation_transfer_backend='mooncake', disaggregation_bootstrap_port=28800, disaggregation_ib_device='mlx5_0,mlx5_1,mlx5_2,mlx5_3', disaggregation_decode_enable_radix_cache=False, disaggregation_decode_enable_offload_kvcache=False, num_reserved_decode_tokens=512, disaggregation_decode_extra_slots=0, disaggregation_decode_polling_interval=1, optimistic_prefill_attempts=0, encoder_only=False, language_only=False, encoder_transfer_backend='zmq_to_tokenizer', encoder_urls=[], encoder_bootstrap_port=8997, encoder_register_urls=[], enable_adaptive_dispatch_to_encoder=False, enable_pdmux=False, pdmux_config_path=None, sm_group_num=8, custom_weight_loader=[], weight_loader_disable_mmap=False, weight_loader_prefetch_checkpoints=False, weight_loader_prefetch_num_threads=4, weight_loader_drop_cache_after_load=False, remote_instance_weight_loader_seed_instance_ip=None, remote_instance_weight_loader_seed_instance_service_port=None, remote_instance_weight_loader_send_weights_group_ports=None, remote_instance_weight_loader_backend='nccl', remote_instance_weight_loader_start_seed_via_transfer_engine=False, engine_info_bootstrap_port=6789, modelexpress_config=None, download_dir=None, model_checksum=None, delete_ckpt_after_loading=False, decrypted_config_file=None, decrypted_draft_config_file=None, checkpoint_engine_wait_weights_before_ready=False, enable_prefill_delayer=False, prefill_delayer_max_delay_passes=30, prefill_delayer_token_usage_low_watermark=None, prefill_delayer_forward_passes_buckets=None, prefill_delayer_wait_seconds_buckets=None, prefill_delayer_queue_min_ratio=None, prefill_delayer_max_delay_ms=None, min_free_slots_delay=None, enable_deterministic_inference=False, rl_on_policy_target=None, kv_canary='none', kv_canary_real_data='none', kv_canary_sweep_interval=0, enable_dynamic_batch_tokenizer=False, dynamic_batch_tokenizer_batch_size=32, dynamic_batch_tokenizer_batch_timeout=0.002, enable_tokenizer_batch_encode=False, disable_tokenizer_batch_decode=False, debug_tensor_dump_output_folder=None, debug_tensor_dump_layers=None, debug_tensor_dump_input_file=None, enable_memory_saver=False, enable_weights_cpu_backup=False, enable_draft_weights_cpu_backup=False, enable_custom_logit_processor=False, enable_return_hidden_states=False, enable_return_routed_experts=False, enable_return_indexer_topk=False, disable_outlines_disk_cache=False, enable_mis=False, weight_cache_mode='off', weight_cache_socket=None, weight_cache_timeout=1800, forward_hooks=None, msprobe_dump_config=None)
+[2026-08-27 04:36:06] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:06] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:06] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:06] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:06] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:06] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:06] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:06] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:24 TP18 EP18] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:24 TP18 EP18] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:24 TP21 EP21] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:24 TP21 EP21] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:24 TP20 EP20] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:24 TP20 EP20] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:25 TP19 EP19] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:25 TP19 EP19] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:25 TP22 EP22] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:25 TP22 EP22] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:25 TP23 EP23] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:25 TP23 EP23] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:26 TP17 EP17] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:26 TP17 EP17] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:26 TP16 EP16] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:26 TP16 EP16] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:27 TP19 EP19] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:27 TP21 EP21] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:27 TP18 EP18] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:27 TP20 EP20] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:27 TP23 EP23] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:27 TP22 EP22] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:27 TP17 EP17] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:27 TP16 EP16] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:29.187384   239 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:29.187532   239 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:29.187558   239 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.7 port: 12001
+I0827 04:36:29.187624   239 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.7:16248
+I0827 04:36:29.187814   239 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:29.191661   235 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:29.191802   235 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:29.191823   235 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.7 port: 12001
+I0827 04:36:29.191901   235 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.7:16408
+I0827 04:36:29.192099   235 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:29.195215   242 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:29.195319   242 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:29.195333   242 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.7 port: 12001
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:29.195333   238 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:29.195374   242 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.7:15148
+I0827 04:36:29.195490   238 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:29.195513   238 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.7 port: 12001
+I0827 04:36:29.195519   242 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:29.195581   238 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.7:16909
+I0827 04:36:29.195815   238 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:29.198207   236 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:29.198346   236 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:29.198370   236 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.7 port: 12001
+I0827 04:36:29.198427   236 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.7:16738
+I0827 04:36:29.198611   236 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:29.200070   242 topology.cpp:127] Device mlx5_0 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:29.201086   237 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:29.201221   237 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:29.201239   237 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.7 port: 12001
+I0827 04:36:29.201300   237 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.7:16522
+I0827 04:36:29.201495   237 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:29.201511   240 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:29.201632   240 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:29.201648   240 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.7 port: 12001
+I0827 04:36:29.201701   240 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.7:16431
+I0827 04:36:29.201871   240 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:29.204236   241 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:29.204355   242 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:29.204380   241 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:29.204393   241 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.7 port: 12001
+I0827 04:36:29.204454   241 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.7:16508
+I0827 04:36:29.204633   241 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:29.204896   238 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:29.207675   237 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:29.208006   240 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:29.208549   242 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:29.212560   242 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:29.214218   242 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:29.214243   242 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:29.214290   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.214713   240 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:29.214757   237 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:29.219102   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:29.219870   239 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:29.220057   235 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:29.220343   242 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:07
+I0827 04:36:29.220369   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.222033   240 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:29.222069   237 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:29.223913   236 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:29.228577   235 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:29.228634   239 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:29.230391   237 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:29.230446   240 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:29.232455   236 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:29.233546   237 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:29.233578   237 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:29.233631   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.234635   240 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:29.234665   240 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:29.234716   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.235774   238 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:29.235827   241 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:29.237140   235 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:29.241117   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:29.241130   236 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:29.242069   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:29.242725   237 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:07
+I0827 04:36:29.242758   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.243661   240 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:07
+I0827 04:36:29.243691   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.245620   235 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:29.247138   238 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:29.247233   241 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:29.248172   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:29.248514   235 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:29.248550   235 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:29.248612   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.248719   236 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:29.249847   242 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:07
+I0827 04:36:29.249886   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.251354   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:29.251672   236 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:29.251719   236 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:29.251781   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.252889   240 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:07
+I0827 04:36:29.252921   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.255584   239 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:29.256263   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:29.257805   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:29.257980   235 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:07
+I0827 04:36:29.258016   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.258512   238 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:29.258567   241 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:29.259447   242 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:07
+I0827 04:36:29.259477   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.259642   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:29.261375   236 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:07
+I0827 04:36:29.261410   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.261518   238 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:29.261557   238 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:29.261615   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.261615   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:29.262753   239 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:29.263202   240 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:07
+I0827 04:36:29.263229   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.265475   239 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:29.265504   239 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:29.265558   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.269449   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:29.271157   236 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:07
+I0827 04:36:29.271186   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.271761   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:29.272920   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:29.273422   237 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:07
+I0827 04:36:29.273456   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.274566   239 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:07
+I0827 04:36:29.274600   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.281176   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:29.282341   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:29.282708   237 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:07
+I0827 04:36:29.282739   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.284137   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:29.284178   239 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:07
+I0827 04:36:29.284219   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.285904   235 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:07
+I0827 04:36:29.285953   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.288185   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:29.289810   242 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:07
+I0827 04:36:29.289940   242 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:29 TP23 EP23] Init torch distributed begin.
+I0827 04:36:29.291602   241 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:29.291882   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:29.291936   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:29.293576   240 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:07
+I0827 04:36:29.293699   240 transfer_engine_impl.cpp:388] installTransport, type=rdma
+I0827 04:36:29.293921   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+[2026-08-27 04:36:29 TP21 EP21] Init torch distributed begin.
+I0827 04:36:29.294121   238 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:07
+I0827 04:36:29.294169   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.294241   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:29.295475   239 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:07
+I0827 04:36:29.295507   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.295553   241 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:29.295591   241 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:29.295648   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.296002   235 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:07
+I0827 04:36:29.296041   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.296232   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:29.298113   236 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:07
+I0827 04:36:29.298154   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.303915   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:29.305596   235 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:07
+I0827 04:36:29.305743   235 transfer_engine_impl.cpp:388] installTransport, type=rdma
+I0827 04:36:29.306217   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+[2026-08-27 04:36:29 TP16 EP16] Init torch distributed begin.
+I0827 04:36:29.307994   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:29.308105   236 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:07
+I0827 04:36:29.308277   236 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:29 TP17 EP17] Init torch distributed begin.
+I0827 04:36:29.310081   237 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:07
+I0827 04:36:29.310261   237 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:29 TP18 EP18] Init torch distributed begin.
+I0827 04:36:29.320421   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:29.321812   239 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:07
+I0827 04:36:29.321919   239 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:29 TP20 EP20] Init torch distributed begin.
+I0827 04:36:29.323829   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:29.323894   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:29.325821   238 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:07
+I0827 04:36:29.325820   241 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:07
+I0827 04:36:29.325871   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.325872   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.360142   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:29.360253   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:29.362092   241 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:07
+I0827 04:36:29.362147   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.362201   238 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:07
+I0827 04:36:29.362243   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.392091   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:29.392130   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:29.394111   238 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:07
+I0827 04:36:29.394127   241 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:07
+I0827 04:36:29.394166   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.394286   238 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:29 TP19 EP19] Init torch distributed begin.
+I0827 04:36:29.420166   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:29.421864   241 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:07
+I0827 04:36:29.421993   241 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:29 TP22 EP22] Init torch distributed begin.
+[2026-08-27 04:36:48 TP23 EP23] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP22 EP22] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP21 EP21] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP20 EP20] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP19 EP19] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP18 EP18] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP17 EP17] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP16 EP16] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP22 EP22] Init torch distributed ends. elapsed=18.81 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP23 EP23] Init torch distributed ends. elapsed=18.94 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP21 EP21] Init torch distributed ends. elapsed=18.93 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP20 EP20] Init torch distributed ends. elapsed=18.91 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP19 EP19] Init torch distributed ends. elapsed=18.83 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP18 EP18] Init torch distributed ends. elapsed=18.92 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP17 EP17] Init torch distributed ends. elapsed=18.92 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP16 EP16] Init torch distributed ends. elapsed=18.92 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP21 EP21] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP16 EP16] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP22 EP22] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP20 EP20] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP23 EP23] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP18 EP18] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP19 EP19] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP22 EP22] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP20 EP20] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP21 EP21] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP16 EP16] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP23 EP23] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP19 EP19] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP18 EP18] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP17 EP17] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP17 EP17] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP20 EP20] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP22 EP22] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP18 EP18] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP23 EP23] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP19 EP19] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP21 EP21] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP20 EP20] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP16 EP16] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP18 EP18] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP22 EP22] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP19 EP19] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP21 EP21] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP23 EP23] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP16 EP16] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP17 EP17] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP17 EP17] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:50 TP23 EP23] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP22 EP22] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP21 EP21] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP19 EP19] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP18 EP18] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP20 EP20] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP16 EP16] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP17 EP17] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:38:53 TP16 EP16] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:53 TP16 EP16] Load weight end. elapsed=124.66 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:53 TP17 EP17] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:53 TP17 EP17] Load weight end. elapsed=124.69 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:53 TP23 EP23] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:53 TP23 EP23] Load weight end. elapsed=124.71 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:53 TP21 EP21] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:53 TP21 EP21] Load weight end. elapsed=124.74 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:53 TP18 EP18] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:53 TP18 EP18] Load weight end. elapsed=124.82 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:53 TP20 EP20] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:53 TP20 EP20] Load weight end. elapsed=124.82 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:53 TP22 EP22] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:53 TP22 EP22] Load weight end. elapsed=124.92 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:53 TP19 EP19] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:53 TP19 EP19] Load weight end. elapsed=125.03 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:39:01 TP17 EP17] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP17 EP17] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP23 EP23] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP23 EP23] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP16 EP16] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP16 EP16] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP18 EP18] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP18 EP18] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP22 EP22] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP22 EP22] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP21 EP21] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP21 EP21] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP20 EP20] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP20 EP20] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP19 EP19] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP19 EP19] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:02 TP23 EP23] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP18 EP18] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP21 EP21] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP22 EP22] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP20 EP20] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP17 EP17] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP19 EP19] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP16 EP16] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP21 EP21] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP22 EP22] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP23 EP23] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP17 EP17] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP16 EP16] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP20 EP20] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP18 EP18] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP19 EP19] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP21 EP21] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP23 EP23] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP22 EP22] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP17 EP17] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP23 EP23] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP21 EP21] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP20 EP20] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP16 EP16] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP22 EP22] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP17 EP17] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP20 EP20] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP16 EP16] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP18 EP18] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP19 EP19] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP18 EP18] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP19 EP19] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP23 EP23] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP23 EP23] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:39:02 TP17 EP17] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP17 EP17] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP16 EP16] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP16 EP16] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:39:02 TP21 EP21] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP21 EP21] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP22 EP22] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP22 EP22] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP20 EP20] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP20 EP20] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:39:02 TP18 EP18] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP18 EP18] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP19 EP19] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP19 EP19] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:42:56 TP19 EP19] Capture target decode CUDA graph end. elapsed=233.80 s, mem usage=0.62 GB, avail mem=8.51 GB.
+[2026-08-27 04:42:56 TP19 EP19] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP16 EP16] Capture target decode CUDA graph end. elapsed=233.84 s, mem usage=0.60 GB, avail mem=8.52 GB.
+[2026-08-27 04:42:56 TP16 EP16] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP22 EP22] Capture target decode CUDA graph end. elapsed=233.84 s, mem usage=0.56 GB, avail mem=8.59 GB.
+[2026-08-27 04:42:56 TP22 EP22] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP17 EP17] Capture target decode CUDA graph end. elapsed=233.88 s, mem usage=0.58 GB, avail mem=8.58 GB.
+[2026-08-27 04:42:56 TP17 EP17] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP21 EP21] Capture target decode CUDA graph end. elapsed=233.90 s, mem usage=0.58 GB, avail mem=8.58 GB.
+[2026-08-27 04:42:56 TP21 EP21] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP20 EP20] Capture target decode CUDA graph end. elapsed=233.91 s, mem usage=0.60 GB, avail mem=8.52 GB.
+[2026-08-27 04:42:56 TP20 EP20] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP23 EP23] Capture target decode CUDA graph end. elapsed=234.01 s, mem usage=0.62 GB, avail mem=8.51 GB.
+[2026-08-27 04:42:56 TP23 EP23] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP18 EP18] Capture target decode CUDA graph end. elapsed=234.00 s, mem usage=0.56 GB, avail mem=8.59 GB.
+[2026-08-27 04:42:56 TP18 EP18] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+W0827 04:42:57.619849  2423 memory_location.cpp:73] Failed to get NUMA node, addr: 0x9680d380, len: 23168: Operation not permitted [1]
+W0827 04:42:57.619855  2422 memory_location.cpp:73] Failed to get NUMA node, addr: 0x9696f4c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.619967  2424 memory_location.cpp:73] Failed to get NUMA node, addr: 0x966bf4c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.620213  2425 memory_location.cpp:73] Failed to get NUMA node, addr: 0x82b481c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.620281  2426 memory_location.cpp:73] Failed to get NUMA node, addr: 0x988cf580, len: 185344: Operation not permitted [1]
+W0827 04:42:57.620285  2427 memory_location.cpp:73] Failed to get NUMA node, addr: 0x988fc9c0, len: 185344: Operation not permitted [1]
+W0827 04:42:57.620325  2428 memory_location.cpp:73] Failed to get NUMA node, addr: 0x96988ec0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.620384  2429 memory_location.cpp:73] Failed to get NUMA node, addr: 0x96949540, len: 46336: Operation not permitted [1]
+W0827 04:42:57.620546  2430 memory_location.cpp:73] Failed to get NUMA node, addr: 0x96790780, len: 23168: Operation not permitted [1]
+W0827 04:42:57.620606  2431 memory_location.cpp:73] Failed to get NUMA node, addr: 0x967ccac0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.651996  2570 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6c8ac200, len: 23168: Operation not permitted [1]
+W0827 04:42:57.652108  2571 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6c6f5f80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.652212  2572 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6c8ea380, len: 23168: Operation not permitted [1]
+W0827 04:42:57.652417  2573 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6d2e8ec0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.652568  2574 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6ea4eb80, len: 185344: Operation not permitted [1]
+W0827 04:42:57.652587  2576 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6cb20d80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.652642  2577 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6c987340, len: 46336: Operation not permitted [1]
+W0827 04:42:57.652667  2575 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6ea7bfc0, len: 185344: Operation not permitted [1]
+W0827 04:42:57.652815  2578 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6c6eb200, len: 23168: Operation not permitted [1]
+W0827 04:42:57.652889  2579 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6cb0a2c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.713306  2681 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8a7feb00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.713331  2680 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ad55240, len: 23168: Operation not permitted [1]
+W0827 04:42:57.713303  2679 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ad28280, len: 23168: Operation not permitted [1]
+W0827 04:42:57.713424  2682 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8abc1200, len: 23168: Operation not permitted [1]
+W0827 04:42:57.713523  2683 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ca64980, len: 185344: Operation not permitted [1]
+W0827 04:42:57.713629  2685 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ca91dc0, len: 185344: Operation not permitted [1]
+W0827 04:42:57.713903  2687 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ac54c00, len: 46336: Operation not permitted [1]
+W0827 04:42:57.714316  2688 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ab93e40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.714335  2689 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8abe8700, len: 23168: Operation not permitted [1]
+W0827 04:42:57.714352  2686 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8a7f7880, len: 23168: Operation not permitted [1]
+W0827 04:42:57.717235  2716 memory_location.cpp:73] Failed to get NUMA node, addr: 0x739fc3c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.717247  2718 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ff359c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.717484  2725 memory_location.cpp:73] Failed to get NUMA node, addr: 0x75e8fa00, len: 185344: Operation not permitted [1]
+W0827 04:42:57.717736  2729 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ff46700, len: 46336: Operation not permitted [1]
+W0827 04:42:57.718014  2728 memory_location.cpp:73] Failed to get NUMA node, addr: 0x73b6a280, len: 23168: Operation not permitted [1]
+W0827 04:42:57.718112  2730 memory_location.cpp:73] Failed to get NUMA node, addr: 0x73dda340, len: 23168: Operation not permitted [1]
+W0827 04:42:57.718163  2731 memory_location.cpp:73] Failed to get NUMA node, addr: 0x73dbf000, len: 23168: Operation not permitted [1]
+W0827 04:42:57.718196  2721 memory_location.cpp:73] Failed to get NUMA node, addr: 0x73a8f640, len: 23168: Operation not permitted [1]
+W0827 04:42:57.718261  2723 memory_location.cpp:73] Failed to get NUMA node, addr: 0x73c5ba40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.724411  2727 memory_location.cpp:73] Failed to get NUMA node, addr: 0x75ebce40, len: 185344: Operation not permitted [1]
+W0827 04:42:57.755818  2858 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5e996800, len: 23168: Operation not permitted [1]
+W0827 04:42:57.755931  2859 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ec52140, len: 23168: Operation not permitted [1]
+W0827 04:42:57.756026  2861 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5e767ec0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.756074  2862 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60bbb440, len: 185344: Operation not permitted [1]
+W0827 04:42:57.756379  2865 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4aecca80, len: 46336: Operation not permitted [1]
+W0827 04:42:57.756517  2863 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60be8880, len: 185344: Operation not permitted [1]
+W0827 04:42:57.756536  2864 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5eb0d880, len: 23168: Operation not permitted [1]
+W0827 04:42:57.756650  2866 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ea7fe80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.757373  2860 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5eb4d800, len: 23168: Operation not permitted [1]
+W0827 04:42:57.757380  2867 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ea97dc0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.818434  3070 memory_location.cpp:73] Failed to get NUMA node, addr: 0x965568c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.818459  3075 memory_location.cpp:73] Failed to get NUMA node, addr: 0x96700fc0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.818470  3072 memory_location.cpp:73] Failed to get NUMA node, addr: 0x96541bc0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.820174  3079 memory_location.cpp:73] Failed to get NUMA node, addr: 0x984fb440, len: 185344: Operation not permitted [1]
+W0827 04:42:57.822142  3084 memory_location.cpp:73] Failed to get NUMA node, addr: 0x962df200, len: 23168: Operation not permitted [1]
+W0827 04:42:57.822369  3086 memory_location.cpp:73] Failed to get NUMA node, addr: 0x831c0780, len: 46336: Operation not permitted [1]
+W0827 04:42:57.822384  3110 memory_location.cpp:73] Failed to get NUMA node, addr: 0x965cd3c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.822431  3082 memory_location.cpp:73] Failed to get NUMA node, addr: 0x98528880, len: 185344: Operation not permitted [1]
+W0827 04:42:57.822527  3127 memory_location.cpp:73] Failed to get NUMA node, addr: 0x9642bbc0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.829262  3077 memory_location.cpp:73] Failed to get NUMA node, addr: 0x9640b580, len: 23168: Operation not permitted [1]
+W0827 04:42:57.832078  3193 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ce567c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.832136  3194 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8cbe8d00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.832149  3195 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8cc94640, len: 23168: Operation not permitted [1]
+W0827 04:42:57.832294  3196 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8d79f140, len: 23168: Operation not permitted [1]
+W0827 04:42:57.832419  3199 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ce4bec0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.832562  3201 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8d7e4b40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.832683  3202 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8cbbd940, len: 23168: Operation not permitted [1]
+W0827 04:42:57.833209  3198 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ef88080, len: 185344: Operation not permitted [1]
+W0827 04:42:57.833267  3200 memory_location.cpp:73] Failed to get NUMA node, addr: 0x79c278c0, len: 46336: Operation not permitted [1]
+W0827 04:42:57.837312  3197 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ef5ac40, len: 185344: Operation not permitted [1]
+W0827 04:42:58.193480  3458 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59a23f40, len: 23168: Operation not permitted [1]
+W0827 04:42:58.193507  3459 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59aa70c0, len: 23168: Operation not permitted [1]
+W0827 04:42:58.193545  3460 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5999c100, len: 23168: Operation not permitted [1]
+W0827 04:42:58.193585  3461 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59bc3040, len: 23168: Operation not permitted [1]
+W0827 04:42:58.193631  3462 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5bd1e5c0, len: 185344: Operation not permitted [1]
+W0827 04:42:58.193667  3463 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5bd4ba00, len: 185344: Operation not permitted [1]
+W0827 04:42:58.193720  3464 memory_location.cpp:73] Failed to get NUMA node, addr: 0x599fa280, len: 23168: Operation not permitted [1]
+W0827 04:42:58.193746  3466 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59dd6100, len: 23168: Operation not permitted [1]
+W0827 04:42:58.193764  3467 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59cbd400, len: 23168: Operation not permitted [1]
+W0827 04:42:58.193776  3465 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59db0340, len: 46336: Operation not permitted [1]
+[2026-08-27 04:43:00] Dummy health check server started in background thread at 0.0.0.0:30000
+I0827 04:48:13.990782  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:13.991083  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.005937  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.006215  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.026726  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.034708  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.038136  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.038393  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.052722  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.052948  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.069641  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.089646  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.092875  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.099138  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.099377  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.099617  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.102869  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.103116  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.103413  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.103614  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.103897  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.104102  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.104316  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.107553  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.107847  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.113867  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.125564  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.131559  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.131772  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.140426  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.140640  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.140889  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.141057  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.144480  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.144747  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.160434  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.160689  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.160965  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.177232  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.188812  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.189050  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.189292  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.189566  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.189749  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.197841  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.198038  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.198292  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.198523  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.198736  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.198998  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.210616  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.210834  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.211103  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.214471  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.216403  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.217782  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.227747  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.228065  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.228319  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.228518  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.230293  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.244758  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.244995  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.245216  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.246076  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.247792  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.248039  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.248274  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.251454  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.255178  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.255391  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.256364  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.256808  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.260069  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.260272  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.260452  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.260666  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.260878  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.266151  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.266335  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.266537  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.266744  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.266952  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.267238  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.267465  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.268203  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.270165  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.272876  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.273144  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.275923  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.276090  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.281610  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.281848  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.285796  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.285949  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.287339  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.289616  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.289852  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.290091  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.290292  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.290467  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.290657  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.290853  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.291211  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.295532  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.297179  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.297688  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.299065  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.299283  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.300338  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.309764  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.310025  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.310344  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.310527  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.310770  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.314235  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.317247  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.318601  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.324312  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.338757  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.343169  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.344022  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.344889  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.353394  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.354696  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.354852  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.360512  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.363333  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.363579  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.363832  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.364054  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.364255  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.364542  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.364830  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.368968  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.369174  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.369390  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.373200  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.374004  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.374215  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.374486  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.379066  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.379289  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.383756  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.384938  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.393990  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.394248  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.394496  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.400096  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.400372  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.400589  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.400769  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.400981  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.401257  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.401494  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.401587  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.402990  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.409467  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.409701  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.410995  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.412505  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.423301  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.423527  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.423753  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.425726  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.426009  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.426232  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.429703  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.432900  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.436066  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.437069  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.437340  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.437568  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.437778  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.438095  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.441175  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.442378  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.443518  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.443746  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.443964  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.444185  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.444403  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.453183  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.454483  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.458822  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.460139  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.460391  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.461473  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.468333  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.469964  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.474977  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.475178  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.475373  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.475994  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.476274  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.482146  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.483319  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.483541  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.483824  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.487998  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.491807  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.492008  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.492403  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.492631  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.492843  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.493103  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.493304  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.493522  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.493757  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.493925  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.493986  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.494186  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.494228  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.494501  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.494679  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.494771  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.494979  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.495187  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.495486  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.495695  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.495930  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.496155  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.496392  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.498075  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.499614  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.499698  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.507068  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.507290  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.515178  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.515554  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.515731  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.515975  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.516929  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.517546  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.519863  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.520967  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.522394  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.522583  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.522799  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.523052  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.523262  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.523473  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.523695  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.523783  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.523902  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.524128  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.524420  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.524629  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.525529  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.525698  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.526990  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.535063  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.535307  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.535513  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.536439  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.536978  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.537204  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.537415  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.537595  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.539846  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.540122  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.540378  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.540510  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.540766  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.541023  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.541293  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.544811  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.547500  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.547943  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.548959  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.551965  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.553215  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.563808  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.564100  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.568046  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.569260  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.569701  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.571031  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.573542  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.579483  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.581025  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.587301  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.587486  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.587728  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.588649  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.594245  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.594436  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.594633  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.594759  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.596094  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.596485  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.596745  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.597023  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.597223  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.597502  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.597754  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.598001  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.598238  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.598425  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.598681  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.598914  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.599107  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.599298  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.601323  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.601538  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.602514  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.606913  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.611363  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.611377  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.612835  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.615579  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.616807  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.616865  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.616990  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.617111  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.617369  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.617590  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.617800  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.617998  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.618206  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.621685  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.623104  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.633111  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.633315  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.634286  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.634464  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.634523  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.634792  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.634966  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.635149  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.635445  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.635624  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.635821  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.636071  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.637221  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.643270  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.644847  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.648780  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.648998  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.656845  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.657044  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.657281  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.658142  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.660220  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.661625  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.664076  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.665403  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.667707  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.669119  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.670938  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.677649  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.677932  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.677983  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.678122  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.678347  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.678530  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.678735  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.678929  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.679240  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.679463  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.679695  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.679894  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.680079  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.680286  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.680457  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.680696  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.681843  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.683002  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.683204  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.683274  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.683362  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.683475  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.683499  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.683585  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.683856  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.684543  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.687750  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.689368  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.690239  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.691855  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.693795  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.696607  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.696817  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.698160  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.703644  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.706352  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.706555  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.706766  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.707012  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.707203  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.707451  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.707659  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.707913  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.708099  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.708461  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.708745  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.708987  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.709229  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.713573  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.714908  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.718238  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.718487  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.718716  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.719660  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.720849  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.722200  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.724702  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.725086  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.725322  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.725628  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.725817  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.726325  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.731271  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.736089  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.736286  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.736464  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.737399  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.740564  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.742311  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.743717  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.746318  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.746944  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.747165  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.748356  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.749109  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.749446  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.749665  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.750972  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.755626  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.756073  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.756265  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.757405  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.762998  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.764305  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.765604  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.765775  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.766307  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.766513  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.766744  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.766980  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.767177  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.767390  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.767621  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.767817  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.775632  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.778723  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.778945  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.779145  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.780236  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.783447  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.786221  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.786506  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.789175  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.789419  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.790586  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.790643  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.790793  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.790974  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.791939  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.796165  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.797827  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.802098  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.813362  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.814442  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.814667  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.814668  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.814859  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.820125  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.820905  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.821156  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.822324  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.832154  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.832475  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.832693  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.832880  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.833066  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.833330  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.833523  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.833789  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.834038  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.834266  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.834484  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.834671  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.834861  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.835052  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.835216  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.835410  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.838090  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.848765  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.849062  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.849324  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.849581  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.849845  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.850172  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.858408  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.858647  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.859768  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.860574  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.860812  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.861027  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.861253  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.861526  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.861723  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.861910  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.862110  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.862299  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.862488  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.864478  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.872546  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.872754  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.872980  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.873731  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.873880  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.873962  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.874325  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.875612  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.877731  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.879717  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.880618  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.885710  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.886065  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.886333  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.890219  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.891563  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.895009  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.903152  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.905467  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.913136  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.915786  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.916013  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.918720  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.918928  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.919982  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.920882  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.921152  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.921317  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.921556  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.921797  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.921981  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.922211  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.924968  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.925161  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.926721  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.930300  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.931375  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.932979  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.933200  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.935876  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.941114  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.943547  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.943744  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.943778  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.943972  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.944057  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.944993  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.954442  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.955266  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.955477  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.955673  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.956692  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.974392  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.974649  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.974893  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.975108  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.985150  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.985353  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.985610  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.985852  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.986055  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.986230  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.986440  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.986611  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.991794  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.991982  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.992164  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.992372  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.004820  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.007462  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.012593  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.012835  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.013056  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.013247  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.013500  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.013749  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.016324  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.016539  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.021533  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.024173  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.024427  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.025575  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.067353  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.067580  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.067747  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.067951  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.068106  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.068390  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.068591  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.068789  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.078953  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.079149  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.079314  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.080277  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.090672  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.090881  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.091123  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.092058  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.132000  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.132241  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.132424  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.132637  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.132885  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.133169  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.133404  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.133606  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.133805  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.133991  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.134207  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.134473  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.134696  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.134949  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.135172  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.135368  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.160171  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.160442  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.160633  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.160794  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.161010  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.161175  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.161370  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.161597  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.161856  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.162032  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.171926  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.172152  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.172313  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.173470  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.217530  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.217757  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.217985  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.218160  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.218354  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.218546  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.218766  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.218966  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.229063  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.229290  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.229496  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.230377  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.240563  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.240742  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.240916  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.241938  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.282371  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.282605  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.282783  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.282986  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.283192  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.283366  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.283558  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.283744  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.283931  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.284118  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.284356  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.284595  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.284804  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.285007  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.285272  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.290516  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.310380  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.310582  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.310774  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.310997  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.311203  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.311424  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.311609  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.311781  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.311981  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.312191  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.322089  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.322285  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.322484  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.323439  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.366820  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.367026  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.367223  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.367421  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.367637  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.367813  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.368016  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.368258  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.378153  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.378351  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.378584  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.379503  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.389467  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.389665  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.389904  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.390815  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.430491  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.430742  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.430959  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.431205  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.431394  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.431623  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.431800  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.432049  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.432286  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.432466  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.432688  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.432901  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.433138  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.433328  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.433526  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.436129  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.458094  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.458309  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.458535  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.458752  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.458948  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.459110  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.459302  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.459517  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.459699  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.459882  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.469916  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.470148  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.470368  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.471292  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.448522  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.450094  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.460287  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.460638  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.462020  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.466852  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.473448  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.481580  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.482398  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.482719  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.483039  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.483215  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.483292  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.483968  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.488898  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.493845  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.494148  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.495286  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.501287  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.504213  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.512477  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.512745  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.513012  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.513972  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.297312  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.299115  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.304275  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.314786  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.316474  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.328058  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.328325  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.328570  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.329540  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.338943  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.339195  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.339458  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.340605  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.341210  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.346589  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.357823  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.358044  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.358083  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.358332  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.358369  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.359782  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.363950  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.366775  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.367130  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.368332  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.369814  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.374266  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.375430  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.378304  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.380679  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.381263  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.381505  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.381752  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.382050  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.382229  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.382687  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.386806  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.386886  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.388373  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.392937  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.394059  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.395437  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.395885  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.396198  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.397209  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.397620  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.398900  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.399183  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.399243  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.400318  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.401305  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.403239  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.404074  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.404656  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.409005  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.411810  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.414434  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.415934  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.417624  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.417887  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.418118  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.420249  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.423964  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.424198  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.424408  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.430177  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.433308  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.433637  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.436582  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.439628  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.440941  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.394086  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.394385  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.395802  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.408921  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.409214  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.410487  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.420368  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.420595  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.421932  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.426435  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.573661  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.574067  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.575153  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.109197  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.110759  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.117308  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.127852  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.129487  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.134262  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.139751  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.141630  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.146772  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.148381  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.148725  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.149887  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.154366  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.161194  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.162659  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.168562  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.654505  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.655917  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.669334  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.669617  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.669867  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.670902  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.679297  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.680840  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.687160  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.692883  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.693225  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.693225  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.693482  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.693727  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.694342  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.694828  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.699385  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.700112  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.701340  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.702790  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.708935  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.719955  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.720265  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.720535  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.721411  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.727005  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.728433  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.462263  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.462824  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.463836  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.475934  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.477473  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.481695  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.487771  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.489451  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.495987  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.185277  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.186878  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.193236  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:47.566026  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:47.575708  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:47.584995  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:48.551697  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:48.561645  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:48.570935  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:49.336337  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:49.347584  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:49.348870  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:49.353668  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:49.362033  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:50.265223  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:50.274431  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:50.283521  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.110039  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.122382  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.127803  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.131213  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.163244  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.171993  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.179919  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.997390  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.007871  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.013643  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.018814  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.023959  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.033485  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.799268  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.810883  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.819684  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:53.708249  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:53.717978  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:53.727499  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:53.735821  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:54.567750  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:54.576763  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:55.320358  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:55.447818  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:55.457465  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:57.588092  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:57.606285  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:57.611370  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:57.639269  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:57.640733  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:57.646224  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:57.660796  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:57.732604  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:57.734110  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:58.280643  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:58.297982  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:59.042151  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:59.051748  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:08.244721  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:08.253723  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:09.194365  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:10.135797  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:10.203588  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:10.219120  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:10.975451  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:10.985549  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:11.843453  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:11.845074  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:12.790898  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:13.633699  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:13.643146  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:15.418679  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:17.010130  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:17.183473  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:17.197549  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:17.986510  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:18.003332  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:18.214159  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:18.224337  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:18.241684  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:18.243786  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:18.258334  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:18.350939  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:18.364851  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:18.893000  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:29.860329  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:46.812764  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-20260827-1325/logs/manual_d_node3.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-20260827-1325/logs/manual_d_node3.log
new file mode 100644
index 0000000..1c0a7ea
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-20260827-1325/logs/manual_d_node3.log
@@ -0,0 +1,1461 @@
+/usr/local/lib/python3.12/dist-packages/torchao/quantization/quant_api.py:1731: SyntaxWarning: invalid escape sequence '\.'
+  """Configuration class for applying different quantization configs to modules or parameters based on their fully qualified names (FQNs).
+/sgl-workspace/sglang/python/sglang/launch_server.py:56: UserWarning: 'python -m sglang.launch_server' is still supported, but 'sglang serve' is the recommended entrypoint.
+  Example: sglang serve --model-path <model> [options]
+  warnings.warn(
+[2026-08-27 04:36:00] KV cache is forced as chunk cache for decode server
+[2026-08-27 04:36:03] Breakable CUDA graph is incompatible with MLA attention; disabling prefill CUDA graph.
+[2026-08-27 04:36:03] Attention backend not specified. Use triton backend by default.
+[2026-08-27 04:36:05] server_args=ServerArgs(model_path='/data/hf_models/Kimi-K3', tokenizer_path='/data/hf_models/Kimi-K3', tokenizer_mode='auto', tokenizer_backend='huggingface', tokenizer_worker_num=1, detokenizer_worker_num=1, skip_tokenizer_init=False, load_format='auto', model_loader_extra_config='{}', trust_remote_code=True, context_length=None, is_embedding=False, enable_multimodal=None, revision=None, model_impl='auto', model_config_parser='auto', json_model_override_args='{}', dtype='auto', quantization=None, quantization_param_path=None, kv_cache_dtype='auto', enable_fp32_lm_head=False, modelopt_quant=None, modelopt_checkpoint_restore_path=None, modelopt_checkpoint_save_path=None, modelopt_export_path=None, quantize_and_serve=False, rl_quant_profile=None, enable_tf32_matmul=False, mem_fraction_static=0.88, max_running_requests=None, max_queued_requests=None, max_total_tokens=None, chunked_prefill_size=8192, enable_dynamic_chunking=False, max_prefill_tokens=16384, prefill_max_requests=None, schedule_policy='fcfs', enable_priority_scheduling=False, disable_priority_preemption=False, default_priority_value=None, abort_on_priority_when_disabled=False, schedule_low_priority_values_first=False, priority_scheduling_preemption_threshold=10, retraction_policy='length', schedule_conservativeness=1.0, page_size=64, swa_full_tokens_ratio=0.8, disable_hybrid_swa_memory=False, radix_eviction_policy='lru', prefill_only_disable_kv_cache=False, disable_radix_cache=True, enable_page_major_kv_layout=False, enable_unified_memory=False, disable_chunked_prefix_cache=False, disable_overlap_schedule=False, num_continuous_decode_steps=1, scheduler_recv_interval=1, enable_mixed_chunk=False, nccl_port=None, dist_timeout=3600, dist_init_addr='174.1.60.5:20000', nnodes=4, node_rank=3, tp_size=32, dcp_size=1, pp_size=1, pp_max_micro_batch_size=None, pp_async_batch_depth=0, dp_size=1, load_balance_method='round_robin', attn_cp_size=1, moe_dp_size=1, dcp_comm_backend='ag_rs', dcp_replicate_q_proj=None, dwdp_size=1, enable_prefill_cp=False, cp_strategy=None, enable_dsa_cache_layer_split=False, enable_dsa_prefill_context_parallel=False, dsa_prefill_cp_mode='round-robin-split', enable_prefill_context_parallel=False, prefill_cp_mode='in-seq-split', enable_cp_decode_attn_tp=False, enable_dp_attention=False, enable_dp_attention_local_control_broadcast=False, enable_dp_lm_head=False, enable_attn_tp_input_scattered=False, disable_attn_tp_gather=False, enable_p2p_check=False, device='cuda', base_gpu_id=0, gpu_id_step=1, random_seed=297832107, watchdog_timeout=300, soft_watchdog_timeout=None, sleep_on_idle=False, use_ray=False, custom_sigquit_handler=None, numa_node=None, gc_threshold=None, host='0.0.0.0', port=30000, fastapi_root_path='', smg_grpc_mode=False, grpc_mode=False, grpc_port=None, sidecar=None, sidecar_args=None, skip_server_warmup=False, warmups=None, enable_http2=False, ssl_keyfile=None, ssl_certfile=None, ssl_ca_certs=None, ssl_keyfile_password=None, enable_ssl_refresh=False, api_key=None, admin_api_key=None, served_model_name='kimi-k3', weight_version='default', chat_template=None, hf_chat_template_name=None, completion_template=None, file_storage_path='sglang_storage', enable_cache_report=False, reasoning_parser=None, default_chat_template_kwargs=None, strip_thinking_cache=False, enable_strict_thinking=False, tool_call_parser=None, tool_server=None, sampling_defaults='model', asr_max_buffer_seconds=60, asr_max_concurrent_sessions=32, preferred_sampling_params=None, allow_auto_truncate=False, stream_interval=1, batch_notify_size=16, stream_response_default_include_usage=False, incremental_streaming_output=False, enable_streaming_session=False, enable_session_radix_cache=False, log_level='info', log_level_http=None, log_requests=False, log_requests_level=2, log_requests_format='text', log_requests_target=None, uvicorn_access_log_exclude_prefixes=[], crash_dump_folder=None, show_time_cost=False, enable_metrics=False, smg_http_sidecar_port=None, enable_mfu_metrics=False, enable_metrics_for_all_schedulers=False, load_snapshot_publish_interval=15, tokenizer_metrics_custom_labels_header='x-custom-labels', tokenizer_metrics_allowed_custom_labels=None, extra_metric_labels=None, bucket_time_to_first_token=None, bucket_inter_token_latency=None, bucket_e2e_request_latency=None, prompt_tokens_buckets=None, generation_tokens_buckets=None, gc_warning_threshold_secs=0.0, decode_log_interval=40, enable_request_time_stats_logging=False, kv_events_config=None, enable_forward_pass_metrics=False, forward_pass_metrics_worker_id='', forward_pass_metrics_ipc_name=None, enable_trace=False, trace_modules='request', otlp_traces_endpoint='localhost:4317', export_metrics_to_file=False, export_metrics_to_file_dir=None, stat_loggers=None, constrained_json_whitespace_pattern=None, constrained_json_disable_any_whitespace=False, attention_backend='triton', decode_attention_backend=None, prefill_attention_backend=None, sampling_backend='flashinfer', grammar_backend='xgrammar', radix_cache_backend=None, mm_attention_backend=None, fp8_gemm_runner_backend='auto', fp4_gemm_runner_backend='auto', bf16_gemm_backend='auto', dsa_prefill_backend=None, dsa_decode_backend=None, dsa_paged_mqa_logits_backend='auto', dsa_topk_backend='sgl-kernel', disable_flashinfer_autotune=False, flashinfer_autotune_skip_ops=None, mamba_backend='triton', cuda_graph_config=CudaGraphConfig(decode=PhaseConfig(backend='full', max_bs=16, bs=[1, 2, 4, 8, 12, 16], tc_compiler='eager', full_prefill_max_req=None), prefill=PhaseConfig(backend='disabled', max_bs=2048, bs=[4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], tc_compiler='eager', full_prefill_max_req=None)), cuda_graph_backend_decode=None, cuda_graph_backend_prefill=None, cuda_graph_max_bs_decode=16, cuda_graph_max_bs_prefill=None, cuda_graph_bs_decode=None, cuda_graph_bs_prefill=None, cuda_graph_tc_compiler=None, disable_prefill_cuda_graph=False, disable_decode_cuda_graph=False, disable_cuda_graph=False, disable_cuda_graph_padding=False, enable_profile_cuda_graph=False, enable_cudagraph_gc=False, debug_cuda_graph=False, enable_layerwise_nvtx_marker=False, enable_nccl_nvls=False, enable_symm_mem=False, triton_attention_reduce_in_fp32=False, triton_attention_num_kv_splits=8, triton_attention_split_tile_size=None, flashinfer_mla_disable_ragged=False, enable_fused_qk_norm_rope=False, enable_precise_embedding_interpolation=False, enable_fused_moe_sum_all_reduce=False, enable_deepseek_v4_fp4_indexer=False, disable_custom_all_reduce=False, enable_mscclpp=False, enable_torch_symm_mem=False, enable_scattered_sconv=False, pre_warm_nccl=False, enable_quant_communications=False, enable_flashinfer_allreduce_fusion=False, enforce_disable_flashinfer_allreduce_fusion=False, flashinfer_allreduce_fusion_backend=None, enable_aiter_allreduce_fusion=False, enable_torch_compile=False, enable_torch_compile_debug_mode=False, torch_compile_max_bs=32, torchao_config='', speculative_algorithm=None, speculative_draft_model_path=None, speculative_draft_model_revision=None, speculative_draft_load_format=None, speculative_num_steps=None, speculative_eagle_topk=None, speculative_num_draft_tokens=None, speculative_dflash_block_size=None, speculative_dspark_block_size=None, speculative_dspark_sps_table_path=None, speculative_dspark_confidence_sts_path=None, speculative_dspark_align_verify_tokens_to_graph_tier=False, speculative_accept_threshold_single=1.0, speculative_accept_threshold_acc=1.0, speculative_use_rejection_sampling=False, speculative_token_map=None, speculative_attention_mode='prefill', speculative_draft_attention_backend=None, speculative_draft_window_size=None, speculative_moe_runner_backend='marlin', speculative_moe_a2a_backend=None, speculative_draft_model_quantization=None, speculative_skip_dp_mlp_sync=False, enable_multi_layer_eagle=False, speculative_adaptive=False, speculative_adaptive_config=None, decoupled_spec_bind_endpoint=None, decoupled_spec_connect_endpoints=None, decoupled_spec_rank=None, decoupled_spec_role='null', spec_trace_dir=None, speculative_ngram_min_bfs_breadth=1, speculative_ngram_max_bfs_breadth=10, speculative_ngram_match_type='BFS', speculative_ngram_max_trie_depth=18, speculative_ngram_capacity=10000000, speculative_ngram_external_corpus_path=None, speculative_ngram_external_sam_budget=0, speculative_ngram_external_corpus_max_tokens=10000000, ep_size=32, moe_a2a_backend='none', moe_runner_backend='marlin', flashinfer_mxfp4_moe_precision='default', deepep_mode='auto', fuseep_mode=2, deepep_dispatcher_output_dtype='auto', ep_num_redundant_experts=0, ep_dispatch_algorithm=None, init_expert_location='trivial', enable_eplb=False, eplb_algorithm='auto', eplb_rebalance_num_iterations=1000, eplb_rebalance_layers_per_chunk=None, eplb_min_rebalancing_utilization_threshold=1.0, expert_distribution_recorder_mode=None, expert_distribution_recorder_buffer_size=1000, enable_expert_distribution_metrics=False, deepep_config=None, moe_dense_tp_size=None, elastic_ep_backend=None, enable_elastic_expert_backup=False, mooncake_ib_device=None, enable_waterfill=False, ep_join_mode=None, ep_join_rank_offset=0, elastic_ep_initial_size=None, max_ep_size=None, elastic_ep_scale_timeout=600, elastic_ep_rejoin=False, disable_flashinfer_cutlass_moe_fp4_allgather=False, disable_shared_experts_fusion=False, enforce_shared_experts_fusion=False, max_mamba_cache_size=None, mamba_ssm_dtype=None, mamba_max_states_per_path=-1, enable_mamba_cache_stochastic_rounding=False, mamba_cache_philox_rounds=0, mamba_full_memory_ratio=0.21, mamba_radix_cache_strategy='extra_buffer_lazy', uses_mamba_radix_cache=False, mamba_track_interval=256, enable_int8_mamba_checkpoint=False, int8_mamba_ckpt_size=None, linear_attn_backend='triton', linear_attn_decode_backend=None, linear_attn_prefill_backend=None, linear_attn_verify_backend=None, enable_linear_replayssm=False, linear_replayssm_cache_len=16, enable_linear_replayssm_spec=False, enable_hierarchical_cache=False, hicache_ratio=2.0, hicache_size=0, hicache_write_policy='write_through', hicache_io_backend='kernel', hicache_mem_layout='page_first', hicache_storage_backend=None, hicache_storage_prefetch_policy='timeout', hicache_storage_backend_extra_config=None, enable_hisparse=False, hisparse_config=None, enable_broadcast_mm_inputs_process=False, enable_prefix_mm_cache=False, mm_enable_dp_encoder=False, mm_process_config={}, mm_processor_worker_num=0, mm_io_worker_num=0, limit_mm_data_per_request=None, enable_mm_global_cache=False, disable_fast_image_processor=False, mm_feature_transport='cpu', keep_mm_feature_on_device=False, enable_lora=None, enable_lora_overlap_loading=None, max_lora_rank=None, lora_target_modules=None, lora_paths=None, max_loaded_loras=None, max_loras_per_batch=8, lora_eviction_policy='lru', lora_backend='csgmv', max_lora_chunk_size=16, experts_shared_outer_loras=None, lora_use_virtual_experts=False, lora_strict_loading=False, lora_drain_wait_threshold=0.0, enable_two_batch_overlap=False, enable_single_batch_overlap=False, tbo_token_distribution_threshold=0.48, cpu_offload_gb=0, offload_group_size=-1, offload_num_in_group=1, offload_prefetch_step=1, offload_mode='cpu', enable_lmcache=False, lmcache_config_file=None, enable_flexkv=False, flexkv_config_file=None, kt_weight_path=None, kt_method='AMXINT4', kt_cpuinfer=None, kt_threadpool_count=2, kt_num_gpu_experts=None, kt_max_deferred_experts_per_token=None, dllm_algorithm=None, dllm_algorithm_config=None, dllm_fdfo=True, disaggregation_mode='decode', disaggregation_transfer_backend='mooncake', disaggregation_bootstrap_port=28800, disaggregation_ib_device='mlx5_0,mlx5_1,mlx5_2,mlx5_3', disaggregation_decode_enable_radix_cache=False, disaggregation_decode_enable_offload_kvcache=False, num_reserved_decode_tokens=512, disaggregation_decode_extra_slots=0, disaggregation_decode_polling_interval=1, optimistic_prefill_attempts=0, encoder_only=False, language_only=False, encoder_transfer_backend='zmq_to_tokenizer', encoder_urls=[], encoder_bootstrap_port=8997, encoder_register_urls=[], enable_adaptive_dispatch_to_encoder=False, enable_pdmux=False, pdmux_config_path=None, sm_group_num=8, custom_weight_loader=[], weight_loader_disable_mmap=False, weight_loader_prefetch_checkpoints=False, weight_loader_prefetch_num_threads=4, weight_loader_drop_cache_after_load=False, remote_instance_weight_loader_seed_instance_ip=None, remote_instance_weight_loader_seed_instance_service_port=None, remote_instance_weight_loader_send_weights_group_ports=None, remote_instance_weight_loader_backend='nccl', remote_instance_weight_loader_start_seed_via_transfer_engine=False, engine_info_bootstrap_port=6789, modelexpress_config=None, download_dir=None, model_checksum=None, delete_ckpt_after_loading=False, decrypted_config_file=None, decrypted_draft_config_file=None, checkpoint_engine_wait_weights_before_ready=False, enable_prefill_delayer=False, prefill_delayer_max_delay_passes=30, prefill_delayer_token_usage_low_watermark=None, prefill_delayer_forward_passes_buckets=None, prefill_delayer_wait_seconds_buckets=None, prefill_delayer_queue_min_ratio=None, prefill_delayer_max_delay_ms=None, min_free_slots_delay=None, enable_deterministic_inference=False, rl_on_policy_target=None, kv_canary='none', kv_canary_real_data='none', kv_canary_sweep_interval=0, enable_dynamic_batch_tokenizer=False, dynamic_batch_tokenizer_batch_size=32, dynamic_batch_tokenizer_batch_timeout=0.002, enable_tokenizer_batch_encode=False, disable_tokenizer_batch_decode=False, debug_tensor_dump_output_folder=None, debug_tensor_dump_layers=None, debug_tensor_dump_input_file=None, enable_memory_saver=False, enable_weights_cpu_backup=False, enable_draft_weights_cpu_backup=False, enable_custom_logit_processor=False, enable_return_hidden_states=False, enable_return_routed_experts=False, enable_return_indexer_topk=False, disable_outlines_disk_cache=False, enable_mis=False, weight_cache_mode='off', weight_cache_socket=None, weight_cache_timeout=1800, forward_hooks=None, msprobe_dump_config=None)
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:16 TP30 EP30] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:16 TP30 EP30] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:19 TP30 EP30] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:19.321749   241 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:19.321810   241 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:19.321820   241 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.8 port: 12001
+I0827 04:36:19.321856   241 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.8:16569
+I0827 04:36:19.321940   241 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:19.343153   241 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:19.347820   241 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:19.373265   241 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:19.380316   241 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:19.383023   241 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:19.383050   241 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:19.383101   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:19.390141   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:19.391583   241 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:08
+I0827 04:36:19.391615   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:19.417495   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:19.418977   241 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:08
+I0827 04:36:19.419003   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:19.445492   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:19.446861   241 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:08
+I0827 04:36:19.446884   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:19.473483   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:19.474817   241 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:08
+I0827 04:36:19.474922   241 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:19 TP30 EP30] Init torch distributed begin.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:20 TP27 EP27] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:20 TP27 EP27] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:21 TP25 EP25] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:21 TP25 EP25] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:21 TP26 EP26] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:21 TP26 EP26] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:21 TP28 EP28] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:21 TP28 EP28] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:21 TP24 EP24] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:21 TP24 EP24] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:21 TP29 EP29] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:21 TP29 EP29] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:21 TP31 EP31] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:21 TP31 EP31] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:22 TP27 EP27] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:22.734625   238 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:22.734687   238 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:22.734699   238 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.8 port: 12001
+I0827 04:36:22.734735   238 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.8:16566
+I0827 04:36:22.734848   238 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:22.760079   238 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:22.768187   238 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:22.777189   238 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:22.786154   238 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:22.789806   238 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:22.789830   238 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:22.789880   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:22.797986   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:22.799417   238 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:08
+I0827 04:36:22.799446   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:22.825555   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:22.826912   238 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:08
+I0827 04:36:22.826946   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:22.853536   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:22.854885   238 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:08
+I0827 04:36:22.854909   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:22.881553   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:22.882910   238 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:08
+I0827 04:36:22.883030   238 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:22 TP27 EP27] Init torch distributed begin.
+[2026-08-27 04:36:22 TP25 EP25] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:22 TP28 EP28] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:22 TP26 EP26] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:23 TP29 EP29] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:23 TP24 EP24] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:23.476856   236 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:23.476951   236 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:23.476967   236 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.8 port: 12001
+I0827 04:36:23.477013   236 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.8:15425
+I0827 04:36:23.477150   236 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:23.477173   239 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:23.477272   239 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:23.477288   239 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.8 port: 12001
+I0827 04:36:23.477344   239 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.8:15710
+I0827 04:36:23.477483   239 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:23.483197   236 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:23.483803   239 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:23.488940   236 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:23.489802   239 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:23.494866   236 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:23.495663   239 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:23.500973   236 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:23.501806   239 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:23.503927   236 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:23.503952   236 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:23.503998   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.504750   239 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:23.504770   239 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:23.504812   239 rdma_context.cpp:205] Using SIEVE endpoint store
+[2026-08-27 04:36:23 TP31 EP31] Multimodal data loading enabled with 16 worker threads (auto).
+I0827 04:36:23.510515   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:23.511322   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:23.511830   236 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:08
+I0827 04:36:23.511855   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.512616   239 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:08
+I0827 04:36:23.512640   239 rdma_context.cpp:205] Using SIEVE endpoint store
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:23.522899   237 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:23.523005   237 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:23.523020   237 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.8 port: 12001
+I0827 04:36:23.523065   237 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.8:16748
+I0827 04:36:23.523202   237 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:23.529469   237 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:23.535485   237 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:23.537460   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:23.537528   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:23.538903   236 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:08
+I0827 04:36:23.538913   239 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:08
+I0827 04:36:23.538928   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.538945   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.541347   237 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:23.546439   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:23.547672   237 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:23.547796   239 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:08
+I0827 04:36:23.547820   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.550041   237 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:23.550061   237 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:23.550102   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.555119   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:23.556490   239 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:08
+I0827 04:36:23.556598   239 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:23 TP28 EP28] Init torch distributed begin.
+I0827 04:36:23.557152   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:23.558562   237 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:08
+I0827 04:36:23.558589   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.565423   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:23.566766   236 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:08
+I0827 04:36:23.566788   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.589463   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:23.590811   237 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:08
+I0827 04:36:23.590835   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.593444   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:23.594800   236 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:08
+I0827 04:36:23.594898   236 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:23 TP25 EP25] Init torch distributed begin.
+I0827 04:36:23.598161   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:23.599563   237 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:08
+I0827 04:36:23.599589   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.625564   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:23.626922   237 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:08
+I0827 04:36:23.627038   237 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:23 TP26 EP26] Init torch distributed begin.
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:23.899124   240 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:23.899245   240 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:23.899261   240 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.8 port: 12001
+I0827 04:36:23.899313   240 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.8:15127
+I0827 04:36:23.899492   240 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:23.907753   235 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:23.907863   235 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:23.907878   235 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.8 port: 12001
+I0827 04:36:23.907927   235 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.8:15330
+I0827 04:36:23.908092   235 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:23.914011   235 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:23.919618   235 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:23.925108   235 topology.cpp:127] Device mlx5_2 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:23.925450   242 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:23.925558   242 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:23.925576   242 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.8 port: 12001
+I0827 04:36:23.925631   242 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.8:16235
+I0827 04:36:23.925693   240 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:23.925761   242 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:23.930419   242 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:23.933247   240 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:23.934639   242 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:23.939190   242 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:23.940605   240 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:23.943866   242 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:23.945531   242 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:23.945554   242 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:23.945595   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.947860   240 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:23.949263   235 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:23.951108   240 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:23.951148   240 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:23.951216   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.952417   235 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:23.952445   235 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:23.952502   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.958555   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:23.959836   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:23.960163   240 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:08
+I0827 04:36:23.960197   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.961452   235 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:08
+I0827 04:36:23.961483   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.969228   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:23.969322   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:23.970860   235 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:08
+I0827 04:36:23.970890   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.971050   242 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:08
+I0827 04:36:23.971084   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.979120   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:23.980706   242 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:08
+I0827 04:36:23.980736   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.989722   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:23.991313   240 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:08
+I0827 04:36:23.991346   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:24.001633   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:24.003211   235 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:08
+I0827 04:36:24.003237   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:24.005753   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:24.007426   242 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:08
+I0827 04:36:24.007457   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:24.021715   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:24.023293   240 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:08
+I0827 04:36:24.023322   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:24.034039   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:24.034071   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:24.035696   235 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:08
+I0827 04:36:24.035816   235 transfer_engine_impl.cpp:388] installTransport, type=rdma
+I0827 04:36:24.035794   242 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:08
+I0827 04:36:24.035987   242 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:24 TP24 EP24] Init torch distributed begin.
+[2026-08-27 04:36:24 TP31 EP31] Init torch distributed begin.
+I0827 04:36:24.049829   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:24.051388   240 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:08
+I0827 04:36:24.051532   240 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:24 TP29 EP29] Init torch distributed begin.
+[2026-08-27 04:36:48 TP31 EP31] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP30 EP30] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP28 EP28] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP29 EP29] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP27 EP27] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP26 EP26] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP25 EP25] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP24 EP24] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP31 EP31] Init torch distributed ends. elapsed=24.20 s, mem usage=0.40 GB
+[2026-08-27 04:36:48 TP30 EP30] Init torch distributed ends. elapsed=28.76 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP29 EP29] Init torch distributed ends. elapsed=24.18 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP28 EP28] Init torch distributed ends. elapsed=24.68 s, mem usage=0.40 GB
+[2026-08-27 04:36:48 TP27 EP27] Init torch distributed ends. elapsed=25.35 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP26 EP26] Init torch distributed ends. elapsed=24.61 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP25 EP25] Init torch distributed ends. elapsed=24.64 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP24 EP24] Init torch distributed ends. elapsed=24.20 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP28 EP28] Load weight begin. avail mem=82.19 GB
+[2026-08-27 04:36:48 TP27 EP27] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP30 EP30] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP24 EP24] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP29 EP29] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP26 EP26] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP25 EP25] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP28 EP28] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP30 EP30] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP27 EP27] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP29 EP29] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP24 EP24] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP26 EP26] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP25 EP25] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP31 EP31] Load weight begin. avail mem=82.19 GB
+[2026-08-27 04:36:48 TP31 EP31] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP27 EP27] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP28 EP28] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP24 EP24] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP27 EP27] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP30 EP30] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP29 EP29] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP26 EP26] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP25 EP25] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP28 EP28] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP24 EP24] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP29 EP29] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP31 EP31] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP30 EP30] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP26 EP26] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP25 EP25] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP31 EP31] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:50 TP29 EP29] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP28 EP28] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP30 EP30] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP26 EP26] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP25 EP25] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP31 EP31] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP27 EP27] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP24 EP24] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:38:54 TP26 EP26] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:54 TP27 EP27] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:54 TP24 EP24] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:54 TP26 EP26] Load weight end. elapsed=125.61 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:54 TP29 EP29] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:54 TP27 EP27] Load weight end. elapsed=125.61 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:54 TP24 EP24] Load weight end. elapsed=125.61 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:54 TP29 EP29] Load weight end. elapsed=125.61 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:54 TP25 EP25] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:54 TP25 EP25] Load weight end. elapsed=125.64 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:54 TP28 EP28] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:54 TP28 EP28] Load weight end. elapsed=125.69 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:54 TP30 EP30] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:54 TP30 EP30] Load weight end. elapsed=125.71 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:54 TP31 EP31] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:54 TP31 EP31] Load weight end. elapsed=125.77 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:39:01 TP26 EP26] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP26 EP26] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP30 EP30] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP30 EP30] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP28 EP28] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP28 EP28] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP27 EP27] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP27 EP27] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP25 EP25] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP25 EP25] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP31 EP31] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP31 EP31] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP24 EP24] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP24 EP24] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP29 EP29] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP29 EP29] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:02 TP29 EP29] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP28 EP28] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP26 EP26] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP27 EP27] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP30 EP30] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP25 EP25] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP31 EP31] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP24 EP24] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP29 EP29] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP30 EP30] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP28 EP28] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP26 EP26] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP27 EP27] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP25 EP25] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP24 EP24] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP31 EP31] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP29 EP29] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP29 EP29] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP28 EP28] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP28 EP28] Memory pool end. avail mem=9.18 GB
+[2026-08-27 04:39:02 TP27 EP27] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP24 EP24] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP26 EP26] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP27 EP27] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP31 EP31] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP24 EP24] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP26 EP26] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP31 EP31] Memory pool end. avail mem=9.18 GB
+[2026-08-27 04:39:02 TP30 EP30] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP25 EP25] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP30 EP30] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP25 EP25] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP29 EP29] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP29 EP29] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP24 EP24] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP24 EP24] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:39:02 TP28 EP28] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP28 EP28] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.17 GB
+[2026-08-27 04:39:02 TP25 EP25] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP31 EP31] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP25 EP25] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP31 EP31] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.17 GB
+[2026-08-27 04:39:02 TP26 EP26] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP26 EP26] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP27 EP27] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP27 EP27] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:39:02 TP30 EP30] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP30 EP30] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:42:56 TP30 EP30] Capture target decode CUDA graph end. elapsed=233.69 s, mem usage=0.56 GB, avail mem=8.59 GB.
+[2026-08-27 04:42:56 TP30 EP30] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP27 EP27] Capture target decode CUDA graph end. elapsed=233.73 s, mem usage=0.62 GB, avail mem=8.51 GB.
+[2026-08-27 04:42:56 TP27 EP27] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP31 EP31] Capture target decode CUDA graph end. elapsed=233.74 s, mem usage=0.62 GB, avail mem=8.55 GB.
+[2026-08-27 04:42:56 TP31 EP31] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP26 EP26] Capture target decode CUDA graph end. elapsed=233.76 s, mem usage=0.56 GB, avail mem=8.59 GB.
+[2026-08-27 04:42:56 TP26 EP26] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP29 EP29] Capture target decode CUDA graph end. elapsed=233.82 s, mem usage=0.58 GB, avail mem=8.58 GB.
+[2026-08-27 04:42:56 TP29 EP29] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP28 EP28] Capture target decode CUDA graph end. elapsed=233.79 s, mem usage=0.60 GB, avail mem=8.56 GB.
+[2026-08-27 04:42:56 TP28 EP28] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP24 EP24] Capture target decode CUDA graph end. elapsed=233.88 s, mem usage=0.60 GB, avail mem=8.52 GB.
+[2026-08-27 04:42:56 TP24 EP24] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP25 EP25] Capture target decode CUDA graph end. elapsed=233.89 s, mem usage=0.58 GB, avail mem=8.58 GB.
+[2026-08-27 04:42:56 TP25 EP25] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+W0827 04:42:57.132246  2431 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5913c7c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.132292  2432 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58be65c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.132460  2435 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5af7a600, len: 185344: Operation not permitted [1]
+W0827 04:42:57.132488  2433 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58ed5900, len: 23168: Operation not permitted [1]
+W0827 04:42:57.132501  2436 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5afa7a40, len: 185344: Operation not permitted [1]
+W0827 04:42:57.132512  2437 memory_location.cpp:73] Failed to get NUMA node, addr: 0x590583c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.132552  2434 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58c15080, len: 23168: Operation not permitted [1]
+W0827 04:42:57.132607  2439 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59046600, len: 23168: Operation not permitted [1]
+W0827 04:42:57.132640  2438 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5afd4e80, len: 46336: Operation not permitted [1]
+W0827 04:42:57.132716  2440 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58e5fc80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.276726  2580 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a339340, len: 23168: Operation not permitted [1]
+W0827 04:42:57.276731  2579 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59705880, len: 23168: Operation not permitted [1]
+W0827 04:42:57.276762  2581 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a32b080, len: 23168: Operation not permitted [1]
+W0827 04:42:57.276839  2582 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a331fc0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.276887  2583 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5bb81840, len: 185344: Operation not permitted [1]
+W0827 04:42:57.276939  2584 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5bbaec80, len: 185344: Operation not permitted [1]
+W0827 04:42:57.276983  2585 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59b7b6c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.277042  2587 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59953f00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.277055  2586 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59517380, len: 46336: Operation not permitted [1]
+W0827 04:42:57.277086  2588 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59b87640, len: 23168: Operation not permitted [1]
+W0827 04:42:57.448414  2727 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a4bd3c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.448436  2728 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ac9eb00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.448488  2729 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a2fe240, len: 23168: Operation not permitted [1]
+W0827 04:42:57.448798  2735 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a6135c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.448812  2734 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59ed25c0, len: 46336: Operation not permitted [1]
+W0827 04:42:57.448840  2733 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a3c3080, len: 23168: Operation not permitted [1]
+W0827 04:42:57.448849  2736 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a181e80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.451195  2730 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5acaf240, len: 23168: Operation not permitted [1]
+W0827 04:42:57.451217  2731 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5c61b840, len: 185344: Operation not permitted [1]
+W0827 04:42:57.455173  2732 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5c648c80, len: 185344: Operation not permitted [1]
+W0827 04:42:57.460536  2816 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a391f40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.460651  2817 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ae89700, len: 23168: Operation not permitted [1]
+W0827 04:42:57.460843  2819 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ae0fa00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.460901  2826 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a4e2c80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.460979  2830 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ae7a300, len: 23168: Operation not permitted [1]
+W0827 04:42:57.461052  2821 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5c6d6880, len: 185344: Operation not permitted [1]
+W0827 04:42:57.461104  2820 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a22ed40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.462260  2823 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5c703cc0, len: 185344: Operation not permitted [1]
+W0827 04:42:57.463069  2827 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4678c140, len: 46336: Operation not permitted [1]
+W0827 04:42:57.463085  2829 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a645e80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.466511  2955 memory_location.cpp:73] Failed to get NUMA node, addr: 0x597592c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.466544  2958 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59908b40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.466676  2960 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a2f9c80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.466849  2953 memory_location.cpp:73] Failed to get NUMA node, addr: 0x599f9a80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.467304  2964 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5b9e4d40, len: 185344: Operation not permitted [1]
+W0827 04:42:57.470024  2966 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5993fa80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.471112  2970 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59898900, len: 23168: Operation not permitted [1]
+W0827 04:42:57.471313  2962 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5b9b7900, len: 185344: Operation not permitted [1]
+W0827 04:42:57.475107  2968 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5991bac0, len: 46336: Operation not permitted [1]
+W0827 04:42:57.475121  2972 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59b5ea40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.493731  3089 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59564740, len: 23168: Operation not permitted [1]
+W0827 04:42:57.493901  3097 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5b355380, len: 185344: Operation not permitted [1]
+W0827 04:42:57.494069  3104 memory_location.cpp:73] Failed to get NUMA node, addr: 0x45dbac00, len: 46336: Operation not permitted [1]
+W0827 04:42:57.494072  3105 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5902f500, len: 23168: Operation not permitted [1]
+W0827 04:42:57.494150  3094 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59498600, len: 23168: Operation not permitted [1]
+W0827 04:42:57.494194  3099 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5b3827c0, len: 185344: Operation not permitted [1]
+W0827 04:42:57.499953  3107 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59486740, len: 23168: Operation not permitted [1]
+W0827 04:42:57.502707  3102 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59016c80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.504236  3091 memory_location.cpp:73] Failed to get NUMA node, addr: 0x594d2200, len: 23168: Operation not permitted [1]
+W0827 04:42:57.504359  3095 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59354e40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.514688  3200 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5885e940, len: 23168: Operation not permitted [1]
+W0827 04:42:57.514716  3202 memory_location.cpp:73] Failed to get NUMA node, addr: 0x587d55c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.514720  3201 memory_location.cpp:73] Failed to get NUMA node, addr: 0x586bd5c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.514801  3204 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5aa56c00, len: 185344: Operation not permitted [1]
+W0827 04:42:57.514810  3205 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5aa84040, len: 185344: Operation not permitted [1]
+W0827 04:42:57.514727  3203 memory_location.cpp:73] Failed to get NUMA node, addr: 0x44d09c40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.514963  3206 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58692200, len: 23168: Operation not permitted [1]
+W0827 04:42:57.515017  3209 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58865100, len: 23168: Operation not permitted [1]
+W0827 04:42:57.515022  3207 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58ae8bc0, len: 46336: Operation not permitted [1]
+W0827 04:42:57.515030  3208 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58683a40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.548924  3433 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5895be80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.548955  3434 memory_location.cpp:73] Failed to get NUMA node, addr: 0x593aa600, len: 23168: Operation not permitted [1]
+W0827 04:42:57.549010  3435 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59399e80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.549088  3436 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58a4fa40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.549110  3437 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ab36800, len: 185344: Operation not permitted [1]
+W0827 04:42:57.549167  3438 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ab63c40, len: 185344: Operation not permitted [1]
+W0827 04:42:57.551182  3439 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5936aac0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.551239  3441 memory_location.cpp:73] Failed to get NUMA node, addr: 0x455ad780, len: 23168: Operation not permitted [1]
+W0827 04:42:57.551265  3442 memory_location.cpp:73] Failed to get NUMA node, addr: 0x594169c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.551280  3440 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5853c980, len: 46336: Operation not permitted [1]
+[2026-08-27 04:42:59] Dummy health check server started in background thread at 0.0.0.0:30000
+I0827 04:48:13.997741  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:13.998107  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.014123  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.014460  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.045612  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.045893  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.063987  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.064270  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.072747  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.073004  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.082927  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.089962  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.090194  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.112046  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.112280  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.112514  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.115789  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.119459  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.119691  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.119886  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.120070  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.120432  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.120639  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.120857  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.127898  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.128136  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.153139  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.153350  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.170001  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.170248  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.170490  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.170712  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.171002  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.171211  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.171445  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.185374  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.185580  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.185820  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.186059  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.186275  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.218560  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.218878  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.219130  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.219380  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.239995  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.240248  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.240523  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.240849  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.241052  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.241318  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.241586  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.241847  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.242057  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.242271  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.242539  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.256910  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.258225  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.261878  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.262174  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.262421  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.262710  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.263056  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.263356  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.263574  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.265990  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.267298  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.281507  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.281759  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.281965  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.282989  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.289211  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.290711  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.291896  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.292208  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.292472  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.292713  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.292904  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.293118  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.293318  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.301455  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.301728  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.301918  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.302150  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.302408  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.302699  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.302891  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.303146  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.303368  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.303570  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.305574  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.306890  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.314224  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.315574  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.319671  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.319937  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.320156  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.326327  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.326542  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.330421  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.330546  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.330736  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.330998  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.331259  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.331496  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.331717  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.331938  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.332140  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.332340  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.332551  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.333663  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.338346  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.339445  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.339735  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.340694  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.345638  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.360560  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.361940  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.369402  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.369591  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.370901  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.377556  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.384063  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.384287  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.384516  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.385501  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.397218  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.397437  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.397625  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.398703  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.398864  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.399148  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.407058  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.407372  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.407644  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.407857  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.408042  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.408264  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.408543  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.408807  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.411617  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.411850  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.412068  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.414391  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.415768  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.423707  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.425092  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.425122  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.430542  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.430770  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.431028  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.431236  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.431470  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.434294  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.434525  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.435901  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.436133  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.437299  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.446563  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.446866  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.451089  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.453542  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.454408  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.455781  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.467486  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.467700  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.467938  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.468218  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.468482  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.468770  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.469058  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.472146  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.473528  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.480324  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.480634  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.481218  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.483843  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.484179  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.484373  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.485363  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.498565  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.498824  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.499142  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.502869  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.508553  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.511684  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.511938  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.513093  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.523473  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.523744  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.523972  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.524215  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.529819  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.531201  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.539283  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.540769  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.542526  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.542757  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.543026  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.543277  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.543532  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.543817  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.544056  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.544275  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.544507  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.544736  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.544970  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.545243  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.545476  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.545689  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.546013  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.551106  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.551363  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.551533  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.551725  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.551961  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.552152  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.552357  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.552618  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.555388  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.555505  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.558743  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.564610  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.564857  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.569089  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.570464  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.581254  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.581478  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.586069  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.586848  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.587337  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.589694  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.589882  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.590091  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.590329  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.590632  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.590873  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.591192  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.591454  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.591703  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.591930  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.592170  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.592396  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.592653  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.595007  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.596319  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.597950  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.599296  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.609437  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.609653  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.609938  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.610687  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.621923  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.622153  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.622318  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.623299  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.648794  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.649142  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.664134  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.664382  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.669473  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.669734  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.672606  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.672832  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.673055  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.673255  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.673466  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.673666  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.676374  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.676550  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.681711  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.689419  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.692066  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.692279  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.694999  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.695231  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.695472  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.698184  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.709776  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.717428  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.720180  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.720382  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.720589  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.720793  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.731559  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.731812  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.731972  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.742295  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.750133  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.750379  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.750625  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.750792  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.778556  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.778743  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.778970  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.779207  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.789633  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.789860  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.790076  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.795363  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.795607  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.795882  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.796195  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.796408  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.796597  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.796849  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.797055  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.797358  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.797609  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.800431  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.808163  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.808368  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.808629  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.808866  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.811561  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.811780  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.828982  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.830327  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.832255  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.835093  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.835287  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.835476  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.836141  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.840837  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.841059  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.841271  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.841498  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.841699  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.841900  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.842151  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.842352  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.842628  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.842826  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.843127  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.843389  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.856551  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.856762  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.856982  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.857259  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.857627  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.858045  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.863099  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.863319  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.882516  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.882750  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.890803  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.891052  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.891294  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.892642  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.892895  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.893175  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.893406  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.893621  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.894102  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.894142  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.896976  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.897236  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.897420  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.900352  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.902928  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.903223  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.903735  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.904199  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.909134  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.909372  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.913172  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.920277  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.920501  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.920786  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.920931  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.921003  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.921264  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.921265  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.921586  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.943581  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.943809  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.944059  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.944257  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.944458  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.944726  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.944954  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.945147  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.959767  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.960093  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.960309  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.960552  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.960808  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.961053  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.961282  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.961503  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.961782  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.962018  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.962278  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.962548  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.962779  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.963052  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.969125  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.969360  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.969631  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.969812  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.969980  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.970170  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.970402  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.970584  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.970847  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.976580  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.977794  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.981958  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.984225  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.984476  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.998291  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.998525  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.998750  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.999001  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.004917  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.005141  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.021842  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.023211  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.028899  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.028923  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.029189  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.029403  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.029655  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.029892  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.030141  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.030356  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.030586  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.030768  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.051952  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.052182  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.052386  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.052623  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.052874  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.053148  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.053403  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.054157  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.056090  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.056403  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.056617  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.056838  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.060328  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.068521  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.071317  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.071501  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.071699  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.077070  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.080605  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.080899  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.081156  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.081415  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.081660  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.081939  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.082155  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.082402  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.082623  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.082731  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.082814  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.083032  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.085912  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.086201  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.096980  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.098359  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.099546  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.102448  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.102720  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.102989  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.103223  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.103464  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.104277  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.108438  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.108711  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.108911  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.109165  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.109467  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.109733  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.110004  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.110239  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.110486  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.110714  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.110724  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.114183  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.122400  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.122675  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.122920  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.123195  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.123414  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.123673  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.123876  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.126001  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.126210  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.126441  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.126644  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.126863  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.127283  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.131070  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.131325  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.131569  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.131783  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.132396  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.134912  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.135169  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.135370  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.136338  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.140527  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.191358  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.191627  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.191875  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.192153  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.192342  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.197578  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.202698  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.202907  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.208132  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.208374  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.211109  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.211313  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.211566  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.214330  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.214540  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.214802  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.215395  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.216657  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.220089  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.220728  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.225381  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.227144  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.228049  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.228245  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.231040  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.231236  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.234068  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.240044  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.240254  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.240460  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.241421  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.244509  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.250010  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.252700  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.252844  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.252897  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.253087  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.253091  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.253288  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.253509  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.253962  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.263445  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.263675  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.269212  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.270474  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.273865  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.276260  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.279080  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.279291  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.279481  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.279681  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.282605  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.295027  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.295212  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.295395  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.298802  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.299988  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.300223  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.300441  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.300705  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.304690  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.307312  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.307541  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.308730  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.310832  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.313612  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.318708  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.318970  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.319183  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.319437  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.319680  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.319883  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.320147  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.320412  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.325534  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.325740  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.325922  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.326180  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.328954  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.329154  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.329488  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.341765  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.341984  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.347205  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.347421  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.347626  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.347848  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.348078  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.348287  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.348484  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.348680  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.350585  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.350865  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.356657  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.356881  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.357982  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.388767  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.389058  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.389303  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.399434  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.399714  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.409890  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.410177  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.410370  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.410568  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.415786  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.416013  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.416213  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.416402  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.421667  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.421880  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.434330  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.434568  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.444756  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.444969  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.445178  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.445397  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.445652  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.450814  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.451030  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.451257  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.451485  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.471215  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.471483  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.476500  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.476737  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.476964  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.477159  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.477352  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.477542  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.477756  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.477946  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.498270  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.498482  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.518587  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.518834  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.519048  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.519256  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.519551  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.519752  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.520037  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.520202  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.522962  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.523137  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.523340  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.523535  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.523741  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.523979  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.524163  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.524384  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.536993  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.542232  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.542461  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.542663  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.542858  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.543066  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.543337  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.543718  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.551376  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.551574  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.552415  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.552659  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.552764  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.552887  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.553094  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.595072  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.595356  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.595624  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.595870  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.596087  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.596268  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.596544  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.596803  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.597059  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.597254  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.614900  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.615161  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.615470  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.615666  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.615878  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.616151  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.616405  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.653784  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.654006  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.654214  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.654405  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.654635  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.654825  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.655045  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.655264  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.655457  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.655639  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.655900  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.656116  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.656322  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.656612  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.656821  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.704331  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.704610  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.704798  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.705024  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.705224  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.705487  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.705684  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.705888  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.706089  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.706321  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.706543  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.706796  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.707023  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.707245  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.707461  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.707670  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.707892  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.708086  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.708282  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.728194  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.728391  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.728615  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.728813  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.729079  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.729300  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.729538  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.729822  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.737394  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.737599  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.738767  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.451416  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.452901  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.467027  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.467362  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.468367  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.468600  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.468681  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.469802  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.473881  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.484158  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.484486  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.484820  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.485107  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.485355  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.485599  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.487003  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.490974  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.493090  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.493911  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.495500  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.501397  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.502763  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.503086  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.503175  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.503513  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.504180  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.504545  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.508555  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.515570  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.516867  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.518400  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.518653  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.518872  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.519848  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.521056  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.527827  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.528055  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.529295  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.544860  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.545089  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.545290  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.545504  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.546707  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.295120  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.296660  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.302729  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.316711  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.318199  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.323024  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.333889  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.334213  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.334910  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.335601  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.336284  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.340574  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.340840  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.342151  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.347932  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.348773  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.349017  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.350518  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.357188  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.358728  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.360765  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.362203  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.363308  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.366835  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.371485  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.371713  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.373103  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.377363  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.377615  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.377725  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.378958  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.379127  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.382095  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.382319  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.383715  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.383874  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.384758  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.388401  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.392554  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.394655  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.396057  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.398231  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.398520  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.399797  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.400022  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.402158  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.402491  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.402809  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.403522  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.404407  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.409164  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.413456  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.414511  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.416069  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.420145  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.421628  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.421854  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.422119  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.423108  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.427431  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.440521  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.440709  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.440948  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.441156  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.441833  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.447186  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.448485  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.392441  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.393929  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.404150  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.404464  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.405719  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.603718  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.605327  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.611670  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.613114  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.138854  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.139194  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.140391  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.148223  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.149956  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.155593  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.157107  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.663831  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.665465  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.672602  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.675256  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.675565  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.676985  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.683609  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.684000  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.685106  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.687062  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.688648  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.689873  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.693089  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.697635  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.697898  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.699429  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.700985  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.703440  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.703691  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.704947  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.707213  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.710541  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.712060  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.718225  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.719748  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.721971  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.723291  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.725860  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.727556  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.461587  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.463192  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.469041  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.485112  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.485399  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.486702  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.491591  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.499243  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.500805  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.507115  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.510190  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.510407  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.511715  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.517388  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.183045  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.184537  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.190621  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:47.592581  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:47.604241  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:48.578392  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:48.587992  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:48.597157  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:48.605240  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:49.361904  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:49.370812  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:49.376947  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:49.384886  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.134733  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.145203  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.152640  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.159878  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.185983  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.195700  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.204260  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.212427  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.024849  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.026432  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.036523  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.047621  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.048969  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.057524  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.825331  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.834313  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.843958  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.852244  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:54.591287  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:54.605118  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:55.477183  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:57.617564  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:57.677883  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:58.316752  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:58.327472  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:59.071226  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:06.744064  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:07.467173  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:09.220778  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:09.244503  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:10.161363  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:10.162961  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:10.231134  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:15.442806  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:18.016840  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:18.924312  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:28.097031  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:33.444605  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:35.220340  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:51.309719  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:51.309983  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:51.311292  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:51.317121  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:52.153962  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:59.389873  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:59.512159  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:23:07.393066  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:23:08.411403  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:23:09.137790  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:23:09.151702  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:23:10.920717  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-20260827-1325/logs/manual_mc_master.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-20260827-1325/logs/manual_mc_master.log
new file mode 100644
index 0000000..c6b9423
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-20260827-1325/logs/manual_mc_master.log
@@ -0,0 +1,413 @@
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+I0827 04:28:16.835302    99 master.cpp:1366] Master service started on port 50051, max_threads=16, enable_metric_reporting=1, metrics_port=9003, default_kv_lease_ttl=10000, default_kv_soft_pin_ttl=1800000, allow_evict_soft_pinned_objects=1, eviction_ratio=0.05, eviction_high_watermark_ratio=0.9, enable_ha=0, enable_offload=0, enable_kv_events=0, kv_events_bind_endpoint=, kv_events_backend_id=, offload_on_evict=0, offload_force_evict=0, offloading_queue_limit=50000, offload_cap_ratio=0.5, ha_backend_type=etcd, ha_backend_connstring=, etcd_endpoints=, client_ttl=10, rpc_thread_num=16, rpc_port=50051, rpc_address=0.0.0.0, rpc_interface=, rpc_conn_timeout_seconds=0, rpc_enable_tcp_no_delay=1, rpc protocol=tcp, cluster_id=mooncake_cluster, root_fs_dir=, global_file_segment_size=9223372036854775807, memory_allocator=offset, enable_http_metadata_server=0, http_metadata_server_port=8080, http_metadata_server_host=0.0.0.0, enable_metadata_cleanup_on_timeout=0, put_start_discard_timeout_sec=30, put_start_release_timeout_sec=600, max_total_finished_tasks=10000, max_total_pending_tasks=10000, max_total_processing_tasks=10000, pending_task_timeout_sec=300, processing_task_timeout_sec=300, enable_snapshot=0, enable_snapshot_restore=0, snapshot_interval_seconds=600, snapshot_backup_dir=, snapshot_object_store_type=, snapshot_catalog_store_type=, snapshot_retention_count=2, max_retry_attempts=10, enable_cxl=0, cxl_path=/dev/dax0.0, cxl_size=8589934592
+I0827 04:28:16.866606   108 master_service.cpp:1520] Task cleanup thread started
+I0827 04:28:16.867491    99 master_admin_service.cpp:311] Master admin server started on port 9003
+I0827 04:28:16.869802   111 master_admin_service.cpp:302] Master Admin Metrics: role=standby, state=starting, service_ready=false, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:28:26.870203   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:28:36.870568   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:28:46.870873   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:28:56.871232   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:29:06.871615   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:29:16.871969   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:29:26.872316   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:29:36.872687   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:29:46.873068   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:29:56.873467   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:30:06.873914   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:30:16.874214   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:30:26.874567   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:30:36.874953   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:30:46.875332   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:30:56.875741   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:31:06.876119   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:31:16.876497   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:31:26.876866   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:31:36.877233   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:31:46.877605   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:31:56.877948   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:32:06.878437   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:32:16.879210   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:32:26.879711   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:32:36.880080   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:32:46.880565   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:32:56.881004   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:33:06.881398   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:33:16.881786   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:33:26.882170   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:33:36.882550   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:33:46.882939   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:33:56.883315   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:34:06.883746   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:34:16.884127   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:34:26.884568   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:34:36.884953   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:34:46.885334   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:34:56.885706   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:35:06.886085   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:35:16.886464   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:35:26.886835   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:35:36.887218   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:35:46.887596   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:35:56.887964   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:36:06.888356   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:36:16.888742   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:36:26.889113   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:36:36.889495   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:36:46.889882   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:36:56.890261   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:37:06.890656   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:37:16.891040   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:37:26.891430   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:37:36.891836   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:37:46.892215   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:37:56.892597   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:38:06.892982   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:38:16.893360   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:38:26.893747   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:38:36.894131   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:38:46.894515   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:38:56.894920   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:39:06.895303   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:39:16.895701   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:39:26.896085   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:39:36.896471   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:39:46.896852   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:39:56.897240   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:40:06.897629   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:40:16.898005   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:40:26.898384   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:40:36.898761   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:40:46.899144   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:40:56.899530   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:41:06.899937   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:41:16.900316   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:41:26.900695   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:41:36.901089   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:41:46.901479   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:41:56.901856   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:42:06.902242   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:42:16.902666   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:42:26.903064   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:42:36.903537   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:42:46.903987   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:42:56.904397   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:43:06.904863   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:43:16.905256   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:43:26.905628   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:43:36.905993   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:43:46.906375   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:43:56.906759   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:44:06.907138   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:44:16.907518   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:44:26.907922   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:44:36.908319   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:44:46.908928   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:44:56.909319   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:45:06.909730   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:45:16.910125   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:45:26.910491   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:45:36.910866   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:45:46.911247   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:45:56.911623   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:46:06.911993   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:46:16.912426   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:46:26.912810   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:46:36.913184   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:46:46.913556   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:46:56.913941   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:47:06.914324   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:47:16.914712   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:47:26.915107   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:47:36.915493   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:47:46.915882   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:47:56.916265   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:48:06.916646   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:48:16.917042   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:48:26.917429   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:48:36.917809   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:48:46.918201   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:48:56.918605   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:49:06.918992   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:49:16.919385   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:49:26.919793   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:49:36.920181   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:49:46.920554   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:49:56.920938   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:50:06.921316   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:50:16.921710   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:50:26.922091   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:50:36.922484   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:50:46.922891   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:50:56.923311   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:51:06.923827   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:51:16.924309   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:51:26.924793   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:51:36.925184   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:51:46.925541   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:51:56.925873   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:52:06.926222   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:52:16.926549   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:52:26.926885   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:52:36.927227   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:52:46.927629   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:52:56.927992   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:53:06.928339   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:53:16.928666   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:53:26.928995   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:53:36.929322   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:53:46.929646   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:53:56.930056   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:54:06.930403   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:54:16.930779   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:54:26.931145   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:54:36.931521   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:54:46.931916   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:54:56.932343   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:55:06.932718   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:55:16.933064   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:55:26.933395   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:55:36.933718   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:55:46.934052   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:55:56.934386   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:56:06.934712   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:56:16.935036   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:56:26.935360   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:56:36.935707   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:56:46.936054   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:56:56.936385   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:57:06.936734   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:57:16.937136   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:57:26.937498   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:57:36.937844   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:57:46.938345   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:57:56.938614   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:58:06.938869   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:58:16.939126   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:58:26.939358   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:58:36.939590   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:58:46.939940   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:58:56.940317   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:59:06.940711   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:59:16.941143   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:59:26.941520   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:59:36.941883   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:59:46.942258   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:59:56.942601   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:00:06.942938   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:00:16.943261   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:00:26.943506   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:00:36.943737   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:00:46.943951   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:00:56.944247   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:01:06.944607   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:01:16.944896   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:01:26.945246   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:01:36.945606   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:01:46.945940   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:01:56.946230   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:02:06.946504   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:02:16.946818   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:02:26.947096   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:02:36.947449   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:02:46.947767   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:02:56.948025   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:03:06.948321   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:03:16.948577   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:03:26.948939   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:03:36.949285   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:03:46.949551   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:03:56.949826   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:04:06.950160   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:04:16.950445   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:04:26.950799   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:04:36.951154   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:04:46.951506   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:04:56.951869   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:05:06.952240   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:05:16.952581   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:05:26.952893   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:05:36.953207   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:05:46.953542   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:05:56.953862   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:06:06.954190   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:06:16.954527   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:06:26.954862   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:06:36.955159   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:06:46.955430   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:06:56.955727   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:07:06.956014   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:07:16.956292   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:07:26.956629   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:07:36.956969   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:07:46.957286   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:07:56.957538   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:08:06.957857   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:08:16.958175   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:08:26.958505   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:08:36.958873   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:08:46.959213   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:08:56.959580   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:09:06.959961   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:09:16.960343   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:09:26.960726   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:09:36.961073   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:09:46.961421   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:09:56.961745   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:10:06.962085   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:10:16.962410   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:10:26.962738   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:10:36.963074   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:10:46.963395   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:10:56.963728   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:11:06.964046   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:11:16.964359   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:11:26.964668   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:11:36.964977   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:11:46.965345   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:11:56.965719   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:12:06.966044   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:12:16.966377   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:12:26.966706   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:12:36.967024   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:12:46.967339   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:12:56.967662   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:13:06.967989   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:13:16.968293   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:13:26.968617   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:13:36.968954   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:13:46.969303   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:13:56.969642   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:14:06.969978   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:14:16.970424   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:14:26.970767   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:14:36.971103   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:14:46.971450   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:14:56.971846   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:15:06.972218   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:15:16.972595   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:15:26.972944   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:15:36.973273   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:15:46.973601   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:15:56.973927   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:16:06.974232   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:16:16.974531   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:16:26.974860   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:16:36.975173   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:16:46.975457   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:16:56.975767   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:17:06.976097   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:17:16.976425   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:17:26.976751   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:17:36.977074   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:17:46.977340   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:17:56.977588   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:18:06.977923   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:18:16.978257   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:18:26.978587   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:18:36.978917   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:18:46.979271   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:18:56.979615   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:19:06.979946   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:19:16.980322   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:19:26.980680   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:19:36.981024   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:19:46.981348   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:19:56.981689   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:20:06.982029   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:20:16.982355   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:20:26.982690   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:20:36.983021   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:20:46.983355   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:20:56.983701   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:21:06.984002   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:21:16.984310   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:21:26.984650   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:21:36.985025   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:21:46.985387   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:21:56.985735   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:22:06.986088   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:22:16.986429   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:22:26.986770   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:22:36.987112   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:22:46.987457   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:22:56.987808   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:23:06.988150   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:23:16.988442   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:23:26.988807   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:23:36.989144   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:23:46.989451   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:23:56.989902   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:24:06.990252   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:24:16.990571   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:24:26.990865   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:24:36.991190   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:24:46.991520   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:24:56.991888   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:25:06.992215   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:25:16.992558   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:25:26.993088   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:25:36.993425   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:25:46.993746   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:25:56.994050   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:26:06.994376   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:26:16.994699   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:26:26.995038   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:26:36.995332   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:26:46.995611   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:26:56.995985   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:27:06.996469   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:27:16.996830   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:27:26.997224   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:27:36.997588   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:27:46.997939   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:27:56.998252   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:28:06.998566   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:28:16.998903   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:28:26.999221   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:28:36.999548   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:28:46.999909   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:28:57.000319   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:29:07.000710   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:29:17.001116   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:29:27.001483   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:29:37.001886   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:29:47.002224   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:29:57.002625   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:30:07.002960   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:30:17.003293   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:30:27.003625   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:30:37.003929   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:30:47.004259   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:30:57.004590   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:31:07.004930   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:31:17.005257   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:31:27.005599   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:31:37.005923   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:31:47.006310   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:31:57.006702   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:32:07.007108   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:32:17.007465   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:32:27.007848   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:32:37.008191   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:32:47.008533   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:32:57.008893   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:33:07.009240   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:33:17.009582   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:33:27.009918   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:33:37.010253   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:33:47.010578   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:33:57.010911   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:34:07.011226   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:34:17.011554   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:34:27.011901   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:34:37.012240   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:34:47.012563   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:34:57.012892   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:35:07.013228   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:35:17.013548   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:35:27.013886   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:35:37.014219   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:35:47.014546   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:35:57.014894   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:36:07.015219   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:36:17.015552   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-20260827-1325/logs/manual_p_node0.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-20260827-1325/logs/manual_p_node0.log
new file mode 100644
index 0000000..fbc5247
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-20260827-1325/logs/manual_p_node0.log
@@ -0,0 +1,1352 @@
+/usr/local/lib/python3.12/dist-packages/torchao/quantization/quant_api.py:1731: SyntaxWarning: invalid escape sequence '\.'
+  """Configuration class for applying different quantization configs to modules or parameters based on their fully qualified names (FQNs).
+/sgl-workspace/sglang/python/sglang/launch_server.py:56: UserWarning: 'python -m sglang.launch_server' is still supported, but 'sglang serve' is the recommended entrypoint.
+  Example: sglang serve --model-path <model> [options]
+  warnings.warn(
+[2026-08-27 04:28:54] Breakable CUDA graph is incompatible with MLA attention; disabling prefill CUDA graph.
+[2026-08-27 04:28:55] Attention backend not specified. Use triton backend by default.
+[2026-08-27 04:28:55] Pipeline parallelism is incompatible with overlap schedule.
+[2026-08-27 04:28:58] server_args=ServerArgs(model_path='/data/hf_models/Kimi-K3', tokenizer_path='/data/hf_models/Kimi-K3', tokenizer_mode='auto', tokenizer_backend='huggingface', tokenizer_worker_num=1, detokenizer_worker_num=1, skip_tokenizer_init=False, load_format='auto', model_loader_extra_config='{}', trust_remote_code=True, context_length=None, is_embedding=False, enable_multimodal=None, revision=None, model_impl='auto', model_config_parser='auto', json_model_override_args='{}', dtype='auto', quantization=None, quantization_param_path=None, kv_cache_dtype='auto', enable_fp32_lm_head=False, modelopt_quant=None, modelopt_checkpoint_restore_path=None, modelopt_checkpoint_save_path=None, modelopt_export_path=None, quantize_and_serve=False, rl_quant_profile=None, enable_tf32_matmul=False, mem_fraction_static=0.88, max_running_requests=None, max_queued_requests=None, max_total_tokens=None, chunked_prefill_size=8192, enable_dynamic_chunking=False, max_prefill_tokens=16384, prefill_max_requests=None, schedule_policy='fcfs', enable_priority_scheduling=False, disable_priority_preemption=False, default_priority_value=None, abort_on_priority_when_disabled=False, schedule_low_priority_values_first=False, priority_scheduling_preemption_threshold=10, retraction_policy='length', schedule_conservativeness=1.0, page_size=64, swa_full_tokens_ratio=0.8, disable_hybrid_swa_memory=False, radix_eviction_policy='lru', prefill_only_disable_kv_cache=False, disable_radix_cache=True, enable_page_major_kv_layout=False, enable_unified_memory=False, disable_chunked_prefix_cache=False, disable_overlap_schedule=True, num_continuous_decode_steps=1, scheduler_recv_interval=1, enable_mixed_chunk=False, nccl_port=None, dist_timeout=3600, dist_init_addr='174.1.60.1:20000', nnodes=4, node_rank=0, tp_size=4, dcp_size=1, pp_size=8, pp_max_micro_batch_size=None, pp_async_batch_depth=0, dp_size=1, load_balance_method='follow_bootstrap_room', attn_cp_size=1, moe_dp_size=1, dcp_comm_backend='ag_rs', dcp_replicate_q_proj=None, dwdp_size=1, enable_prefill_cp=False, cp_strategy=None, enable_dsa_cache_layer_split=False, enable_dsa_prefill_context_parallel=False, dsa_prefill_cp_mode='round-robin-split', enable_prefill_context_parallel=False, prefill_cp_mode='in-seq-split', enable_cp_decode_attn_tp=False, enable_dp_attention=False, enable_dp_attention_local_control_broadcast=False, enable_dp_lm_head=False, enable_attn_tp_input_scattered=False, disable_attn_tp_gather=False, enable_p2p_check=False, device='cuda', base_gpu_id=0, gpu_id_step=1, random_seed=984607766, watchdog_timeout=300, soft_watchdog_timeout=None, sleep_on_idle=False, use_ray=False, custom_sigquit_handler=None, numa_node=None, gc_threshold=None, host='0.0.0.0', port=30000, fastapi_root_path='', smg_grpc_mode=False, grpc_mode=False, grpc_port=None, sidecar=None, sidecar_args=None, skip_server_warmup=False, warmups=None, enable_http2=False, ssl_keyfile=None, ssl_certfile=None, ssl_ca_certs=None, ssl_keyfile_password=None, enable_ssl_refresh=False, api_key=None, admin_api_key=None, served_model_name='kimi-k3', weight_version='default', chat_template=None, hf_chat_template_name=None, completion_template=None, file_storage_path='sglang_storage', enable_cache_report=False, reasoning_parser=None, default_chat_template_kwargs=None, strip_thinking_cache=False, enable_strict_thinking=False, tool_call_parser=None, tool_server=None, sampling_defaults='model', asr_max_buffer_seconds=60, asr_max_concurrent_sessions=32, preferred_sampling_params=None, allow_auto_truncate=False, stream_interval=1, batch_notify_size=16, stream_response_default_include_usage=False, incremental_streaming_output=False, enable_streaming_session=False, enable_session_radix_cache=False, log_level='info', log_level_http=None, log_requests=False, log_requests_level=2, log_requests_format='text', log_requests_target=None, uvicorn_access_log_exclude_prefixes=[], crash_dump_folder=None, show_time_cost=False, enable_metrics=False, smg_http_sidecar_port=None, enable_mfu_metrics=False, enable_metrics_for_all_schedulers=False, load_snapshot_publish_interval=15, tokenizer_metrics_custom_labels_header='x-custom-labels', tokenizer_metrics_allowed_custom_labels=None, extra_metric_labels=None, bucket_time_to_first_token=None, bucket_inter_token_latency=None, bucket_e2e_request_latency=None, prompt_tokens_buckets=None, generation_tokens_buckets=None, gc_warning_threshold_secs=0.0, decode_log_interval=40, enable_request_time_stats_logging=False, kv_events_config=None, enable_forward_pass_metrics=False, forward_pass_metrics_worker_id='', forward_pass_metrics_ipc_name=None, enable_trace=False, trace_modules='request', otlp_traces_endpoint='localhost:4317', export_metrics_to_file=False, export_metrics_to_file_dir=None, stat_loggers=None, constrained_json_whitespace_pattern=None, constrained_json_disable_any_whitespace=False, attention_backend='triton', decode_attention_backend=None, prefill_attention_backend=None, sampling_backend='flashinfer', grammar_backend='xgrammar', radix_cache_backend=None, mm_attention_backend=None, fp8_gemm_runner_backend='auto', fp4_gemm_runner_backend='auto', bf16_gemm_backend='auto', dsa_prefill_backend=None, dsa_decode_backend=None, dsa_paged_mqa_logits_backend='auto', dsa_topk_backend='sgl-kernel', disable_flashinfer_autotune=False, flashinfer_autotune_skip_ops=None, mamba_backend='triton', cuda_graph_config=CudaGraphConfig(decode=PhaseConfig(backend='disabled', max_bs=16, bs=[1, 2, 4, 8, 12, 16], tc_compiler='eager', full_prefill_max_req=None), prefill=PhaseConfig(backend='disabled', max_bs=2048, bs=[4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], tc_compiler='eager', full_prefill_max_req=None)), cuda_graph_backend_decode=None, cuda_graph_backend_prefill=None, cuda_graph_max_bs_decode=16, cuda_graph_max_bs_prefill=None, cuda_graph_bs_decode=None, cuda_graph_bs_prefill=None, cuda_graph_tc_compiler=None, disable_prefill_cuda_graph=False, disable_decode_cuda_graph=False, disable_cuda_graph=False, disable_cuda_graph_padding=False, enable_profile_cuda_graph=False, enable_cudagraph_gc=False, debug_cuda_graph=False, enable_layerwise_nvtx_marker=False, enable_nccl_nvls=False, enable_symm_mem=False, triton_attention_reduce_in_fp32=False, triton_attention_num_kv_splits=8, triton_attention_split_tile_size=None, flashinfer_mla_disable_ragged=False, enable_fused_qk_norm_rope=False, enable_precise_embedding_interpolation=False, enable_fused_moe_sum_all_reduce=False, enable_deepseek_v4_fp4_indexer=False, disable_custom_all_reduce=False, enable_mscclpp=False, enable_torch_symm_mem=False, enable_scattered_sconv=False, pre_warm_nccl=False, enable_quant_communications=False, enable_flashinfer_allreduce_fusion=False, enforce_disable_flashinfer_allreduce_fusion=False, flashinfer_allreduce_fusion_backend=None, enable_aiter_allreduce_fusion=False, enable_torch_compile=False, enable_torch_compile_debug_mode=False, torch_compile_max_bs=32, torchao_config='', speculative_algorithm=None, speculative_draft_model_path=None, speculative_draft_model_revision=None, speculative_draft_load_format=None, speculative_num_steps=None, speculative_eagle_topk=None, speculative_num_draft_tokens=None, speculative_dflash_block_size=None, speculative_dspark_block_size=None, speculative_dspark_sps_table_path=None, speculative_dspark_confidence_sts_path=None, speculative_dspark_align_verify_tokens_to_graph_tier=False, speculative_accept_threshold_single=1.0, speculative_accept_threshold_acc=1.0, speculative_use_rejection_sampling=False, speculative_token_map=None, speculative_attention_mode='prefill', speculative_draft_attention_backend=None, speculative_draft_window_size=None, speculative_moe_runner_backend='flashinfer_mxfp4', speculative_moe_a2a_backend=None, speculative_draft_model_quantization=None, speculative_skip_dp_mlp_sync=False, enable_multi_layer_eagle=False, speculative_adaptive=False, speculative_adaptive_config=None, decoupled_spec_bind_endpoint=None, decoupled_spec_connect_endpoints=None, decoupled_spec_rank=None, decoupled_spec_role='null', spec_trace_dir=None, speculative_ngram_min_bfs_breadth=1, speculative_ngram_max_bfs_breadth=10, speculative_ngram_match_type='BFS', speculative_ngram_max_trie_depth=18, speculative_ngram_capacity=10000000, speculative_ngram_external_corpus_path=None, speculative_ngram_external_sam_budget=0, speculative_ngram_external_corpus_max_tokens=10000000, ep_size=4, moe_a2a_backend='none', moe_runner_backend='flashinfer_mxfp4', flashinfer_mxfp4_moe_precision='default', deepep_mode='auto', fuseep_mode=2, deepep_dispatcher_output_dtype='auto', ep_num_redundant_experts=0, ep_dispatch_algorithm=None, init_expert_location='trivial', enable_eplb=False, eplb_algorithm='auto', eplb_rebalance_num_iterations=1000, eplb_rebalance_layers_per_chunk=None, eplb_min_rebalancing_utilization_threshold=1.0, expert_distribution_recorder_mode=None, expert_distribution_recorder_buffer_size=1000, enable_expert_distribution_metrics=False, deepep_config=None, moe_dense_tp_size=None, elastic_ep_backend=None, enable_elastic_expert_backup=False, mooncake_ib_device=None, enable_waterfill=False, ep_join_mode=None, ep_join_rank_offset=0, elastic_ep_initial_size=None, max_ep_size=None, elastic_ep_scale_timeout=600, elastic_ep_rejoin=False, disable_flashinfer_cutlass_moe_fp4_allgather=False, disable_shared_experts_fusion=False, enforce_shared_experts_fusion=False, max_mamba_cache_size=None, mamba_ssm_dtype=None, mamba_max_states_per_path=-1, enable_mamba_cache_stochastic_rounding=False, mamba_cache_philox_rounds=0, mamba_full_memory_ratio=0.36, mamba_radix_cache_strategy='extra_buffer_lazy', uses_mamba_radix_cache=False, mamba_track_interval=256, enable_int8_mamba_checkpoint=False, int8_mamba_ckpt_size=None, linear_attn_backend='triton', linear_attn_decode_backend=None, linear_attn_prefill_backend=None, linear_attn_verify_backend=None, enable_linear_replayssm=False, linear_replayssm_cache_len=16, enable_linear_replayssm_spec=False, enable_hierarchical_cache=False, hicache_ratio=2.0, hicache_size=0, hicache_write_policy='write_through', hicache_io_backend='kernel', hicache_mem_layout='page_first', hicache_storage_backend=None, hicache_storage_prefetch_policy='timeout', hicache_storage_backend_extra_config=None, enable_hisparse=False, hisparse_config=None, enable_broadcast_mm_inputs_process=False, enable_prefix_mm_cache=False, mm_enable_dp_encoder=False, mm_process_config={}, mm_processor_worker_num=0, mm_io_worker_num=0, limit_mm_data_per_request=None, enable_mm_global_cache=False, disable_fast_image_processor=False, mm_feature_transport='cpu', keep_mm_feature_on_device=False, enable_lora=None, enable_lora_overlap_loading=None, max_lora_rank=None, lora_target_modules=None, lora_paths=None, max_loaded_loras=None, max_loras_per_batch=8, lora_eviction_policy='lru', lora_backend='csgmv', max_lora_chunk_size=16, experts_shared_outer_loras=None, lora_use_virtual_experts=False, lora_strict_loading=False, lora_drain_wait_threshold=0.0, enable_two_batch_overlap=False, enable_single_batch_overlap=False, tbo_token_distribution_threshold=0.48, cpu_offload_gb=0, offload_group_size=-1, offload_num_in_group=1, offload_prefetch_step=1, offload_mode='cpu', enable_lmcache=False, lmcache_config_file=None, enable_flexkv=False, flexkv_config_file=None, kt_weight_path=None, kt_method='AMXINT4', kt_cpuinfer=None, kt_threadpool_count=2, kt_num_gpu_experts=None, kt_max_deferred_experts_per_token=None, dllm_algorithm=None, dllm_algorithm_config=None, dllm_fdfo=True, disaggregation_mode='prefill', disaggregation_transfer_backend='mooncake', disaggregation_bootstrap_port=28800, disaggregation_ib_device='mlx5_0,mlx5_1,mlx5_2,mlx5_3', disaggregation_decode_enable_radix_cache=False, disaggregation_decode_enable_offload_kvcache=False, num_reserved_decode_tokens=512, disaggregation_decode_extra_slots=None, disaggregation_decode_polling_interval=1, optimistic_prefill_attempts=0, encoder_only=False, language_only=False, encoder_transfer_backend='zmq_to_tokenizer', encoder_urls=[], encoder_bootstrap_port=8997, encoder_register_urls=[], enable_adaptive_dispatch_to_encoder=False, enable_pdmux=False, pdmux_config_path=None, sm_group_num=8, custom_weight_loader=[], weight_loader_disable_mmap=False, weight_loader_prefetch_checkpoints=False, weight_loader_prefetch_num_threads=4, weight_loader_drop_cache_after_load=False, remote_instance_weight_loader_seed_instance_ip=None, remote_instance_weight_loader_seed_instance_service_port=None, remote_instance_weight_loader_send_weights_group_ports=None, remote_instance_weight_loader_backend='nccl', remote_instance_weight_loader_start_seed_via_transfer_engine=False, engine_info_bootstrap_port=6789, modelexpress_config=None, download_dir=None, model_checksum=None, delete_ckpt_after_loading=False, decrypted_config_file=None, decrypted_draft_config_file=None, checkpoint_engine_wait_weights_before_ready=False, enable_prefill_delayer=False, prefill_delayer_max_delay_passes=30, prefill_delayer_token_usage_low_watermark=None, prefill_delayer_forward_passes_buckets=None, prefill_delayer_wait_seconds_buckets=None, prefill_delayer_queue_min_ratio=None, prefill_delayer_max_delay_ms=None, min_free_slots_delay=None, enable_deterministic_inference=False, rl_on_policy_target=None, kv_canary='none', kv_canary_real_data='none', kv_canary_sweep_interval=0, enable_dynamic_batch_tokenizer=False, dynamic_batch_tokenizer_batch_size=32, dynamic_batch_tokenizer_batch_timeout=0.002, enable_tokenizer_batch_encode=False, disable_tokenizer_batch_decode=False, debug_tensor_dump_output_folder=None, debug_tensor_dump_layers=None, debug_tensor_dump_input_file=None, enable_memory_saver=False, enable_weights_cpu_backup=False, enable_draft_weights_cpu_backup=False, enable_custom_logit_processor=False, enable_return_hidden_states=False, enable_return_routed_experts=False, enable_return_indexer_topk=False, disable_outlines_disk_cache=False, enable_mis=False, weight_cache_mode='off', weight_cache_socket=None, weight_cache_timeout=1800, forward_hooks=None, msprobe_dump_config=None)
+[2026-08-27 04:28:58] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:58] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:58] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:58] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:58] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:58] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:58] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:58] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:01] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:01] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:01] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:03] Multimodal processor concurrency enabled with 2 isolated worker threads (auto).
+[2026-08-27 04:29:03] CommonKVBootstrapServer started successfully on 0.0.0.0:28800
+[2026-08-27 04:29:03] No HuggingFace chat template found
+[2026-08-27 04:29:03] No chat template found, defaulting to 'string' content format
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:15 PP0 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:15 PP0 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:16] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:16] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:16 PP1 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:16 PP1 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+/sgl-workspace/sglang/python/sglang/srt/utils/hf_transformers/tokenizer.py:423: UserWarning: Using a slow tokenizer. This might cause a significant slowdown. Consider using a fast tokenizer instead.
+  warnings.warn(
+[2026-08-27 04:29:16] Applying special tokens cache patch for Kimi tokenizer: <class 'transformers_modules.Kimi_hyphen_K3.caf40b2773f9029e.tokenization_kimi.TikTokenTokenizer'>
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:16 PP1 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:16 PP1 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:16 PP0 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:16 PP0 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:16 PP1 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:16 PP1 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:17 PP0 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:17 PP0 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:17 PP1 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:17 PP1 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:17 PP0 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:17 PP0 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:17 PP0 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:17.587481   237 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:17.587555   237 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:17.587566   237 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.1 port: 12001
+I0827 04:29:17.587602   237 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.1:16830
+I0827 04:29:17.587723   237 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:17.613691   237 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:17.620473   237 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:17.627439   237 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:17.653980   237 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:17.656663   237 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:17.656692   237 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:17.656746   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:17.663833   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:17.665359   237 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:01
+I0827 04:29:17.665387   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:17.690173   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:17.691654   237 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:01
+I0827 04:29:17.691689   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:17.718215   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:17.719650   237 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:01
+I0827 04:29:17.719686   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:17.746297   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:17.747779   237 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:01
+I0827 04:29:17.747891   237 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:17 PP0 TP1 EP1] Init torch distributed begin.
+[2026-08-27 04:29:17 PP1 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:18.219898   241 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:18.219964   241 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:18.219973   241 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.1 port: 12001
+I0827 04:29:18.220007   241 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.1:15425
+I0827 04:29:18.220109   241 transfer_engine_impl.cpp:258] Auto-discovering topology...
+[2026-08-27 04:29:18 PP1 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+I0827 04:29:18.244532   241 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:18.252967   241 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:18.261787   241 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:18.289944   241 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:18.293610   241 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:18.293633   241 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:18.293681   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.301677   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:18.303126   241 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:01
+I0827 04:29:18.303155   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.330181   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:18.331590   241 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:01
+I0827 04:29:18.331614   241 rdma_context.cpp:205] Using SIEVE endpoint store
+[2026-08-27 04:29:18 PP0 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+I0827 04:29:18.358213   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:18.359581   241 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:01
+I0827 04:29:18.359604   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.386204   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:18.387562   241 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:01
+I0827 04:29:18.387681   241 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:18 PP1 TP1 EP1] Init torch distributed begin.
+[2026-08-27 04:29:18 PP1 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:18.588883   242 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:18.588970   242 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:18.588984   242 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.1 port: 12001
+I0827 04:29:18.589022   242 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.1:15650
+I0827 04:29:18.589143   242 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:18.594419   242 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:18.599439   242 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:18.604553   242 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:18.609825   242 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:18.612036   242 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:18.612056   242 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:18.612097   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.618021   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:18.619308   242 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:01
+I0827 04:29:18.619333   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.646080   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:18.647465   242 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:01
+I0827 04:29:18.647495   242 rdma_context.cpp:205] Using SIEVE endpoint store
+[2026-08-27 04:29:18 PP0 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+I0827 04:29:18.674199   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:18.675598   242 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:01
+I0827 04:29:18.675621   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.682866   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:18.684276   242 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:01
+I0827 04:29:18.684391   242 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:18 PP1 TP2 EP2] Init torch distributed begin.
+[2026-08-27 04:29:18 PP1 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:18.695660   238 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:18.695755   238 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:18.695767   238 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.1 port: 12001
+I0827 04:29:18.695806   238 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.1:15123
+I0827 04:29:18.695909   238 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:18.701093   238 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:18.706161   238 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:18.711323   238 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:18.737910   238 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:18.740556   238 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:18.740602   238 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:18.740651   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.747717   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:18.749193   238 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:01
+I0827 04:29:18.749220   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.774145   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:18.775534   238 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:01
+I0827 04:29:18.775558   238 rdma_context.cpp:205] Using SIEVE endpoint store
+[2026-08-27 04:29:18 PP0 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+I0827 04:29:18.802196   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:18.803623   238 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:01
+I0827 04:29:18.803649   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.810976   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:18.812341   238 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:01
+I0827 04:29:18.812443   238 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:18 PP0 TP2 EP2] Init torch distributed begin.
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:19.084213   243 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:19.084332   243 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:19.084348   243 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.1 port: 12001
+I0827 04:29:19.084400   243 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.1:15411
+I0827 04:29:19.084602   243 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:19.114387   243 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:19.121773   243 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:19.146405   243 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:19.153759   243 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:19.156515   243 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:19.156548   243 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:19.156603   243 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.163982   243 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:19.165694   243 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:01
+I0827 04:29:19.165726   243 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.173307   243 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:19.174904   243 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:01
+I0827 04:29:19.174932   243 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.206454   243 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:19.208142   243 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:01
+I0827 04:29:19.208168   243 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.215909   243 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:19.217583   243 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:01
+I0827 04:29:19.217706   243 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:19 PP1 TP3 EP3] Init torch distributed begin.
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:19.239588   236 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:19.239699   236 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:19.239715   236 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.1 port: 12001
+I0827 04:29:19.239763   236 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.1:15605
+I0827 04:29:19.239933   236 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:19.240546   240 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:19.240644   240 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:19.240658   240 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.1 port: 12001
+I0827 04:29:19.240702   240 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.1:15487
+I0827 04:29:19.240862   240 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:19.242301   239 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:19.242408   239 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:19.242422   239 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.1 port: 12001
+I0827 04:29:19.242470   239 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.1:15552
+I0827 04:29:19.242636   239 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:19.246201   240 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:19.248451   239 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:19.251204   240 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:19.254226   239 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:19.256254   240 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:19.259852   239 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:19.261372   240 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:19.263411   240 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:19.263432   240 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:19.263473   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.265838   239 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:19.268059   239 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:19.268082   239 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:19.268126   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.269106   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:19.270298   236 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:19.270401   240 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:01
+I0827 04:29:19.270423   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.274480   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:19.275991   239 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:01
+I0827 04:29:19.276019   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.276877   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:19.278193   240 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:01
+I0827 04:29:19.278220   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.297820   236 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:19.302485   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:19.302623   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:19.303958   239 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:01
+I0827 04:29:19.303983   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.304121   240 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:01
+I0827 04:29:19.304147   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.311621   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:19.313341   240 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:01
+I0827 04:29:19.313478   240 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:19 PP1 TP0 EP0] Init torch distributed begin.
+I0827 04:29:19.322126   236 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:19.330281   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:19.331759   239 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:01
+I0827 04:29:19.331784   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.339274   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:19.340766   239 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:01
+I0827 04:29:19.340884   239 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:19 PP0 TP3 EP3] Init torch distributed begin.
+I0827 04:29:19.345958   236 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:19.348632   236 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:19.348659   236 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:19.348712   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.355962   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:19.417273   236 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:01
+I0827 04:29:19.417309   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.424894   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:19.426437   236 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:01
+I0827 04:29:19.426491   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.434231   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:19.435868   236 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:01
+I0827 04:29:19.435896   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.443578   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:19.445148   236 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:01
+I0827 04:29:19.445264   236 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:19 PP0 TP0 EP0] Init torch distributed begin.
+[2026-08-27 04:29:26 PP0 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:26 PP1 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:27 PP1 TP2 EP2] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP1 TP0 EP0] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP1 TP3 EP3] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP1 TP1 EP1] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP0 TP3 EP3] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP0 TP0 EP0] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP0 TP1 EP1] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP0 TP2 EP2] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP0 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:27 PP0 TP3 EP3] sglang is using nccl==2.28.9
+[2026-08-27 04:29:27 PP0 TP1 EP1] sglang is using nccl==2.28.9
+[2026-08-27 04:29:27 PP0 TP2 EP2] sglang is using nccl==2.28.9
+[2026-08-27 04:29:27 PP0 TP0 EP0] Init torch distributed ends. elapsed=8.22 s, mem usage=0.46 GB
+[2026-08-27 04:29:27 PP1 TP3 EP3] Init torch distributed ends. elapsed=8.45 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP1 TP1 EP1] Init torch distributed ends. elapsed=9.28 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP1 TP2 EP2] Init torch distributed ends. elapsed=8.98 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP1 TP0 EP0] Init torch distributed ends. elapsed=8.35 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP0 TP2 EP2] Init torch distributed ends. elapsed=8.85 s, mem usage=0.46 GB
+[2026-08-27 04:29:27 PP0 TP3 EP3] Init torch distributed ends. elapsed=8.32 s, mem usage=0.46 GB
+[2026-08-27 04:29:27 PP0 TP1 EP1] Init torch distributed ends. elapsed=9.92 s, mem usage=0.46 GB
+[2026-08-27 04:29:27 PP1 TP0 EP0] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP1 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP0 TP0 EP0] Load weight begin. avail mem=82.12 GB
+[2026-08-27 04:29:27 PP1 TP3 EP3] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP1 TP2 EP2] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP0 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP1 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP1 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP1 TP1 EP1] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP0 TP2 EP2] Load weight begin. avail mem=82.12 GB
+[2026-08-27 04:29:27 PP1 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP0 TP3 EP3] Load weight begin. avail mem=82.12 GB
+[2026-08-27 04:29:27 PP0 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP0 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP1 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:27 PP1 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:27 PP0 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:27 PP1 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:27 PP1 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:27 PP1 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:27 PP0 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:27 PP1 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:27 PP1 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:27 PP1 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:27 PP0 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:27 PP0 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:27 PP0 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:27 PP0 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:27 PP0 TP1 EP1] Load weight begin. avail mem=82.12 GB
+[2026-08-27 04:29:27 PP0 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP0 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP0 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+
Multi-thread loading shards:   0% Completed | 0/96 [00:00<?, ?it/s]
Multi-thread loading shards:   1% Completed | 1/96 [00:00<01:01,  1.55it/s]
Multi-thread loading shards:   2% Completed | 2/96 [00:09<08:49,  5.63s/it]
Multi-thread loading shards:   3% Completed | 3/96 [00:12<06:35,  4.25s/it]
Multi-thread loading shards:   4% Completed | 4/96 [00:14<05:26,  3.55s/it]
Multi-thread loading shards:   5% Completed | 5/96 [00:17<04:43,  3.11s/it]
Multi-thread loading shards:   6% Completed | 6/96 [00:19<04:17,  2.86s/it]
Multi-thread loading shards:   7% Completed | 7/96 [00:21<04:00,  2.71s/it]
Multi-thread loading shards:   8% Completed | 8/96 [00:24<03:49,  2.61s/it]
Multi-thread loading shards:   9% Completed | 9/96 [00:26<03:44,  2.58s/it]
Multi-thread loading shards:  10% Completed | 10/96 [00:29<03:40,  2.56s/it]
Multi-thread loading shards:  11% Completed | 11/96 [00:31<03:36,  2.54s/it]
Multi-thread loading shards:  12% Completed | 12/96 [00:32<02:43,  1.95s/it]
Multi-thread loading shards:  14% Completed | 13/96 [00:32<02:05,  1.52s/it]
Multi-thread loading shards:  15% Completed | 14/96 [00:33<01:33,  1.14s/it]
Multi-thread loading shards:  16% Completed | 15/96 [00:33<01:21,  1.01s/it]
Multi-thread loading shards:  17% Completed | 16/96 [00:34<01:19,  1.00it/s]
Multi-thread loading shards:  18% Completed | 17/96 [00:35<01:02,  1.26it/s]
Multi-thread loading shards:  19% Completed | 18/96 [00:35<00:51,  1.53it/s]
Multi-thread loading shards:  20% Completed | 19/96 [00:35<00:42,  1.80it/s]
Multi-thread loading shards:  21% Completed | 20/96 [00:36<00:37,  2.01it/s]
Multi-thread loading shards:  22% Completed | 21/96 [00:36<00:33,  2.23it/s]
Multi-thread loading shards:  23% Completed | 22/96 [00:36<00:30,  2.43it/s]
Multi-thread loading shards:  24% Completed | 23/96 [00:37<00:27,  2.68it/s]
Multi-thread loading shards:  25% Completed | 24/96 [00:37<00:25,  2.80it/s]
Multi-thread loading shards:  26% Completed | 25/96 [00:37<00:24,  2.89it/s]
Multi-thread loading shards:  27% Completed | 26/96 [00:38<00:23,  2.99it/s]
Multi-thread loading shards:  28% Completed | 27/96 [00:38<00:22,  3.01it/s]
Multi-thread loading shards:  29% Completed | 28/96 [00:38<00:22,  2.99it/s]
Multi-thread loading shards:  30% Completed | 29/96 [00:39<00:22,  2.98it/s]
Multi-thread loading shards:  31% Completed | 30/96 [00:39<00:22,  2.98it/s]
Multi-thread loading shards:  32% Completed | 31/96 [00:39<00:21,  3.01it/s]
Multi-thread loading shards:  33% Completed | 32/96 [00:40<00:21,  3.00it/s]
Multi-thread loading shards:  34% Completed | 33/96 [00:41<00:35,  1.79it/s]
Multi-thread loading shards:  35% Completed | 34/96 [00:41<00:30,  2.02it/s]
Multi-thread loading shards:  36% Completed | 35/96 [00:41<00:27,  2.23it/s]
Multi-thread loading shards:  38% Completed | 36/96 [00:42<00:24,  2.41it/s]
Multi-thread loading shards:  39% Completed | 37/96 [00:42<00:22,  2.57it/s]
Multi-thread loading shards:  40% Completed | 38/96 [00:42<00:21,  2.64it/s]
Multi-thread loading shards:  41% Completed | 39/96 [00:43<00:20,  2.73it/s]
Multi-thread loading shards:  42% Completed | 40/96 [00:43<00:20,  2.80it/s]
Multi-thread loading shards:  43% Completed | 41/96 [00:43<00:19,  2.79it/s]
Multi-thread loading shards:  44% Completed | 42/96 [00:44<00:19,  2.80it/s]
Multi-thread loading shards:  45% Completed | 43/96 [00:44<00:22,  2.39it/s]
Multi-thread loading shards:  46% Completed | 44/96 [00:45<00:20,  2.51it/s]
Multi-thread loading shards:  47% Completed | 45/96 [00:45<00:19,  2.64it/s]
Multi-thread loading shards:  48% Completed | 46/96 [00:45<00:18,  2.67it/s]
Multi-thread loading shards:  49% Completed | 47/96 [00:46<00:18,  2.61it/s]
Multi-thread loading shards:  50% Completed | 48/96 [00:46<00:17,  2.72it/s]
Multi-thread loading shards:  51% Completed | 49/96 [00:47<00:16,  2.84it/s]
Multi-thread loading shards:  52% Completed | 50/96 [00:47<00:16,  2.81it/s]
Multi-thread loading shards:  53% Completed | 51/96 [00:47<00:15,  2.93it/s]
Multi-thread loading shards:  54% Completed | 52/96 [00:48<00:21,  2.05it/s]
Multi-thread loading shards:  55% Completed | 53/96 [00:49<00:33,  1.30it/s]
Multi-thread loading shards:  56% Completed | 54/96 [00:50<00:27,  1.55it/s]
Multi-thread loading shards:  57% Completed | 55/96 [00:50<00:22,  1.82it/s]
Multi-thread loading shards:  58% Completed | 56/96 [00:50<00:19,  2.06it/s]
Multi-thread loading shards:  59% Completed | 57/96 [00:51<00:17,  2.25it/s]
Multi-thread loading shards:  60% Completed | 58/96 [00:51<00:15,  2.40it/s]
Multi-thread loading shards:  61% Completed | 59/96 [00:51<00:14,  2.56it/s]
Multi-thread loading shards:  62% Completed | 60/96 [00:52<00:13,  2.67it/s]
Multi-thread loading shards:  64% Completed | 61/96 [00:52<00:12,  2.77it/s]
Multi-thread loading shards:  65% Completed | 62/96 [00:52<00:11,  2.84it/s]
Multi-thread loading shards:  66% Completed | 63/96 [00:53<00:11,  2.89it/s]
Multi-thread loading shards:  67% Completed | 64/96 [00:53<00:11,  2.87it/s]
Multi-thread loading shards:  68% Completed | 65/96 [00:54<00:10,  2.89it/s]
Multi-thread loading shards:  69% Completed | 66/96 [00:54<00:10,  2.91it/s]
Multi-thread loading shards:  70% Completed | 67/96 [00:54<00:10,  2.90it/s]
Multi-thread loading shards:  71% Completed | 68/96 [00:55<00:09,  2.90it/s]
Multi-thread loading shards:  72% Completed | 69/96 [00:55<00:09,  2.89it/s]
Multi-thread loading shards:  73% Completed | 70/96 [00:56<00:14,  1.85it/s]
Multi-thread loading shards:  74% Completed | 71/96 [00:56<00:11,  2.10it/s]
Multi-thread loading shards:  75% Completed | 72/96 [00:57<00:10,  2.23it/s]
Multi-thread loading shards:  76% Completed | 73/96 [00:57<00:09,  2.39it/s]
Multi-thread loading shards:  77% Completed | 74/96 [00:57<00:08,  2.48it/s]
Multi-thread loading shards:  78% Completed | 75/96 [00:58<00:09,  2.15it/s]
Multi-thread loading shards:  79% Completed | 76/96 [00:59<00:10,  1.89it/s]
Multi-thread loading shards:  80% Completed | 77/96 [00:59<00:09,  2.05it/s]
Multi-thread loading shards:  81% Completed | 78/96 [00:59<00:08,  2.24it/s]
Multi-thread loading shards:  82% Completed | 79/96 [01:00<00:07,  2.39it/s]
Multi-thread loading shards:  83% Completed | 80/96 [01:00<00:06,  2.51it/s]
Multi-thread loading shards:  84% Completed | 81/96 [01:00<00:05,  2.65it/s]
Multi-thread loading shards:  85% Completed | 82/96 [01:01<00:04,  2.84it/s]
Multi-thread loading shards:  86% Completed | 83/96 [01:02<00:06,  1.99it/s]
Multi-thread loading shards:  88% Completed | 84/96 [01:02<00:06,  1.99it/s]
Multi-thread loading shards:  90% Completed | 86/96 [01:02<00:03,  3.27it/s]
Multi-thread loading shards:  92% Completed | 88/96 [01:02<00:01,  4.72it/s]
Multi-thread loading shards:  94% Completed | 90/96 [01:02<00:00,  6.49it/s]
Multi-thread loading shards:  96% Completed | 92/96 [01:03<00:00,  8.37it/s]
Multi-thread loading shards:  98% Completed | 94/96 [01:03<00:00,  8.91it/s]
Multi-thread loading shards: 100% Completed | 96/96 [01:03<00:00,  9.35it/s]
Multi-thread loading shards: 100% Completed | 96/96 [01:03<00:00,  1.51it/s]
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:33 PP0 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:33 PP0 TP2 EP2] Load weight end. elapsed=65.63 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=39.03 GB, mem usage=43.09 GB.
+[2026-08-27 04:30:33 PP0 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:33 PP0 TP0 EP0] Load weight end. elapsed=65.81 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=39.03 GB, mem usage=43.09 GB.
+[2026-08-27 04:30:35 PP0 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:35 PP0 TP3 EP3] Load weight end. elapsed=67.37 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=39.03 GB, mem usage=43.09 GB.
+[2026-08-27 04:30:35 PP0 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:35 PP0 TP1 EP1] Load weight end. elapsed=67.46 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=39.03 GB, mem usage=43.09 GB.
+[2026-08-27 04:30:35 PP0 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:35 PP0 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:35 PP0 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:35 PP0 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:35 PP0 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:35 PP0 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:35 PP0 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:35 PP0 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:36 PP1 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:36 PP1 TP3 EP3] Load weight end. elapsed=68.32 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=36.12 GB, mem usage=45.92 GB.
+[2026-08-27 04:30:36 PP1 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP1 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP1 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP1 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP1 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP1 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP1 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP1 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:37 PP1 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:37 PP1 TP1 EP1] Load weight end. elapsed=69.55 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=36.12 GB, mem usage=45.92 GB.
+[2026-08-27 04:30:37 PP1 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:37 PP1 TP0 EP0] Load weight end. elapsed=69.66 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=36.12 GB, mem usage=45.92 GB.
+[2026-08-27 04:30:39 PP1 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:39 PP1 TP2 EP2] Load weight end. elapsed=71.33 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=36.12 GB, mem usage=45.92 GB.
+[2026-08-27 04:30:39 PP1 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:39 PP1 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:39 PP1 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:39 PP1 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:39 PP1 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:39 PP1 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:39 PP1 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:39 PP1 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:53 PP0 TP0 EP0] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:53 PP1 TP3 EP3] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:53 PP1 TP2 EP2] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:53 PP1 TP1 EP1] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:53 PP1 TP0 EP0] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:53 PP0 TP2 EP2] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:53 PP0 TP3 EP3] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:53 PP0 TP1 EP1] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:53 PP0 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:53 PP1 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:53 PP1 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:53 PP1 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:53 PP0 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:53 PP1 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:53 PP0 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:53 PP0 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:53 PP0 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 8.16 GB
+[2026-08-27 04:30:53 PP0 TP0 EP0] Memory pool end. avail mem=30.14 GB
+[2026-08-27 04:30:53 PP0 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 8.16 GB
+[2026-08-27 04:30:53 PP0 TP2 EP2] Memory pool end. avail mem=30.14 GB
+[2026-08-27 04:30:53 PP1 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:53 PP1 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:53 PP1 TP0 EP0] Memory pool end. avail mem=23.24 GB
+[2026-08-27 04:30:53 PP1 TP2 EP2] Memory pool end. avail mem=23.24 GB
+[2026-08-27 04:30:53 PP0 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 8.16 GB
+[2026-08-27 04:30:53 PP1 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:53 PP0 TP3 EP3] Memory pool end. avail mem=30.14 GB
+[2026-08-27 04:30:53 PP1 TP1 EP1] Memory pool end. avail mem=23.24 GB
+[2026-08-27 04:30:53 PP1 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:53 PP0 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 8.16 GB
+[2026-08-27 04:30:53 PP1 TP3 EP3] Memory pool end. avail mem=23.24 GB
+[2026-08-27 04:30:53 PP0 TP1 EP1] Memory pool end. avail mem=30.14 GB
+[2026-08-27 04:30:54 PP0 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:30:54 PP0 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:30:54 PP0 TP0 EP0] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp0_pp0_dp0.json
+[2026-08-27 04:30:54 PP1 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:30:54 PP1 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:30:54 PP0 TP2 EP2] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp2_pp0_dp0.json
+[2026-08-27 04:30:54 PP1 TP2 EP2] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp2_pp1_dp0.json
+[2026-08-27 04:30:54 PP1 TP0 EP0] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp0_pp1_dp0.json
+[2026-08-27 04:30:54 PP1 TP1 EP1] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp1_pp1_dp0.json
+[2026-08-27 04:30:54 PP1 TP3 EP3] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp3_pp1_dp0.json
+[2026-08-27 04:30:54 PP0 TP1 EP1] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp1_pp0_dp0.json
+[2026-08-27 04:30:54 PP0 TP3 EP3] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp3_pp0_dp0.json
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.46profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.32profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.51profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  6.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.09profile/s]
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.07profile/s]
+[TensorRT-LLM][INFO] Set logger level to INFO
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.48profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.11profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.29profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.37profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.16profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.82profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.18profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.58profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:01<00:02,  1.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.66profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.03profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.49profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:01<00:01,  2.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.25profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:01,  2.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:02<00:00,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:02<00:00,  3.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.20profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.01profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  9.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  9.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  9.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 12.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 12.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 11.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 11.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.78profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.16profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  3.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.75profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.50profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.92profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.68profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  3.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.97profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.87profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 10.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.51profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.74profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 11.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.30profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.70profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.72profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.63profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.53profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.29profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.58profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  8.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 10.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.71profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.62profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.43profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.18profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.29profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.37profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.53profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.37profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.35profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.00profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.71profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.60profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.78profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.79profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  8.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.86profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  8.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.16profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.78profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  3.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.15profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.94profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.93profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.25profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.23profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.97profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.79profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.96profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.64profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.12profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.75profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.12profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.25profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.91profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.05profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.66profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.90profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  9.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.73profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.55profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.49profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.13profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.74profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.17profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  3.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.72profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.57profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.59profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.64profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.83profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 13.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.49profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 12.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  9.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  7.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.32profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.93profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.29profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.72profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.75profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.46profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.51profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  3.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.90profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.16profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.72profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.63profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.83profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.70profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.59profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.55profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  6.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.93profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.33profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.75profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.30profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  6.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  7.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.79profile/s]
+[2026-08-27 04:31:21 PP0 TP1 EP1] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.53profile/s][2026-08-27 04:31:21 PP0 TP0 EP0] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.97profile/s]
+[2026-08-27 04:31:21 PP0 TP3 EP3] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.43profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.06profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s][2026-08-27 04:31:21 PP0 TP2 EP2] FlashInfer autotune completed.
+[2026-08-27 04:31:21 PP0 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:21 PP0 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:21 PP0 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:21 PP0 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:21 PP0 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:21 PP0 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:21 PP0 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:21 PP0 TP0 EP0] max_total_num_tokens=3801408, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=54, context_len=1048576, available_gpu_mem=27.35 GB
+[2026-08-27 04:31:21 PP0 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.86profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.58profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:01<00:07,  1.18s/profile]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  2.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.27profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.55profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  3.96profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.86profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  6.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  7.01profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.23profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.05profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  6.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  7.06profile/s]
+[2026-08-27 04:31:26 PP1 TP1 EP1] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.40profile/s]
+[2026-08-27 04:31:26 PP1 TP2 EP2] FlashInfer autotune completed.
+[2026-08-27 04:31:26 PP1 TP0 EP0] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.81profile/s]
+[2026-08-27 04:31:26 PP1 TP3 EP3] FlashInfer autotune completed.
+[2026-08-27 04:31:26 PP1 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:26 PP1 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:26 PP1 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:26 PP1 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:26 PP1 TP0 EP0] max_total_num_tokens=3801408, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=54, context_len=1048576, available_gpu_mem=20.45 GB
+[2026-08-27 04:31:26 PP1 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:26 PP1 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:26 PP1 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:26 PP1 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+W0827 04:31:30.805795  2607 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fe48940, len: 6912: Operation not permitted [1]
+W0827 04:31:30.805797  2608 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4f684400, len: 6912: Operation not permitted [1]
+W0827 04:31:30.805945  2609 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4f962700, len: 6912: Operation not permitted [1]
+W0827 04:31:30.805979  2610 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4f9e8a00, len: 6912: Operation not permitted [1]
+W0827 04:31:30.806130  2612 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fe5f640, len: 55296: Operation not permitted [1]
+W0827 04:31:30.806133  2611 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4f83f000, len: 55296: Operation not permitted [1]
+W0827 04:31:30.806264  2613 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fe328c0, len: 6912: Operation not permitted [1]
+W0827 04:31:30.806334  2614 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fa75cc0, len: 13824: Operation not permitted [1]
+W0827 04:31:30.806449  2615 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fa87f80, len: 6912: Operation not permitted [1]
+W0827 04:31:30.806493  2616 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fa29a80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.203712  2635 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50c65e80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.203859  2636 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50c6bb40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.203975  2637 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50d1ffc0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.204025  2638 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50a0c300, len: 6912: Operation not permitted [1]
+W0827 04:31:31.204075  2640 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50de1940, len: 55296: Operation not permitted [1]
+W0827 04:31:31.204082  2639 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50dc93c0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.204438  2641 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50c87940, len: 6912: Operation not permitted [1]
+W0827 04:31:31.204561  2642 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50cffb00, len: 13824: Operation not permitted [1]
+W0827 04:31:31.204641  2644 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50dc7380, len: 6912: Operation not permitted [1]
+W0827 04:31:31.204722  2643 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50c243c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.323861  2661 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50b13c40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.323863  2662 memory_location.cpp:73] Failed to get NUMA node, addr: 0x511191c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.323912  2663 memory_location.cpp:73] Failed to get NUMA node, addr: 0x511b37c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.324069  2664 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50b2da80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.324110  2666 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5115c980, len: 55296: Operation not permitted [1]
+W0827 04:31:31.324155  2665 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5116ac40, len: 55296: Operation not permitted [1]
+W0827 04:31:31.324479  2667 memory_location.cpp:73] Failed to get NUMA node, addr: 0x511a7e80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.324587  2668 memory_location.cpp:73] Failed to get NUMA node, addr: 0x511b8cc0, len: 13824: Operation not permitted [1]
+W0827 04:31:31.324621  2669 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50fc6800, len: 6912: Operation not permitted [1]
+W0827 04:31:31.324682  2670 memory_location.cpp:73] Failed to get NUMA node, addr: 0x511bc780, len: 6912: Operation not permitted [1]
+W0827 04:31:31.543082  2689 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4902e440, len: 6912: Operation not permitted [1]
+W0827 04:31:31.543203  2690 memory_location.cpp:73] Failed to get NUMA node, addr: 0x49484200, len: 6912: Operation not permitted [1]
+W0827 04:31:31.543314  2691 memory_location.cpp:73] Failed to get NUMA node, addr: 0x49119600, len: 6912: Operation not permitted [1]
+W0827 04:31:31.543375  2692 memory_location.cpp:73] Failed to get NUMA node, addr: 0x493dc780, len: 6912: Operation not permitted [1]
+W0827 04:31:31.543467  2693 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51ac5fc0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.543601  2694 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51a9e780, len: 55296: Operation not permitted [1]
+W0827 04:31:31.543751  2695 memory_location.cpp:73] Failed to get NUMA node, addr: 0x49149a40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.543912  2696 memory_location.cpp:73] Failed to get NUMA node, addr: 0x518c0840, len: 13824: Operation not permitted [1]
+W0827 04:31:31.543947  2697 memory_location.cpp:73] Failed to get NUMA node, addr: 0x49303540, len: 6912: Operation not permitted [1]
+W0827 04:31:31.543990  2698 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51951a80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681525  2717 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50376e40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681525  2718 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5058ee40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681525  2716 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5008b980, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681548  2715 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fe0e840, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681727  2719 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5040eb80, len: 55296: Operation not permitted [1]
+W0827 04:31:31.681779  2724 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50209ac0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681811  2720 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5041c3c0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.681828  2722 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5048cc40, len: 13824: Operation not permitted [1]
+W0827 04:31:31.681844  2723 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fde7a00, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681851  2721 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50386580, len: 6912: Operation not permitted [1]
+W0827 04:31:31.730942  2745 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fce4240, len: 6912: Operation not permitted [1]
+W0827 04:31:31.730918  2744 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fed7d40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.730950  2743 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fdcaf80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.731007  2748 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5074c400, len: 55296: Operation not permitted [1]
+W0827 04:31:31.731073  2747 memory_location.cpp:73] Failed to get NUMA node, addr: 0x507bd9c0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.731086  2746 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4f931480, len: 6912: Operation not permitted [1]
+W0827 04:31:31.731294  2749 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd958c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.731361  2751 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd31d80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.731361  2750 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4ffabe00, len: 13824: Operation not permitted [1]
+W0827 04:31:31.731566  2752 memory_location.cpp:73] Failed to get NUMA node, addr: 0x508527c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.808167  2778 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd3ce40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.808190  2777 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd20740, len: 6912: Operation not permitted [1]
+W0827 04:31:31.808192  2776 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4ff86c80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.808637  2782 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fe88780, len: 6912: Operation not permitted [1]
+W0827 04:31:31.808724  2780 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fe48c80, len: 55296: Operation not permitted [1]
+W0827 04:31:31.808760  2781 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fe564c0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.809056  2785 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4ffa7e40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.809049  2779 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4feaa740, len: 6912: Operation not permitted [1]
+W0827 04:31:31.809075  2784 memory_location.cpp:73] Failed to get NUMA node, addr: 0x502fabc0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.809239  2783 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4ffa1f80, len: 13824: Operation not permitted [1]
+W0827 04:31:31.851475  2807 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4ff6fcc0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.851596  2808 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fc0a280, len: 6912: Operation not permitted [1]
+W0827 04:31:31.851615  2809 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd7d700, len: 6912: Operation not permitted [1]
+W0827 04:31:31.851732  2810 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd761c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.851951  2811 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fde6cc0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.851970  2812 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4feeb000, len: 55296: Operation not permitted [1]
+W0827 04:31:31.851962  2813 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fbe2680, len: 6912: Operation not permitted [1]
+W0827 04:31:31.851987  2816 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fbf07c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.852048  2815 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4ffa9940, len: 6912: Operation not permitted [1]
+W0827 04:31:31.852046  2814 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd235c0, len: 13824: Operation not permitted [1]
+[2026-08-27 04:31:32] INFO:     Started server process [1]
+[2026-08-27 04:31:32] INFO:     Waiting for application startup.
+[2026-08-27 04:31:32] Calling super().encode with {'add_special_tokens': False}
+[2026-08-27 04:31:32] Calling super().encode with {'add_special_tokens': False}
+[2026-08-27 04:31:32] INFO:     Application startup complete.
+[2026-08-27 04:31:32] INFO:     Uvicorn running on http://0.0.0.0:30000 (Press CTRL+C to quit)
+[2026-08-27 04:31:33] INFO:     127.0.0.1:60348 - "GET /model_info HTTP/1.1" 200 OK
+[2026-08-27 04:31:33] Start of pd disaggregation warmup ...
+[2026-08-27 04:31:37] INFO:     174.1.60.1:33088 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:38] INFO:     174.1.60.1:33092 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:39] INFO:     174.1.60.1:33106 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:40] INFO:     174.1.60.1:33110 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:41] INFO:     174.1.60.1:42028 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:42] INFO:     174.1.60.1:42030 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:43] INFO:     174.1.60.1:42046 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:44] INFO:     174.1.60.1:42050 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:45] INFO:     174.1.60.1:42060 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:46] INFO:     174.1.60.1:42072 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:47] INFO:     174.1.60.1:42086 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:48] INFO:     174.1.60.1:42092 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:49] INFO:     174.1.60.1:42098 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:50] INFO:     174.1.60.1:42100 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:51] INFO:     174.1.60.1:46852 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:52] INFO:     174.1.60.1:46856 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:53] INFO:     174.1.60.1:46872 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:54] INFO:     174.1.60.1:46876 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:55] INFO:     174.1.60.1:46886 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:56] INFO:     174.1.60.1:46890 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:57] INFO:     174.1.60.1:46898 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:58] INFO:     174.1.60.1:46912 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:59] INFO:     174.1.60.1:46928 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:00] INFO:     174.1.60.1:46936 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:01] INFO:     174.1.60.1:57672 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:02] INFO:     174.1.60.1:57676 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:03] INFO:     174.1.60.1:57686 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:04] INFO:     174.1.60.1:57688 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:05] INFO:     174.1.60.1:57696 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:06] INFO:     174.1.60.1:57708 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:12] INFO:     174.1.60.1:39714 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:13] INFO:     174.1.60.1:39728 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:14] INFO:     174.1.60.1:39736 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:15] INFO:     174.1.60.1:39746 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:16] INFO:     174.1.60.1:39760 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:17] INFO:     174.1.60.1:39772 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:18] INFO:     174.1.60.1:39782 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:19] INFO:     174.1.60.1:39786 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:20] INFO:     174.1.60.1:39800 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:21] INFO:     174.1.60.1:41036 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:22] INFO:     174.1.60.1:41052 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:23] INFO:     174.1.60.1:41058 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:24] INFO:     174.1.60.1:41068 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:25] INFO:     174.1.60.1:41074 - "GET /health HTTP/1.1" 503 Service Unavailable
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank1]:[W827 04:32:26.019157362 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank0]:[W827 04:32:26.019963648 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank3]:[W827 04:32:26.020028119 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank2]:[W827 04:32:26.020057305 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank6]:[W827 04:32:26.045976535 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank7]:[W827 04:32:26.046176399 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank5]:[W827 04:32:26.048033943 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank4]:[W827 04:32:26.048331224 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[2026-08-27 04:32:26] INFO:     174.1.60.1:41078 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:27] INFO:     174.1.60.1:41090 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:28] INFO:     174.1.60.1:41102 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:29] INFO:     174.1.60.1:41104 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:30] INFO:     174.1.60.1:41120 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:31] INFO:     174.1.60.1:57002 - "GET /health HTTP/1.1" 503 Service Unavailable
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank6]:[W827 04:32:31.709756552 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank7]:[W827 04:32:31.709839299 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank5]:[W827 04:32:31.709991082 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank4]:[W827 04:32:31.710123265 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[2026-08-27 04:32:32] INFO:     174.1.60.1:57016 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:33] INFO:     174.1.60.1:57030 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:34] INFO:     174.1.60.1:57046 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:35] INFO:     174.1.60.1:57056 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:36] INFO:     174.1.60.1:57066 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:37] INFO:     174.1.60.1:57070 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:38] INFO:     174.1.60.1:57072 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:39] INFO:     174.1.60.1:57082 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:40] INFO:     174.1.60.1:57098 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:41] INFO:     174.1.60.1:56606 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:50] INFO:     174.1.60.1:56620 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:51] INFO:     174.1.60.1:45658 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:52] INFO:     174.1.60.1:45672 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:53] INFO:     174.1.60.1:45674 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:54] INFO:     174.1.60.1:45690 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:55] INFO:     174.1.60.1:45704 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:56] INFO:     174.1.60.1:45708 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:57] INFO:     174.1.60.1:45714 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:58] INFO:     174.1.60.1:45722 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:00] INFO:     174.1.60.1:45736 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:01] INFO:     174.1.60.1:45746 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:02] INFO:     174.1.60.1:33310 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:03] INFO:     174.1.60.1:33320 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:04] INFO:     174.1.60.1:33324 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:05] INFO:     174.1.60.1:33328 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:06] INFO:     174.1.60.1:33336 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:07] INFO:     174.1.60.1:33350 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:08] INFO:     174.1.60.1:33354 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:09] INFO:     174.1.60.1:33356 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:10] INFO:     174.1.60.1:33368 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:11] INFO:     174.1.60.1:39318 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:12] INFO:     174.1.60.1:39328 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:13] INFO:     174.1.60.1:39336 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:14] INFO:     174.1.60.1:39344 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:15] INFO:     174.1.60.1:39346 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:16] INFO:     174.1.60.1:39352 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:17] INFO:     174.1.60.1:39364 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:18] INFO:     174.1.60.1:39378 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:19] INFO:     174.1.60.1:39390 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:20] INFO:     174.1.60.1:39402 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:25] INFO:     174.1.60.1:37792 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:26] INFO:     174.1.60.1:37804 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:27] INFO:     174.1.60.1:37812 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:28] INFO:     174.1.60.1:37826 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:29] INFO:     174.1.60.1:37830 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:30] INFO:     174.1.60.1:37844 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:31] INFO:     174.1.60.1:44208 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:32] INFO:     174.1.60.1:44220 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:33] INFO:     174.1.60.1:44226 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:34] INFO:     174.1.60.1:44228 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:35] INFO:     174.1.60.1:44238 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:37] INFO:     174.1.60.1:44248 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:38] INFO:     174.1.60.1:44258 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:39] INFO:     174.1.60.1:44272 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:40] INFO:     174.1.60.1:44276 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:41] INFO:     174.1.60.1:44194 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:42] INFO:     174.1.60.1:44198 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:43] INFO:     174.1.60.1:44212 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:44] INFO:     174.1.60.1:44216 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:45] INFO:     174.1.60.1:44222 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:46] INFO:     174.1.60.1:44236 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:47] INFO:     174.1.60.1:44246 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:48] INFO:     174.1.60.1:44252 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:49] INFO:     174.1.60.1:44264 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:50] INFO:     174.1.60.1:44278 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:51] INFO:     174.1.60.1:35728 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:52] INFO:     174.1.60.1:35736 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:53] INFO:     174.1.60.1:35752 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:54] INFO:     174.1.60.1:35754 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:55] INFO:     174.1.60.1:35758 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:04] INFO:     174.1.60.1:44514 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:05] INFO:     174.1.60.1:44518 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:06] INFO:     174.1.60.1:44520 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:07] INFO:     174.1.60.1:44526 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:08] INFO:     174.1.60.1:44530 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:09] INFO:     174.1.60.1:44546 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:10] INFO:     174.1.60.1:44550 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:11] INFO:     174.1.60.1:45098 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:12] INFO:     127.0.0.1:59722 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:12] INFO:     174.1.60.1:45106 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:13] INFO:     174.1.60.1:45122 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:14] INFO:     174.1.60.1:45124 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:15] INFO:     174.1.60.1:45136 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:16] INFO:     174.1.60.1:45146 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:17] INFO:     174.1.60.1:45152 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:19] INFO:     174.1.60.1:45158 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:20] INFO:     174.1.60.1:45170 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:21] INFO:     174.1.60.1:45182 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:22] INFO:     174.1.60.1:39946 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:23] INFO:     174.1.60.1:39958 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:24] INFO:     174.1.60.1:39962 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:25] INFO:     174.1.60.1:39978 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:26] INFO:     174.1.60.1:39990 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:27] INFO:     174.1.60.1:39994 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:28] INFO:     174.1.60.1:39996 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:29] INFO:     174.1.60.1:40002 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:30] INFO:     174.1.60.1:40012 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:31] INFO:     174.1.60.1:46978 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:32] INFO:     174.1.60.1:46988 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:33] INFO:     174.1.60.1:46996 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:34] INFO:     174.1.60.1:47004 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:39] INFO:     174.1.60.1:47020 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:40] INFO:     174.1.60.1:47022 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:41] INFO:     174.1.60.1:59158 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:42] INFO:     174.1.60.1:59170 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:43] INFO:     174.1.60.1:59172 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:44] INFO:     174.1.60.1:59180 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:45] INFO:     174.1.60.1:59184 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:46] INFO:     174.1.60.1:59186 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:47] INFO:     174.1.60.1:59200 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:48] INFO:     174.1.60.1:59202 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:49] INFO:     174.1.60.1:59214 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:50] INFO:     174.1.60.1:59220 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:52] INFO:     174.1.60.1:40566 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:53] INFO:     174.1.60.1:40576 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:54] INFO:     174.1.60.1:40588 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:55] INFO:     174.1.60.1:40596 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:56] INFO:     174.1.60.1:40598 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:57] INFO:     174.1.60.1:40604 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:58] INFO:     174.1.60.1:40608 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:59] INFO:     174.1.60.1:40610 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:00] INFO:     174.1.60.1:40622 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:01] INFO:     174.1.60.1:41102 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:02] INFO:     174.1.60.1:41108 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:03] INFO:     174.1.60.1:41122 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:04] INFO:     174.1.60.1:41124 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:05] INFO:     174.1.60.1:41134 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:06] INFO:     174.1.60.1:41140 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:07] INFO:     174.1.60.1:41142 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:08] INFO:     174.1.60.1:41158 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:09] INFO:     174.1.60.1:41160 - "GET /health HTTP/1.1" 503 Service Unavailable
+[rank3]:[W827 04:35:10.359210278 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank2]:[W827 04:35:10.359272236 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank0]:[W827 04:35:10.360047094 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank1]:[W827 04:35:10.370610409 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[2026-08-27 04:35:10 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.28
+[2026-08-27 04:35:10 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.29
+[2026-08-27 04:35:10] INFO:     127.0.0.1:60364 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 04:35:10] Disaggregation warmup requests completed for all 1 DP ranks
+[2026-08-27 04:35:10] End of disaggregation warmup
+[2026-08-27 04:35:10] The server is fired up and ready to roll!
+[2026-08-27 04:35:29 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.39
+[2026-08-27 04:35:29 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.39
+[2026-08-27 04:35:31 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25.83
+[2026-08-27 04:35:31 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25.84
+[2026-08-27 04:35:32] INFO:     174.1.60.1:49902 - "GET /health HTTP/1.1" 200 OK
+[2026-08-27 04:43:22 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.14
+[2026-08-27 04:43:22 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.14
+[2026-08-27 04:43:23] INFO:     174.1.60.1:33240 - "GET /health HTTP/1.1" 200 OK
+/sgl-workspace/sglang/python/sglang/srt/entrypoints/http_server.py:450: FastAPIDeprecationWarning: ORJSONResponse is deprecated, FastAPI now serializes data directly to JSON bytes via Pydantic when a return type or response model is set, which is faster and doesn't need a custom response class. Read more in the FastAPI docs: https://fastapi.tiangolo.com/advanced/custom-response/#orjson-or-response-model and https://fastapi.tiangolo.com/tutorial/response-model/
+  return await original_handler(ORJSONRequest(request.scope, request.receive))
+[2026-08-27 04:45:43] INFO:     174.1.60.5:35536 - "GET /v1/models HTTP/1.1" 200 OK
+[2026-08-27 04:45:43] INFO:     174.1.60.5:35542 - "GET /v1/models HTTP/1.1" 200 OK
+[2026-08-27 04:48:11] INFO:     174.1.60.5:47258 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 04:48:13 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.22
+[2026-08-27 04:48:13 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.22
+[2026-08-27 04:48:13 PP0 TP1 EP1] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:13 PP0 TP2 EP2] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:13 PP0 TP0 EP0] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:13 PP0 TP3 EP3] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP1 TP1 EP1] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP1 TP2 EP2] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP1 TP0 EP0] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP1 TP3 EP3] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:16] INFO:     174.1.60.5:50348 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 04:48:16] INFO:     174.1.60.5:50364 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 04:54:46] INFO:     174.1.60.5:51618 - "GET /v1/models HTTP/1.1" 200 OK
+[2026-08-27 04:54:46] INFO:     174.1.60.5:51626 - "GET /v1/models HTTP/1.1" 200 OK
+[2026-08-27 04:57:13] INFO:     174.1.60.5:48050 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 04:57:14 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.12
+[2026-08-27 04:57:14 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.12
+[2026-08-27 04:57:14] INFO:     174.1.60.5:48060 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 04:57:14] INFO:     174.1.60.5:48074 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 04:57:40] INFO:     174.1.60.5:51792 - "GET /v1/models HTTP/1.1" 200 OK
+[2026-08-27 04:57:40] INFO:     174.1.60.5:51798 - "GET /v1/models HTTP/1.1" 200 OK
+[2026-08-27 05:00:07] INFO:     174.1.60.5:48432 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:04:21 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1088, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2.55
+[2026-08-27 05:04:21 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1088, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2.55
+[2026-08-27 05:04:22] INFO:     174.1.60.5:36022 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:04:22] INFO:     174.1.60.5:36028 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:04:41 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.19
+[2026-08-27 05:04:41 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.19
+[2026-08-27 05:04:41] INFO:     174.1.60.5:33310 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:04:49] INFO:     174.1.60.5:39866 - "GET /v1/models HTTP/1.1" 200 OK
+[2026-08-27 05:04:49] INFO:     174.1.60.5:39874 - "GET /v1/models HTTP/1.1" 200 OK
+[2026-08-27 05:05:38] INFO:     174.1.60.5:57632 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:05:39 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1.10
+[2026-08-27 05:05:39 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1.10
+[2026-08-27 05:07:16] INFO:     174.1.60.5:42880 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:07:30 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 73.88
+[2026-08-27 05:07:32 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 72.71
+[2026-08-27 05:07:32 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3187.73
+[2026-08-27 05:07:33 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9987.21
+[2026-08-27 05:07:33 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 5011.85
+[2026-08-27 05:07:33 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7713.95
+[2026-08-27 05:07:33] INFO:     174.1.60.5:49324 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:07:38 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8673, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1473.19
+[2026-08-27 05:07:39 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8673, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1235.01
+[2026-08-27 05:07:40 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 481, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4257.97
+[2026-08-27 05:07:40 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 481, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9114.58
+[2026-08-27 05:07:40 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4643.44
+[2026-08-27 05:07:40 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7350.46
+[2026-08-27 05:07:40] INFO:     174.1.60.5:49326 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:07:46 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8656, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1470.23
+[2026-08-27 05:07:47 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8656, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1236.74
+[2026-08-27 05:07:48 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 464, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4291.82
+[2026-08-27 05:07:48 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 464, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9091.63
+[2026-08-27 05:07:48 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4126.73
+[2026-08-27 05:07:48 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6652.98
+[2026-08-27 05:07:48] INFO:     174.1.60.5:38928 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:07:53 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8663, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1477.97
+[2026-08-27 05:07:54 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8663, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1240.34
+[2026-08-27 05:07:55 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 471, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4270.28
+[2026-08-27 05:07:55 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 471, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9111.26
+[2026-08-27 05:07:55 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4676.01
+[2026-08-27 05:07:55 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7851.58
+[2026-08-27 05:07:56] INFO:     174.1.60.5:34214 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:08:01 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8659, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1474.57
+[2026-08-27 05:08:02 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8659, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1237.36
+[2026-08-27 05:08:03 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 467, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4273.37
+[2026-08-27 05:08:03 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 467, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9096.53
+[2026-08-27 05:08:03 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4082.13
+[2026-08-27 05:08:03 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6360.56
+[2026-08-27 05:08:03] INFO:     174.1.60.5:38204 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:08:08 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8692, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1476.50
+[2026-08-27 05:08:10 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8692, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1239.29
+[2026-08-27 05:08:10 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 500, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4267.60
+[2026-08-27 05:08:10 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 500, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8989.39
+[2026-08-27 05:08:11 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4419.86
+[2026-08-27 05:08:11 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8656.56
+[2026-08-27 05:08:11] INFO:     174.1.60.5:47402 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:08:16 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8557, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1477.68
+[2026-08-27 05:08:17 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8557, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1239.64
+[2026-08-27 05:08:18 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 365, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4273.32
+[2026-08-27 05:08:18 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 365, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9138.56
+[2026-08-27 05:08:18 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4206.66
+[2026-08-27 05:08:18 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8364.41
+[2026-08-27 05:08:18] INFO:     174.1.60.5:47404 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:08:24 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8683, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1475.28
+[2026-08-27 05:08:25 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8683, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1237.96
+[2026-08-27 05:08:26 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 491, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4263.38
+[2026-08-27 05:08:26 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 491, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9074.22
+[2026-08-27 05:08:26 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4537.22
+[2026-08-27 05:08:26 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7174.96
+[2026-08-27 05:08:26] INFO:     174.1.60.5:48174 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:08:26] INFO:     174.1.60.5:48178 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:09:17] INFO:     174.1.60.5:41772 - "GET /v1/models HTTP/1.1" 200 OK
+[2026-08-27 05:09:17] INFO:     174.1.60.5:41776 - "GET /v1/models HTTP/1.1" 200 OK
+[2026-08-27 05:11:44] INFO:     174.1.60.5:45272 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:11:45 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.32
+[2026-08-27 05:11:45 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.32
+E0827 05:11:45.751729  1489 worker_pool.cpp:1217] CQ completion timeout diagnostic: context=mlx5_3, outstanding=2, outstanding_age_ms=0, poll_gap_ms=199552, last_poll_interval_ms=0, max_poll_interval_ms=539649, submitted=34398, processed=34396
+[2026-08-27 05:11:46] INFO:     174.1.60.5:45274 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:11:46] INFO:     174.1.60.5:45276 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:12:14] INFO:     174.1.60.5:34478 - "GET /v1/models HTTP/1.1" 200 OK
+[2026-08-27 05:12:14] INFO:     174.1.60.5:34486 - "GET /v1/models HTTP/1.1" 200 OK
+[2026-08-27 05:14:42 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1088, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6.16
+[2026-08-27 05:14:42 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1088, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6.16
+[2026-08-27 05:14:42] INFO:     174.1.60.5:46142 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:14:43] INFO:     174.1.60.5:46156 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:14:43] INFO:     174.1.60.5:46162 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:15:12] INFO:     174.1.60.5:60686 - "GET /v1/models HTTP/1.1" 200 OK
+[2026-08-27 05:15:12] INFO:     174.1.60.5:60696 - "GET /v1/models HTTP/1.1" 200 OK
+[2026-08-27 05:17:40] INFO:     174.1.60.5:49298 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:17:45 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 44.69
+[2026-08-27 05:17:46 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 44.43
+[2026-08-27 05:17:47 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4291.89
+[2026-08-27 05:17:47 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9142.33
+[2026-08-27 05:17:47 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4792.42
+[2026-08-27 05:17:47 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8482.06
+[2026-08-27 05:17:47] INFO:     174.1.60.5:45358 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:17:53 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8673, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1475.80
+[2026-08-27 05:17:54 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8673, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1239.62
+[2026-08-27 05:17:55 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 481, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4273.03
+[2026-08-27 05:17:55 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 481, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9047.83
+[2026-08-27 05:17:55 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4673.57
+[2026-08-27 05:17:55 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8029.35
+[2026-08-27 05:17:55] INFO:     174.1.60.5:37922 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:18:00 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8656, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1479.36
+[2026-08-27 05:18:01 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8656, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1239.75
+[2026-08-27 05:18:02 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 464, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4268.49
+[2026-08-27 05:18:02 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 464, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9151.27
+[2026-08-27 05:18:02 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4773.15
+[2026-08-27 05:18:02 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8723.97
+[2026-08-27 05:18:02] INFO:     174.1.60.5:59842 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:18:08 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8663, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1480.27
+[2026-08-27 05:18:09 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8663, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1240.59
+[2026-08-27 05:18:10 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 471, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4269.06
+[2026-08-27 05:18:10 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 471, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9108.64
+[2026-08-27 05:18:10 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4695.76
+[2026-08-27 05:18:10 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8458.08
+[2026-08-27 05:18:10] INFO:     174.1.60.5:59848 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:18:15 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8659, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1476.05
+[2026-08-27 05:18:17 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8659, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1238.88
+[2026-08-27 05:18:17 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 467, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4269.01
+[2026-08-27 05:18:17 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 467, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9060.08
+[2026-08-27 05:18:17 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4685.85
+[2026-08-27 05:18:17 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8293.92
+[2026-08-27 05:18:18] INFO:     174.1.60.5:46946 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:18:23 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8692, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1478.44
+[2026-08-27 05:18:24 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8692, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1238.63
+[2026-08-27 05:18:25 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 500, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4261.28
+[2026-08-27 05:18:25 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 500, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9140.66
+[2026-08-27 05:18:25 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4667.18
+[2026-08-27 05:18:25 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8159.39
+[2026-08-27 05:18:25] INFO:     174.1.60.5:42086 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:18:31 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8557, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1480.63
+[2026-08-27 05:18:32 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8557, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1240.79
+[2026-08-27 05:18:32 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 365, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4272.82
+[2026-08-27 05:18:33 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 365, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9138.60
+[2026-08-27 05:18:33 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4248.66
+[2026-08-27 05:18:33 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8746.98
+[2026-08-27 05:18:33] INFO:     174.1.60.5:36136 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:18:38 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8683, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1480.90
+[2026-08-27 05:18:39 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8683, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1240.92
+[2026-08-27 05:18:40 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 491, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4261.25
+[2026-08-27 05:18:40 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 491, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9102.59
+[2026-08-27 05:18:40 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4532.61
+[2026-08-27 05:18:40 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8049.42
+[2026-08-27 05:18:40] INFO:     174.1.60.5:36144 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:18:40] INFO:     174.1.60.5:36150 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:19:09] INFO:     174.1.60.5:57946 - "GET /v1/models HTTP/1.1" 200 OK
+[2026-08-27 05:19:09] INFO:     174.1.60.5:57958 - "GET /v1/models HTTP/1.1" 200 OK
+[2026-08-27 05:21:37] INFO:     174.1.60.5:50774 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:37] INFO:     174.1.60.5:50788 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:37] INFO:     174.1.60.5:50802 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:37] INFO:     174.1.60.5:50814 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:37] INFO:     174.1.60.5:50826 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:37] INFO:     174.1.60.5:50840 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:37] INFO:     174.1.60.5:50844 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:37] INFO:     174.1.60.5:50846 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:43 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 2, #pending-token: 8652, #bootstrap-req: 2, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 44.89
+[2026-08-27 05:21:44 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 2, #pending-token: 8652, #bootstrap-req: 2, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 44.68
+[2026-08-27 05:21:44 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 3, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 5768.82
+[2026-08-27 05:21:45 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 3, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 5827.47
+[2026-08-27 05:21:45 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 3, #pending-token: 9185, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9671.83
+[2026-08-27 05:21:45 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 3, #pending-token: 9185, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 19750.66
+[2026-08-27 05:21:46 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 2, #pending-token: 993, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6916.66
+[2026-08-27 05:21:46 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 2, #pending-token: 993, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8672.27
+[2026-08-27 05:21:47 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 2, #pending-token: 9680, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10637.56
+[2026-08-27 05:21:47 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 2, #pending-token: 9680, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10354.39
+[2026-08-27 05:21:48 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1488, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8222.75
+[2026-08-27 05:21:48 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1488, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8344.22
+[2026-08-27 05:21:49 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 10199, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10480.00
+[2026-08-27 05:21:49 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 10199, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10387.32
+[2026-08-27 05:21:50 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 18858, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8578.91
+[2026-08-27 05:21:50 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 18858, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8578.44
+[2026-08-27 05:21:50 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 27591, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9988.39
+[2026-08-27 05:21:51 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 27591, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9972.71
+[2026-08-27 05:21:51 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 53023, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8837.11
+[2026-08-27 05:21:51] INFO:     174.1.60.5:43264 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:52 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 53023, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9171.43
+[2026-08-27 05:21:52 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 44876, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9896.54
+[2026-08-27 05:21:52 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 44876, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10032.97
+[2026-08-27 05:21:53 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 36684, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9174.07
+[2026-08-27 05:21:53] INFO:     174.1.60.5:43266 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:53 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 36684, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9172.94
+[2026-08-27 05:21:54 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 28504, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9586.45
+[2026-08-27 05:21:54 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 28504, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9826.60
+[2026-08-27 05:21:55 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 20312, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9357.29
+[2026-08-27 05:21:55] INFO:     174.1.60.5:43280 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:55 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 20312, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9123.76
+[2026-08-27 05:21:56 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 12139, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9447.89
+[2026-08-27 05:21:56 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 12139, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9658.04
+[2026-08-27 05:21:57 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 3947, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9443.97
+[2026-08-27 05:21:57] INFO:     174.1.60.5:43282 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:57 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 3947, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9638.68
+[2026-08-27 05:21:57 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 3968, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7366.63
+[2026-08-27 05:21:57 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 3968, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8902.23
+[2026-08-27 05:21:57] INFO:     174.1.60.5:43298 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:57] INFO:     174.1.60.5:43310 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:59] INFO:     174.1.60.5:43312 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:00] INFO:     174.1.60.5:43328 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:02 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 0, #pending-token: 8608, #bootstrap-req: 4, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1515.73
+[2026-08-27 05:22:03 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 0, #pending-token: 8608, #bootstrap-req: 4, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1303.59
+[2026-08-27 05:22:04 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 416, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4709.52
+[2026-08-27 05:22:04 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 416, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8416.60
+[2026-08-27 05:22:05 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 9149, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10991.56
+[2026-08-27 05:22:05 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 9149, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9867.38
+[2026-08-27 05:22:06 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 957, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7879.22
+[2026-08-27 05:22:06 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 957, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7915.28
+[2026-08-27 05:22:07 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 9638, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10712.86
+[2026-08-27 05:22:07 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 9638, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11079.28
+[2026-08-27 05:22:08 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1446, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8251.19
+[2026-08-27 05:22:08 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1446, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8053.96
+[2026-08-27 05:22:09 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 10033, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10494.65
+[2026-08-27 05:22:09 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 10033, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11035.63
+[2026-08-27 05:22:09 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1841, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8664.83
+[2026-08-27 05:22:10 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1841, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8259.51
+[2026-08-27 05:22:10 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1856, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6733.51
+[2026-08-27 05:22:10 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1856, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2578.77
+[2026-08-27 05:22:11 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8540, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7967.20
+[2026-08-27 05:22:11] INFO:     174.1.60.5:53576 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:11 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8540, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9138.35
+[2026-08-27 05:22:12 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 17066, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6059.77
+[2026-08-27 05:22:12 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 17066, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8568.79
+[2026-08-27 05:22:13 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8910, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10873.13
+[2026-08-27 05:22:13] INFO:     174.1.60.5:53584 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:13 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8910, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9603.70
+[2026-08-27 05:22:14 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 17466, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7794.12
+[2026-08-27 05:22:14 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 17466, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8622.72
+[2026-08-27 05:22:15 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9324, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10734.99
+[2026-08-27 05:22:15] INFO:     174.1.60.5:53590 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:15 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9324, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9808.25
+[2026-08-27 05:22:16 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 17751, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8115.61
+[2026-08-27 05:22:16 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 17751, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8814.22
+[2026-08-27 05:22:16 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9579, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10679.21
+[2026-08-27 05:22:17] INFO:     174.1.60.5:53596 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:17 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9579, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9729.78
+[2026-08-27 05:22:17 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1387, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8283.85
+[2026-08-27 05:22:18 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1387, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9429.57
+[2026-08-27 05:22:18 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1408, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6343.40
+[2026-08-27 05:22:18 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1408, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8479.29
+[2026-08-27 05:22:18] INFO:     174.1.60.5:53608 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:18] INFO:     174.1.60.5:53620 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:20] INFO:     174.1.60.5:53636 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:21] INFO:     174.1.60.5:44034 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:23 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8547, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1526.94
+[2026-08-27 05:22:24 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8547, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1315.46
+[2026-08-27 05:22:25 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 355, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4760.70
+[2026-08-27 05:22:25 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 355, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8428.33
+[2026-08-27 05:22:26 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8932, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10829.34
+[2026-08-27 05:22:26 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8932, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9556.73
+[2026-08-27 05:22:27 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 740, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7781.14
+[2026-08-27 05:22:27 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 740, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7932.50
+[2026-08-27 05:22:27 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9259, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10727.88
+[2026-08-27 05:22:28 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9259, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10866.67
+[2026-08-27 05:22:28 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1067, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8113.36
+[2026-08-27 05:22:29 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1067, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8138.49
+[2026-08-27 05:22:29 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9745, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10655.24
+[2026-08-27 05:22:29 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9745, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10900.49
+[2026-08-27 05:22:30 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1553, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8338.75
+[2026-08-27 05:22:30 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1553, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8227.97
+[2026-08-27 05:22:30 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1600, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6665.39
+[2026-08-27 05:22:31 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1600, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2199.04
+[2026-08-27 05:22:31 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8602, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7710.31
+[2026-08-27 05:22:31] INFO:     174.1.60.5:37214 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:32 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8602, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9261.97
+[2026-08-27 05:22:33 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 410, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6121.64
+[2026-08-27 05:22:33 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 410, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8576.46
+[2026-08-27 05:22:34 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8696, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10967.17
+[2026-08-27 05:22:34] INFO:     174.1.60.5:37230 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:34 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8696, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9567.88
+[2026-08-27 05:22:35 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 504, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7845.65
+[2026-08-27 05:22:35 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 504, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8659.79
+[2026-08-27 05:22:35 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9164, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10983.83
+[2026-08-27 05:22:35] INFO:     174.1.60.5:37246 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:36 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9164, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9874.30
+[2026-08-27 05:22:36 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 972, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7934.46
+[2026-08-27 05:22:36 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 972, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8762.34
+[2026-08-27 05:22:37 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9322, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10758.34
+[2026-08-27 05:22:37] INFO:     174.1.60.5:37262 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:37 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9322, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9763.35
+[2026-08-27 05:22:38 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1130, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8224.94
+[2026-08-27 05:22:38 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1130, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9463.33
+[2026-08-27 05:22:38 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1152, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6381.46
+[2026-08-27 05:22:38 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1152, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8785.04
+[2026-08-27 05:22:38] INFO:     174.1.60.5:37274 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:39] INFO:     174.1.60.5:37280 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:40] INFO:     174.1.60.5:37284 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:42] INFO:     174.1.60.5:57706 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:44 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8602, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1531.42
+[2026-08-27 05:22:44 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8602, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1318.33
+[2026-08-27 05:22:45 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 410, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4775.67
+[2026-08-27 05:22:45 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 410, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8534.85
+[2026-08-27 05:22:46 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 9105, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10895.13
+[2026-08-27 05:22:46 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 9105, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9621.85
+[2026-08-27 05:22:47 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 913, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7859.49
+[2026-08-27 05:22:47 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 913, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8012.84
+[2026-08-27 05:22:48 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9375, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10691.27
+[2026-08-27 05:22:48 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9375, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10846.99
+[2026-08-27 05:22:49 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1183, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8180.68
+[2026-08-27 05:22:49 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1183, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8142.25
+[2026-08-27 05:22:50 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9868, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10633.74
+[2026-08-27 05:22:50 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9868, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10953.99
+[2026-08-27 05:22:51 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1676, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8427.87
+[2026-08-27 05:22:51 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1676, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8243.07
+[2026-08-27 05:22:51 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6603.73
+[2026-08-27 05:22:52 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2380.37
+[2026-08-27 05:22:52 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8467, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7941.05
+[2026-08-27 05:22:52] INFO:     174.1.60.5:59718 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:52 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8467, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9276.71
+[2026-08-27 05:22:53 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 275, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6098.49
+[2026-08-27 05:22:53 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 275, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8603.25
+[2026-08-27 05:22:54 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8988, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11035.64
+[2026-08-27 05:22:54] INFO:     174.1.60.5:59730 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:54 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8988, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9672.42
+[2026-08-27 05:22:55 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 796, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7825.33
+[2026-08-27 05:22:55 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 796, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8635.70
+[2026-08-27 05:22:56 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9363, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10703.37
+[2026-08-27 05:22:56] INFO:     174.1.60.5:59738 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:56 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9363, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9819.19
+[2026-08-27 05:22:57 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 1171, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8175.11
+[2026-08-27 05:22:57 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 1171, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8823.65
+[2026-08-27 05:22:58 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9880, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10599.05
+[2026-08-27 05:22:58] INFO:     174.1.60.5:59744 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:58 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9880, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9922.77
+[2026-08-27 05:22:59 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1688, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8423.21
+[2026-08-27 05:22:59 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1688, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9407.16
+[2026-08-27 05:22:59 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6506.01
+[2026-08-27 05:22:59 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8452.04
+[2026-08-27 05:22:59] INFO:     174.1.60.5:59746 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:23:00] INFO:     174.1.60.5:59748 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:23:01] INFO:     174.1.60.5:59088 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:23:03] INFO:     174.1.60.5:59094 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:23:04 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8601, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1533.96
+[2026-08-27 05:23:05 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8601, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1320.52
+[2026-08-27 05:23:06 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 409, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4797.37
+[2026-08-27 05:23:06 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 409, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8455.98
+[2026-08-27 05:23:07 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 9104, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10736.15
+[2026-08-27 05:23:07 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 9104, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9859.62
+[2026-08-27 05:23:08 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 912, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8011.36
+[2026-08-27 05:23:08 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 912, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8059.42
+[2026-08-27 05:23:08 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9485, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10534.48
+[2026-08-27 05:23:09 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9485, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10717.32
+[2026-08-27 05:23:09 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1293, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8282.08
+[2026-08-27 05:23:10 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1293, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8188.84
+[2026-08-27 05:23:10 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9900, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10556.34
+[2026-08-27 05:23:10 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9900, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10839.28
+[2026-08-27 05:23:11 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1708, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8457.93
+[2026-08-27 05:23:11 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1708, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8302.17
+[2026-08-27 05:23:11 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6651.64
+[2026-08-27 05:23:12 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2387.61
+[2026-08-27 05:23:12 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8525, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7923.97
+[2026-08-27 05:23:13 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8525, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9210.87
+[2026-08-27 05:23:14 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 333, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6091.62
+[2026-08-27 05:23:14 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 333, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8517.86
+[2026-08-27 05:23:15 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8909, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10878.98
+[2026-08-27 05:23:15 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8909, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9736.23
+[2026-08-27 05:23:16 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 717, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7902.78
+[2026-08-27 05:23:16 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 717, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8689.24
+[2026-08-27 05:23:16 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9452, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10764.37
+[2026-08-27 05:23:17 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9452, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9778.82
+[2026-08-27 05:23:17 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 1260, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8087.43
+[2026-08-27 05:23:18 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 1260, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8848.19
+[2026-08-27 05:23:18 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9830, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10642.74
+[2026-08-27 05:23:18 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9830, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9957.02
+[2026-08-27 05:23:19 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1638, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8506.01
+[2026-08-27 05:23:19 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1638, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9493.47
+[2026-08-27 05:23:19 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1664, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6758.94
+[2026-08-27 05:23:19 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1664, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9006.24
+[2026-08-27 05:23:20] INFO:     174.1.60.5:56558 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:23:20] INFO:     174.1.60.5:56572 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:32:52 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.11
+[2026-08-27 05:32:52 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.11
+[2026-08-27 05:32:53] INFO:     174.1.60.1:38326 - "GET /health HTTP/1.1" 200 OK
+[2026-08-27 05:33:19] INFO:     174.1.60.5:50438 - "GET /v1/models HTTP/1.1" 200 OK
+[2026-08-27 05:33:19] INFO:     174.1.60.5:50442 - "GET /v1/models HTTP/1.1" 200 OK
+[2026-08-27 05:35:46] INFO:     174.1.60.5:53276 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:35:52 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 45.65
+[2026-08-27 05:35:53 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 45.39
+[2026-08-27 05:35:54 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4281.94
+[2026-08-27 05:35:54 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9046.45
+[2026-08-27 05:35:54 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4655.95
+[2026-08-27 05:35:54 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8048.57
+[2026-08-27 05:36:14] INFO:     174.1.60.5:41824 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:36:14] INFO:     174.1.60.5:41836 - "GET /server_info HTTP/1.1" 200 OK
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-20260827-1325/logs/manual_p_node1.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-20260827-1325/logs/manual_p_node1.log
new file mode 100644
index 0000000..809e03a
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-20260827-1325/logs/manual_p_node1.log
@@ -0,0 +1,1051 @@
+/usr/local/lib/python3.12/dist-packages/torchao/quantization/quant_api.py:1731: SyntaxWarning: invalid escape sequence '\.'
+  """Configuration class for applying different quantization configs to modules or parameters based on their fully qualified names (FQNs).
+/sgl-workspace/sglang/python/sglang/launch_server.py:56: UserWarning: 'python -m sglang.launch_server' is still supported, but 'sglang serve' is the recommended entrypoint.
+  Example: sglang serve --model-path <model> [options]
+  warnings.warn(
+[2026-08-27 04:28:43] Breakable CUDA graph is incompatible with MLA attention; disabling prefill CUDA graph.
+[2026-08-27 04:28:43] Attention backend not specified. Use triton backend by default.
+[2026-08-27 04:28:43] Pipeline parallelism is incompatible with overlap schedule.
+[2026-08-27 04:28:46] server_args=ServerArgs(model_path='/data/hf_models/Kimi-K3', tokenizer_path='/data/hf_models/Kimi-K3', tokenizer_mode='auto', tokenizer_backend='huggingface', tokenizer_worker_num=1, detokenizer_worker_num=1, skip_tokenizer_init=False, load_format='auto', model_loader_extra_config='{}', trust_remote_code=True, context_length=None, is_embedding=False, enable_multimodal=None, revision=None, model_impl='auto', model_config_parser='auto', json_model_override_args='{}', dtype='auto', quantization=None, quantization_param_path=None, kv_cache_dtype='auto', enable_fp32_lm_head=False, modelopt_quant=None, modelopt_checkpoint_restore_path=None, modelopt_checkpoint_save_path=None, modelopt_export_path=None, quantize_and_serve=False, rl_quant_profile=None, enable_tf32_matmul=False, mem_fraction_static=0.88, max_running_requests=None, max_queued_requests=None, max_total_tokens=None, chunked_prefill_size=8192, enable_dynamic_chunking=False, max_prefill_tokens=16384, prefill_max_requests=None, schedule_policy='fcfs', enable_priority_scheduling=False, disable_priority_preemption=False, default_priority_value=None, abort_on_priority_when_disabled=False, schedule_low_priority_values_first=False, priority_scheduling_preemption_threshold=10, retraction_policy='length', schedule_conservativeness=1.0, page_size=64, swa_full_tokens_ratio=0.8, disable_hybrid_swa_memory=False, radix_eviction_policy='lru', prefill_only_disable_kv_cache=False, disable_radix_cache=True, enable_page_major_kv_layout=False, enable_unified_memory=False, disable_chunked_prefix_cache=False, disable_overlap_schedule=True, num_continuous_decode_steps=1, scheduler_recv_interval=1, enable_mixed_chunk=False, nccl_port=None, dist_timeout=3600, dist_init_addr='174.1.60.1:20000', nnodes=4, node_rank=1, tp_size=4, dcp_size=1, pp_size=8, pp_max_micro_batch_size=None, pp_async_batch_depth=0, dp_size=1, load_balance_method='follow_bootstrap_room', attn_cp_size=1, moe_dp_size=1, dcp_comm_backend='ag_rs', dcp_replicate_q_proj=None, dwdp_size=1, enable_prefill_cp=False, cp_strategy=None, enable_dsa_cache_layer_split=False, enable_dsa_prefill_context_parallel=False, dsa_prefill_cp_mode='round-robin-split', enable_prefill_context_parallel=False, prefill_cp_mode='in-seq-split', enable_cp_decode_attn_tp=False, enable_dp_attention=False, enable_dp_attention_local_control_broadcast=False, enable_dp_lm_head=False, enable_attn_tp_input_scattered=False, disable_attn_tp_gather=False, enable_p2p_check=False, device='cuda', base_gpu_id=0, gpu_id_step=1, random_seed=577755302, watchdog_timeout=300, soft_watchdog_timeout=None, sleep_on_idle=False, use_ray=False, custom_sigquit_handler=None, numa_node=None, gc_threshold=None, host='0.0.0.0', port=30000, fastapi_root_path='', smg_grpc_mode=False, grpc_mode=False, grpc_port=None, sidecar=None, sidecar_args=None, skip_server_warmup=False, warmups=None, enable_http2=False, ssl_keyfile=None, ssl_certfile=None, ssl_ca_certs=None, ssl_keyfile_password=None, enable_ssl_refresh=False, api_key=None, admin_api_key=None, served_model_name='kimi-k3', weight_version='default', chat_template=None, hf_chat_template_name=None, completion_template=None, file_storage_path='sglang_storage', enable_cache_report=False, reasoning_parser=None, default_chat_template_kwargs=None, strip_thinking_cache=False, enable_strict_thinking=False, tool_call_parser=None, tool_server=None, sampling_defaults='model', asr_max_buffer_seconds=60, asr_max_concurrent_sessions=32, preferred_sampling_params=None, allow_auto_truncate=False, stream_interval=1, batch_notify_size=16, stream_response_default_include_usage=False, incremental_streaming_output=False, enable_streaming_session=False, enable_session_radix_cache=False, log_level='info', log_level_http=None, log_requests=False, log_requests_level=2, log_requests_format='text', log_requests_target=None, uvicorn_access_log_exclude_prefixes=[], crash_dump_folder=None, show_time_cost=False, enable_metrics=False, smg_http_sidecar_port=None, enable_mfu_metrics=False, enable_metrics_for_all_schedulers=False, load_snapshot_publish_interval=15, tokenizer_metrics_custom_labels_header='x-custom-labels', tokenizer_metrics_allowed_custom_labels=None, extra_metric_labels=None, bucket_time_to_first_token=None, bucket_inter_token_latency=None, bucket_e2e_request_latency=None, prompt_tokens_buckets=None, generation_tokens_buckets=None, gc_warning_threshold_secs=0.0, decode_log_interval=40, enable_request_time_stats_logging=False, kv_events_config=None, enable_forward_pass_metrics=False, forward_pass_metrics_worker_id='', forward_pass_metrics_ipc_name=None, enable_trace=False, trace_modules='request', otlp_traces_endpoint='localhost:4317', export_metrics_to_file=False, export_metrics_to_file_dir=None, stat_loggers=None, constrained_json_whitespace_pattern=None, constrained_json_disable_any_whitespace=False, attention_backend='triton', decode_attention_backend=None, prefill_attention_backend=None, sampling_backend='flashinfer', grammar_backend='xgrammar', radix_cache_backend=None, mm_attention_backend=None, fp8_gemm_runner_backend='auto', fp4_gemm_runner_backend='auto', bf16_gemm_backend='auto', dsa_prefill_backend=None, dsa_decode_backend=None, dsa_paged_mqa_logits_backend='auto', dsa_topk_backend='sgl-kernel', disable_flashinfer_autotune=False, flashinfer_autotune_skip_ops=None, mamba_backend='triton', cuda_graph_config=CudaGraphConfig(decode=PhaseConfig(backend='disabled', max_bs=16, bs=[1, 2, 4, 8, 12, 16], tc_compiler='eager', full_prefill_max_req=None), prefill=PhaseConfig(backend='disabled', max_bs=2048, bs=[4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], tc_compiler='eager', full_prefill_max_req=None)), cuda_graph_backend_decode=None, cuda_graph_backend_prefill=None, cuda_graph_max_bs_decode=16, cuda_graph_max_bs_prefill=None, cuda_graph_bs_decode=None, cuda_graph_bs_prefill=None, cuda_graph_tc_compiler=None, disable_prefill_cuda_graph=False, disable_decode_cuda_graph=False, disable_cuda_graph=False, disable_cuda_graph_padding=False, enable_profile_cuda_graph=False, enable_cudagraph_gc=False, debug_cuda_graph=False, enable_layerwise_nvtx_marker=False, enable_nccl_nvls=False, enable_symm_mem=False, triton_attention_reduce_in_fp32=False, triton_attention_num_kv_splits=8, triton_attention_split_tile_size=None, flashinfer_mla_disable_ragged=False, enable_fused_qk_norm_rope=False, enable_precise_embedding_interpolation=False, enable_fused_moe_sum_all_reduce=False, enable_deepseek_v4_fp4_indexer=False, disable_custom_all_reduce=False, enable_mscclpp=False, enable_torch_symm_mem=False, enable_scattered_sconv=False, pre_warm_nccl=False, enable_quant_communications=False, enable_flashinfer_allreduce_fusion=False, enforce_disable_flashinfer_allreduce_fusion=False, flashinfer_allreduce_fusion_backend=None, enable_aiter_allreduce_fusion=False, enable_torch_compile=False, enable_torch_compile_debug_mode=False, torch_compile_max_bs=32, torchao_config='', speculative_algorithm=None, speculative_draft_model_path=None, speculative_draft_model_revision=None, speculative_draft_load_format=None, speculative_num_steps=None, speculative_eagle_topk=None, speculative_num_draft_tokens=None, speculative_dflash_block_size=None, speculative_dspark_block_size=None, speculative_dspark_sps_table_path=None, speculative_dspark_confidence_sts_path=None, speculative_dspark_align_verify_tokens_to_graph_tier=False, speculative_accept_threshold_single=1.0, speculative_accept_threshold_acc=1.0, speculative_use_rejection_sampling=False, speculative_token_map=None, speculative_attention_mode='prefill', speculative_draft_attention_backend=None, speculative_draft_window_size=None, speculative_moe_runner_backend='flashinfer_mxfp4', speculative_moe_a2a_backend=None, speculative_draft_model_quantization=None, speculative_skip_dp_mlp_sync=False, enable_multi_layer_eagle=False, speculative_adaptive=False, speculative_adaptive_config=None, decoupled_spec_bind_endpoint=None, decoupled_spec_connect_endpoints=None, decoupled_spec_rank=None, decoupled_spec_role='null', spec_trace_dir=None, speculative_ngram_min_bfs_breadth=1, speculative_ngram_max_bfs_breadth=10, speculative_ngram_match_type='BFS', speculative_ngram_max_trie_depth=18, speculative_ngram_capacity=10000000, speculative_ngram_external_corpus_path=None, speculative_ngram_external_sam_budget=0, speculative_ngram_external_corpus_max_tokens=10000000, ep_size=4, moe_a2a_backend='none', moe_runner_backend='flashinfer_mxfp4', flashinfer_mxfp4_moe_precision='default', deepep_mode='auto', fuseep_mode=2, deepep_dispatcher_output_dtype='auto', ep_num_redundant_experts=0, ep_dispatch_algorithm=None, init_expert_location='trivial', enable_eplb=False, eplb_algorithm='auto', eplb_rebalance_num_iterations=1000, eplb_rebalance_layers_per_chunk=None, eplb_min_rebalancing_utilization_threshold=1.0, expert_distribution_recorder_mode=None, expert_distribution_recorder_buffer_size=1000, enable_expert_distribution_metrics=False, deepep_config=None, moe_dense_tp_size=None, elastic_ep_backend=None, enable_elastic_expert_backup=False, mooncake_ib_device=None, enable_waterfill=False, ep_join_mode=None, ep_join_rank_offset=0, elastic_ep_initial_size=None, max_ep_size=None, elastic_ep_scale_timeout=600, elastic_ep_rejoin=False, disable_flashinfer_cutlass_moe_fp4_allgather=False, disable_shared_experts_fusion=False, enforce_shared_experts_fusion=False, max_mamba_cache_size=None, mamba_ssm_dtype=None, mamba_max_states_per_path=-1, enable_mamba_cache_stochastic_rounding=False, mamba_cache_philox_rounds=0, mamba_full_memory_ratio=0.36, mamba_radix_cache_strategy='extra_buffer_lazy', uses_mamba_radix_cache=False, mamba_track_interval=256, enable_int8_mamba_checkpoint=False, int8_mamba_ckpt_size=None, linear_attn_backend='triton', linear_attn_decode_backend=None, linear_attn_prefill_backend=None, linear_attn_verify_backend=None, enable_linear_replayssm=False, linear_replayssm_cache_len=16, enable_linear_replayssm_spec=False, enable_hierarchical_cache=False, hicache_ratio=2.0, hicache_size=0, hicache_write_policy='write_through', hicache_io_backend='kernel', hicache_mem_layout='page_first', hicache_storage_backend=None, hicache_storage_prefetch_policy='timeout', hicache_storage_backend_extra_config=None, enable_hisparse=False, hisparse_config=None, enable_broadcast_mm_inputs_process=False, enable_prefix_mm_cache=False, mm_enable_dp_encoder=False, mm_process_config={}, mm_processor_worker_num=0, mm_io_worker_num=0, limit_mm_data_per_request=None, enable_mm_global_cache=False, disable_fast_image_processor=False, mm_feature_transport='cpu', keep_mm_feature_on_device=False, enable_lora=None, enable_lora_overlap_loading=None, max_lora_rank=None, lora_target_modules=None, lora_paths=None, max_loaded_loras=None, max_loras_per_batch=8, lora_eviction_policy='lru', lora_backend='csgmv', max_lora_chunk_size=16, experts_shared_outer_loras=None, lora_use_virtual_experts=False, lora_strict_loading=False, lora_drain_wait_threshold=0.0, enable_two_batch_overlap=False, enable_single_batch_overlap=False, tbo_token_distribution_threshold=0.48, cpu_offload_gb=0, offload_group_size=-1, offload_num_in_group=1, offload_prefetch_step=1, offload_mode='cpu', enable_lmcache=False, lmcache_config_file=None, enable_flexkv=False, flexkv_config_file=None, kt_weight_path=None, kt_method='AMXINT4', kt_cpuinfer=None, kt_threadpool_count=2, kt_num_gpu_experts=None, kt_max_deferred_experts_per_token=None, dllm_algorithm=None, dllm_algorithm_config=None, dllm_fdfo=True, disaggregation_mode='prefill', disaggregation_transfer_backend='mooncake', disaggregation_bootstrap_port=28800, disaggregation_ib_device='mlx5_0,mlx5_1,mlx5_2,mlx5_3', disaggregation_decode_enable_radix_cache=False, disaggregation_decode_enable_offload_kvcache=False, num_reserved_decode_tokens=512, disaggregation_decode_extra_slots=None, disaggregation_decode_polling_interval=1, optimistic_prefill_attempts=0, encoder_only=False, language_only=False, encoder_transfer_backend='zmq_to_tokenizer', encoder_urls=[], encoder_bootstrap_port=8997, encoder_register_urls=[], enable_adaptive_dispatch_to_encoder=False, enable_pdmux=False, pdmux_config_path=None, sm_group_num=8, custom_weight_loader=[], weight_loader_disable_mmap=False, weight_loader_prefetch_checkpoints=False, weight_loader_prefetch_num_threads=4, weight_loader_drop_cache_after_load=False, remote_instance_weight_loader_seed_instance_ip=None, remote_instance_weight_loader_seed_instance_service_port=None, remote_instance_weight_loader_send_weights_group_ports=None, remote_instance_weight_loader_backend='nccl', remote_instance_weight_loader_start_seed_via_transfer_engine=False, engine_info_bootstrap_port=6789, modelexpress_config=None, download_dir=None, model_checksum=None, delete_ckpt_after_loading=False, decrypted_config_file=None, decrypted_draft_config_file=None, checkpoint_engine_wait_weights_before_ready=False, enable_prefill_delayer=False, prefill_delayer_max_delay_passes=30, prefill_delayer_token_usage_low_watermark=None, prefill_delayer_forward_passes_buckets=None, prefill_delayer_wait_seconds_buckets=None, prefill_delayer_queue_min_ratio=None, prefill_delayer_max_delay_ms=None, min_free_slots_delay=None, enable_deterministic_inference=False, rl_on_policy_target=None, kv_canary='none', kv_canary_real_data='none', kv_canary_sweep_interval=0, enable_dynamic_batch_tokenizer=False, dynamic_batch_tokenizer_batch_size=32, dynamic_batch_tokenizer_batch_timeout=0.002, enable_tokenizer_batch_encode=False, disable_tokenizer_batch_decode=False, debug_tensor_dump_output_folder=None, debug_tensor_dump_layers=None, debug_tensor_dump_input_file=None, enable_memory_saver=False, enable_weights_cpu_backup=False, enable_draft_weights_cpu_backup=False, enable_custom_logit_processor=False, enable_return_hidden_states=False, enable_return_routed_experts=False, enable_return_indexer_topk=False, disable_outlines_disk_cache=False, enable_mis=False, weight_cache_mode='off', weight_cache_socket=None, weight_cache_timeout=1800, forward_hooks=None, msprobe_dump_config=None)
+[2026-08-27 04:28:46] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:46] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:46] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:46] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:46] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:46] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:46] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:46] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:02 PP3 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:02 PP3 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:03 PP3 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:03 PP3 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:03 PP3 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:03 PP3 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:03 PP2 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:03 PP2 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:03 PP2 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:03 PP2 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:04 PP2 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:04 PP2 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:04 PP2 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:04 PP2 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:04 PP3 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:04 PP3 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:05 PP3 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:05 PP2 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:05 PP3 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:05 PP3 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:05 PP2 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:06.044592   241 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:06.044694   241 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:06.044710   241 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.2 port: 12001
+I0827 04:29:06.044759   241 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.2:16343
+I0827 04:29:06.044883   241 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:06.053642   242 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:06.053738   242 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:06.053754   242 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.2 port: 12001
+I0827 04:29:06.053797   242 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.2:15170
+I0827 04:29:06.053925   242 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:06.058039   237 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:06.058135   237 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:06.058149   237 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.2 port: 12001
+I0827 04:29:06.058192   237 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.2:15486
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:06.058215   236 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:06.058310   236 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:06.058316   237 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:06.058326   236 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.2 port: 12001
+I0827 04:29:06.058367   236 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.2:16140
+I0827 04:29:06.058492   236 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:06.059734   239 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:06.059828   239 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:06.059844   239 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.2 port: 12001
+I0827 04:29:06.059890   239 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.2:15755
+I0827 04:29:06.060027   239 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:06.064433   237 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:06.064513   236 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:06.065588   239 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:06.069945   237 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:06.070852   236 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:06.071113   239 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:06.071978   241 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:06.079348   236 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:06.079967   242 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:06.080070   241 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:06.087044   241 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:06.087074   242 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:06.094024   242 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:06.095021   241 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:06.095822   237 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:06.097810   241 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:06.097839   241 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:06.097889   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.099939   239 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:06.100984   242 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:06.103683   242 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:06.103708   242 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:06.103756   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.103785   236 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:06.103833   237 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:06.106014   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:06.106982   236 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:06.107007   236 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:06.107053   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.107446   241 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:02
+I0827 04:29:06.107475   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.107936   239 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:06.107971   237 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:06.107997   237 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:06.108048   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.110760   239 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:06.110790   239 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:06.110847   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.111824   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:06.113214   242 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:02
+I0827 04:29:06.113242   242 rdma_context.cpp:205] Using SIEVE endpoint store
+[2026-08-27 04:29:06 PP2 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+I0827 04:29:06.114821   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:06.115196   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+[2026-08-27 04:29:06 PP2 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+I0827 04:29:06.116189   241 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:02
+I0827 04:29:06.116214   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.116621   237 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:02
+I0827 04:29:06.116650   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.118325   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:06.119874   239 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:02
+I0827 04:29:06.119906   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.121601   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:06.122952   242 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:02
+I0827 04:29:06.122977   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.127393   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:06.128841   239 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:02
+I0827 04:29:06.128867   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.132040   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:06.133428   236 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:02
+I0827 04:29:06.133458   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.136312   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:06.137817   239 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:02
+I0827 04:29:06.137846   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.144285   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:06.144373   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:06.145742   237 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:02
+I0827 04:29:06.145776   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.145833   241 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:02
+I0827 04:29:06.145866   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.152084   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:06.153820   242 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:02
+I0827 04:29:06.153860   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.154352   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:06.155728   241 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:02
+I0827 04:29:06.155835   241 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:06 PP3 TP2 EP2] Init torch distributed begin.
+I0827 04:29:06.163965   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:06.164443   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:06.164605   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:06.165319   236 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:02
+I0827 04:29:06.165345   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.165951   239 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:02
+I0827 04:29:06.166070   239 transfer_engine_impl.cpp:388] installTransport, type=rdma
+I0827 04:29:06.166186   242 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:02
+I0827 04:29:06.166311   242 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:06 PP3 TP0 EP0] Init torch distributed begin.
+[2026-08-27 04:29:06 PP3 TP3 EP3] Init torch distributed begin.
+I0827 04:29:06.172065   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:06.173699   237 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:02
+I0827 04:29:06.173743   237 rdma_context.cpp:205] Using SIEVE endpoint store
+[2026-08-27 04:29:06 PP3 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+I0827 04:29:06.192251   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:06.193763   236 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:02
+I0827 04:29:06.193797   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.200259   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:06.201810   237 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:02
+I0827 04:29:06.201937   237 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:06 PP2 TP2 EP2] Init torch distributed begin.
+I0827 04:29:06.220070   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:06.221505   236 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:02
+I0827 04:29:06.221653   236 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:06 PP2 TP1 EP1] Init torch distributed begin.
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:06.613179   235 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:06.613286   235 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:06.613301   235 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.2 port: 12001
+I0827 04:29:06.613344   235 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.2:16382
+I0827 04:29:06.613492   235 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:06.618471   235 topology.cpp:127] Device mlx5_0 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:06.618719   238 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:06.618824   238 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:06.618839   238 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.2 port: 12001
+I0827 04:29:06.618882   238 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.2:16170
+I0827 04:29:06.619020   238 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:06.624203   238 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:06.629066   238 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:06.633975   238 topology.cpp:127] Device mlx5_2 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:06.636366   240 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:06.636454   240 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:06.636466   240 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.2 port: 12001
+I0827 04:29:06.636509   240 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.2:15840
+I0827 04:29:06.636675   240 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:06.638876   238 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:06.640780   238 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:06.640802   238 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:06.640842   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.641396   240 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:06.643196   235 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:06.645701   240 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:06.649896   235 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:06.650310   240 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:06.654819   240 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:06.656375   240 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:06.656396   240 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:06.656432   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.657156   235 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:06.659863   235 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:06.659895   235 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:06.659950   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.661671   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:06.662956   240 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:02
+I0827 04:29:06.662979   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.664074   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:06.665762   238 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:02
+I0827 04:29:06.665796   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.667351   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:06.668808   235 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:02
+I0827 04:29:06.668838   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.676294   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:06.677721   235 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:02
+I0827 04:29:06.677748   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.692129   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:06.693621   240 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:02
+I0827 04:29:06.693648   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.696089   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:06.697444   238 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:02
+I0827 04:29:06.697470   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.707720   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:06.709362   235 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:02
+I0827 04:29:06.709394   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.719987   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:06.721463   240 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:02
+I0827 04:29:06.721489   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.728108   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:06.728855   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:06.729467   238 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:02
+I0827 04:29:06.729491   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.730392   240 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:02
+I0827 04:29:06.730533   240 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:06 PP3 TP1 EP1] Init torch distributed begin.
+I0827 04:29:06.736888   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:06.738317   238 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:02
+I0827 04:29:06.738422   238 transfer_engine_impl.cpp:388] installTransport, type=rdma
+I0827 04:29:06.738466   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+[2026-08-27 04:29:06 PP2 TP3 EP3] Init torch distributed begin.
+I0827 04:29:06.739869   235 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:02
+I0827 04:29:06.739982   235 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:06 PP2 TP0 EP0] Init torch distributed begin.
+[2026-08-27 04:29:26 PP3 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:26 PP2 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:27 PP2 TP3 EP3] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP2 TP1 EP1] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP2 TP2 EP2] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP2 TP0 EP0] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP3 TP3 EP3] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP3 TP0 EP0] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP3 TP2 EP2] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP3 TP1 EP1] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP3 TP3 EP3] Init torch distributed ends. elapsed=21.61 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP3 TP2 EP2] Init torch distributed ends. elapsed=21.62 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP3 TP1 EP1] Init torch distributed ends. elapsed=21.04 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP3 TP0 EP0] Init torch distributed ends. elapsed=21.61 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP2 TP2 EP2] Init torch distributed ends. elapsed=21.57 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP2 TP1 EP1] Init torch distributed ends. elapsed=21.55 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP2 TP0 EP0] Init torch distributed ends. elapsed=21.03 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP2 TP3 EP3] Init torch distributed ends. elapsed=21.04 s, mem usage=0.54 GB
+[2026-08-27 04:29:28 PP3 TP3 EP3] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP3 TP0 EP0] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP2 TP0 EP0] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP3 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP2 TP1 EP1] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP3 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP2 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP3 TP2 EP2] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP2 TP3 EP3] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP2 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP2 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP3 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP2 TP2 EP2] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP3 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP3 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP2 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP3 TP1 EP1] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP2 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP3 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP2 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP3 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP3 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP2 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP3 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP2 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP3 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP2 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP2 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP2 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP2 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP3 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP3 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:30:31 PP2 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:31 PP2 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:31 PP2 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:31 PP2 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:31 PP2 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:31 PP2 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:31 PP2 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:31 PP2 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP2 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:32 PP2 TP0 EP0] Load weight end. elapsed=64.92 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=36.14 GB, mem usage=45.91 GB.
+[2026-08-27 04:30:33 PP3 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:33 PP3 TP1 EP1] Load weight end. elapsed=65.31 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:34 PP2 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:34 PP2 TP1 EP1] Load weight end. elapsed=66.07 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=36.14 GB, mem usage=45.91 GB.
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:35 PP3 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:35 PP3 TP0 EP0] Load weight end. elapsed=67.41 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:36 PP2 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:36 PP2 TP3 EP3] Load weight end. elapsed=68.04 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=36.14 GB, mem usage=45.91 GB.
+[2026-08-27 04:30:36 PP2 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:36 PP2 TP2 EP2] Load weight end. elapsed=68.23 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=36.14 GB, mem usage=45.91 GB.
+[2026-08-27 04:30:36 PP2 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:36 PP2 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:36 PP2 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:36 PP2 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:36 PP2 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:36 PP2 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:36 PP2 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:36 PP2 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:36 PP3 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:36 PP3 TP3 EP3] Load weight end. elapsed=68.86 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:37 PP3 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:37 PP3 TP2 EP2] Load weight end. elapsed=69.68 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:38 PP3 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:38 PP3 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:38 PP3 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:38 PP3 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:38 PP3 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:38 PP3 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:38 PP3 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:38 PP3 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:54 PP3 TP3 EP3] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP3 TP2 EP2] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP3 TP1 EP1] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP2 TP3 EP3] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP3 TP0 EP0] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP2 TP2 EP2] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP2 TP1 EP1] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP2 TP0 EP0] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP2 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:54 PP2 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:54 PP3 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP3 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP2 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:54 PP3 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP3 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP2 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:54 PP3 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP3 TP3 EP3] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP2 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP2 TP2 EP2] Memory pool end. avail mem=23.25 GB
+[2026-08-27 04:30:54 PP2 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP2 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP2 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP3 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP3 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP2 TP3 EP3] Memory pool end. avail mem=23.25 GB
+[2026-08-27 04:30:54 PP2 TP1 EP1] Memory pool end. avail mem=23.25 GB
+[2026-08-27 04:30:54 PP3 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP3 TP1 EP1] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP3 TP2 EP2] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP2 TP0 EP0] Memory pool end. avail mem=23.25 GB
+[2026-08-27 04:30:54 PP3 TP0 EP0] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP3 TP3 EP3] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp3_pp3_dp0.json
+[2026-08-27 04:30:54 PP2 TP1 EP1] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp1_pp2_dp0.json
+[2026-08-27 04:30:54 PP2 TP2 EP2] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp2_pp2_dp0.json
+[2026-08-27 04:30:54 PP3 TP1 EP1] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp1_pp3_dp0.json
+[2026-08-27 04:30:54 PP2 TP3 EP3] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp3_pp2_dp0.json
+[2026-08-27 04:30:54 PP3 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:30:54 PP3 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:30:54 PP3 TP2 EP2] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp2_pp3_dp0.json
+[2026-08-27 04:30:54 PP2 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:30:54 PP2 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:30:54 PP3 TP0 EP0] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp0_pp3_dp0.json
+[2026-08-27 04:30:54 PP2 TP0 EP0] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp0_pp2_dp0.json
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO[TensorRT-LLM][INFO] Set logger level to INFO
+
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.65profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.60profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.79profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.51profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.86profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.27profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.62profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.98profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.86profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 12.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 13.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.93profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.92profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.09profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.63profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  6.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  7.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.33profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.58profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  8.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.56profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.23profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.63profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.36profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.09profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.97profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  8.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.30profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 13.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 13.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  8.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  8.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 11.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 11.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.23profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.48profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.23profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.53profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.27profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.59profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.62profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.97profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.60profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.46profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.12profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.68profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.77profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.62profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  6.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.87profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.30profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.40profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.83profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.94profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.59profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.66profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.29profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.23profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:01,  3.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.68profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.75profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  7.01profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.22profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.65profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.62profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.46profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.22profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  7.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:01<00:06,  1.03s/profile]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:01,  3.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.50profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.93profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.48profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.11profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.12profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.30profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.68profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  7.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 11.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  9.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 12.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 11.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 11.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 11.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.11profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.31profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.65profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.31profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.35profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.49profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:01<00:01,  2.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.65profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.68profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  7.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.09profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 11.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.88profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.98profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.50profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.62profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.51profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.91profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:01,  3.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.31profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.86profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.27profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.59profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.17profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.76profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.73profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  9.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.37profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.15profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.98profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.65profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.97profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.23profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.46profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  9.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.03profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.57profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.40profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.40profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.57profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.13profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.56profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  6.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.73profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.23profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  3.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.86profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.77profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.92profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.37profile/s]
+[2026-08-27 04:31:26 PP2 TP3 EP3] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.39profile/s]
+[2026-08-27 04:31:26 PP2 TP2 EP2] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.75profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.46profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.35profile/s]
+[2026-08-27 04:31:26 PP2 TP1 EP1] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.12profile/s]
+[2026-08-27 04:31:26 PP2 TP0 EP0] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.66profile/s][2026-08-27 04:31:26 PP2 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:26 PP2 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:26 PP2 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:26 PP2 TP0 EP0] max_total_num_tokens=3801408, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=54, context_len=1048576, available_gpu_mem=20.47 GB
+[2026-08-27 04:31:26 PP2 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:26 PP2 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:26 PP2 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:26 PP2 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:26 PP2 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.93profile/s]
+[2026-08-27 04:31:27 PP3 TP1 EP1] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.16profile/s]
+[2026-08-27 04:31:27 PP3 TP0 EP0] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.43profile/s]
+[2026-08-27 04:31:27 PP3 TP2 EP2] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.05profile/s]
+[2026-08-27 04:31:27 PP3 TP3 EP3] FlashInfer autotune completed.
+[2026-08-27 04:31:27 PP3 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:27 PP3 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:27 PP3 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:27 PP3 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:27 PP3 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:27 PP3 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:27 PP3 TP0 EP0] max_total_num_tokens=3801408, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=54, context_len=1048576, available_gpu_mem=16.54 GB
+[2026-08-27 04:31:27 PP3 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:27 PP3 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+W0827 04:31:30.590376  2521 memory_location.cpp:73] Failed to get NUMA node, addr: 0x513c5ec0, len: 6912: Operation not permitted [1]
+W0827 04:31:30.590432  2522 memory_location.cpp:73] Failed to get NUMA node, addr: 0x514a0900, len: 6912: Operation not permitted [1]
+W0827 04:31:30.590495  2525 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51226dc0, len: 55296: Operation not permitted [1]
+W0827 04:31:30.590492  2523 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5128db40, len: 6912: Operation not permitted [1]
+W0827 04:31:30.590533  2524 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5156ff80, len: 6912: Operation not permitted [1]
+W0827 04:31:30.590600  2526 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51320ac0, len: 55296: Operation not permitted [1]
+W0827 04:31:30.590636  2527 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51768540, len: 6912: Operation not permitted [1]
+W0827 04:31:30.590826  2528 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51483dc0, len: 13824: Operation not permitted [1]
+W0827 04:31:30.590914  2529 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51119380, len: 6912: Operation not permitted [1]
+W0827 04:31:30.590979  2530 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50fe1e00, len: 6912: Operation not permitted [1]
+W0827 04:31:30.846120  2547 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50ceebc0, len: 6912: Operation not permitted [1]
+W0827 04:31:30.846271  2548 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50c57640, len: 6912: Operation not permitted [1]
+W0827 04:31:30.846328  2549 memory_location.cpp:73] Failed to get NUMA node, addr: 0x502ed680, len: 6912: Operation not permitted [1]
+W0827 04:31:30.846354  2550 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50d1c800, len: 6912: Operation not permitted [1]
+W0827 04:31:30.846448  2551 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50510a40, len: 55296: Operation not permitted [1]
+W0827 04:31:30.846473  2552 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5051e280, len: 55296: Operation not permitted [1]
+W0827 04:31:30.846571  2553 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50d01340, len: 6912: Operation not permitted [1]
+W0827 04:31:30.846689  2554 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50ce3440, len: 13824: Operation not permitted [1]
+W0827 04:31:30.846858  2555 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5030f440, len: 6912: Operation not permitted [1]
+W0827 04:31:30.846926  2556 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50cae980, len: 6912: Operation not permitted [1]
+W0827 04:31:31.198441  2575 memory_location.cpp:73] Failed to get NUMA node, addr: 0x514f46c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.198501  2576 memory_location.cpp:73] Failed to get NUMA node, addr: 0x509c6400, len: 6912: Operation not permitted [1]
+W0827 04:31:31.198501  2577 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51513300, len: 6912: Operation not permitted [1]
+W0827 04:31:31.198679  2578 memory_location.cpp:73] Failed to get NUMA node, addr: 0x506eef40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.198787  2579 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50dc9180, len: 55296: Operation not permitted [1]
+W0827 04:31:31.198915  2580 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50904700, len: 55296: Operation not permitted [1]
+W0827 04:31:31.198982  2581 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50aa9440, len: 6912: Operation not permitted [1]
+W0827 04:31:31.199081  2582 memory_location.cpp:73] Failed to get NUMA node, addr: 0x515012c0, len: 13824: Operation not permitted [1]
+W0827 04:31:31.199148  2583 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5161ac80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.199185  2584 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50dd69c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.763885  2603 memory_location.cpp:73] Failed to get NUMA node, addr: 0x519f1500, len: 6912: Operation not permitted [1]
+W0827 04:31:31.763998  2605 memory_location.cpp:73] Failed to get NUMA node, addr: 0x517df940, len: 6912: Operation not permitted [1]
+W0827 04:31:31.763923  2604 memory_location.cpp:73] Failed to get NUMA node, addr: 0x514ab440, len: 6912: Operation not permitted [1]
+W0827 04:31:31.764139  2606 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51556180, len: 6912: Operation not permitted [1]
+W0827 04:31:31.764169  2608 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50ed8140, len: 55296: Operation not permitted [1]
+W0827 04:31:31.764184  2607 memory_location.cpp:73] Failed to get NUMA node, addr: 0x515df040, len: 55296: Operation not permitted [1]
+W0827 04:31:31.764356  2609 memory_location.cpp:73] Failed to get NUMA node, addr: 0x517e3580, len: 6912: Operation not permitted [1]
+W0827 04:31:31.764564  2610 memory_location.cpp:73] Failed to get NUMA node, addr: 0x519cdb80, len: 13824: Operation not permitted [1]
+W0827 04:31:31.764624  2611 memory_location.cpp:73] Failed to get NUMA node, addr: 0x519dbf80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.764714  2612 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50f08d80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.868036  2629 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50a7a680, len: 6912: Operation not permitted [1]
+W0827 04:31:31.868046  2630 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50b6e400, len: 6912: Operation not permitted [1]
+W0827 04:31:31.868139  2631 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50e5f580, len: 6912: Operation not permitted [1]
+W0827 04:31:31.868261  2632 memory_location.cpp:73] Failed to get NUMA node, addr: 0x503c7100, len: 6912: Operation not permitted [1]
+W0827 04:31:31.868330  2633 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50c94ac0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.868432  2635 memory_location.cpp:73] Failed to get NUMA node, addr: 0x509b2880, len: 6912: Operation not permitted [1]
+W0827 04:31:31.868502  2634 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50b14c80, len: 55296: Operation not permitted [1]
+W0827 04:31:31.868564  2636 memory_location.cpp:73] Failed to get NUMA node, addr: 0x503ee840, len: 13824: Operation not permitted [1]
+W0827 04:31:31.868708  2637 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51057d80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.868835  2638 memory_location.cpp:73] Failed to get NUMA node, addr: 0x509cd940, len: 6912: Operation not permitted [1]
+W0827 04:31:32.003983  2656 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50715b00, len: 6912: Operation not permitted [1]
+W0827 04:31:32.004096  2657 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51346e80, len: 6912: Operation not permitted [1]
+W0827 04:31:32.004005  2655 memory_location.cpp:73] Failed to get NUMA node, addr: 0x507176c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.004179  2658 memory_location.cpp:73] Failed to get NUMA node, addr: 0x507e3a40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.004302  2659 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5078aa80, len: 55296: Operation not permitted [1]
+W0827 04:31:32.004357  2660 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5135c980, len: 55296: Operation not permitted [1]
+W0827 04:31:32.004410  2661 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5099f180, len: 6912: Operation not permitted [1]
+W0827 04:31:32.004585  2662 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50b287c0, len: 13824: Operation not permitted [1]
+W0827 04:31:32.004739  2663 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5070b6c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.004812  2664 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50753540, len: 6912: Operation not permitted [1]
+W0827 04:31:32.144867  2688 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50aed700, len: 6912: Operation not permitted [1]
+W0827 04:31:32.144946  2689 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50792740, len: 6912: Operation not permitted [1]
+W0827 04:31:32.145030  2690 memory_location.cpp:73] Failed to get NUMA node, addr: 0x508fca40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.145135  2691 memory_location.cpp:73] Failed to get NUMA node, addr: 0x505d7c40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.145318  2693 memory_location.cpp:73] Failed to get NUMA node, addr: 0x505ec340, len: 55296: Operation not permitted [1]
+W0827 04:31:32.145387  2692 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50602cc0, len: 55296: Operation not permitted [1]
+W0827 04:31:32.145479  2694 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50748c00, len: 6912: Operation not permitted [1]
+W0827 04:31:32.145632  2695 memory_location.cpp:73] Failed to get NUMA node, addr: 0x507e3480, len: 13824: Operation not permitted [1]
+W0827 04:31:32.145767  2696 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50316400, len: 6912: Operation not permitted [1]
+W0827 04:31:32.145864  2697 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50788600, len: 6912: Operation not permitted [1]
+W0827 04:31:32.177534  2726 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fff3c80, len: 6912: Operation not permitted [1]
+W0827 04:31:32.177623  2728 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fad6780, len: 6912: Operation not permitted [1]
+W0827 04:31:32.177625  2727 memory_location.cpp:73] Failed to get NUMA node, addr: 0x501dc2c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.178279  2729 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50596480, len: 6912: Operation not permitted [1]
+W0827 04:31:32.178463  2730 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50222dc0, len: 55296: Operation not permitted [1]
+W0827 04:31:32.178498  2731 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5038f6c0, len: 55296: Operation not permitted [1]
+W0827 04:31:32.178581  2732 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4ffd7ec0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.178695  2733 memory_location.cpp:73] Failed to get NUMA node, addr: 0x506d84c0, len: 13824: Operation not permitted [1]
+W0827 04:31:32.178884  2734 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5078efc0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.178922  2735 memory_location.cpp:73] Failed to get NUMA node, addr: 0x502c7b00, len: 6912: Operation not permitted [1]
+[2026-08-27 04:31:32] Dummy health check server started in background thread at 0.0.0.0:30000
+[rank10]:[W827 04:32:32.605391940 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank11]:[W827 04:32:32.605815131 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank9]:[W827 04:32:32.606041691 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank8]:[W827 04:32:32.608313727 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank11]:[W827 04:33:19.698792619 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank8]:[W827 04:33:19.699392665 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank9]:[W827 04:33:19.699501937 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank10]:[W827 04:33:19.699589651 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank14]:[W827 04:33:19.728965278 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank12]:[W827 04:33:19.729026815 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank13]:[W827 04:33:19.729092243 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank15]:[W827 04:33:19.730463091 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank14]:[W827 04:33:25.763484084 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank13]:[W827 04:33:25.764518868 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank15]:[W827 04:33:25.764776298 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank12]:[W827 04:33:25.765271843 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[2026-08-27 04:35:10 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.29
+[2026-08-27 04:35:10 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.29
+[2026-08-27 04:35:29 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.39
+[2026-08-27 04:35:29 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.39
+[2026-08-27 04:35:31 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25.86
+[2026-08-27 04:35:31 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25.87
+[2026-08-27 04:43:23 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.14
+[2026-08-27 04:43:23 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.14
+[2026-08-27 04:48:13 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.22
+[2026-08-27 04:48:13 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.22
+[2026-08-27 04:48:14 PP2 TP1 EP1] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP2 TP2 EP2] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP2 TP0 EP0] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP3 TP2 EP2] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP3 TP0 EP0] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP3 TP1 EP1] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP2 TP3 EP3] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP3 TP3 EP3] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:57:14 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.12
+[2026-08-27 04:57:14 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.12
+[2026-08-27 05:04:21 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1088, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2.55
+[2026-08-27 05:04:21 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1088, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2.55
+[2026-08-27 05:04:41 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.19
+[2026-08-27 05:04:41 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.19
+[2026-08-27 05:05:39 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1.10
+[2026-08-27 05:05:39 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1.10
+[2026-08-27 05:07:32 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 72.23
+[2026-08-27 05:07:33 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 72.17
+[2026-08-27 05:07:33 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 78417.87
+[2026-08-27 05:07:33 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 501194.56
+[2026-08-27 05:07:33 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 30385.21
+[2026-08-27 05:07:33 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32725.49
+[2026-08-27 05:07:40 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8673, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1097.27
+[2026-08-27 05:07:40 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8673, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1082.51
+[2026-08-27 05:07:40 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 481, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 68995.34
+[2026-08-27 05:07:40 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 481, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 494490.91
+[2026-08-27 05:07:40 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33832.33
+[2026-08-27 05:07:40 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33008.60
+[2026-08-27 05:07:48 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8656, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1098.69
+[2026-08-27 05:07:48 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8656, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1082.21
+[2026-08-27 05:07:48 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 464, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 62757.93
+[2026-08-27 05:07:48 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 464, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 451735.55
+[2026-08-27 05:07:48 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 30632.65
+[2026-08-27 05:07:48 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 31628.03
+[2026-08-27 05:07:55 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8663, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1101.49
+[2026-08-27 05:07:55 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8663, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1087.16
+[2026-08-27 05:07:55 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 471, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 71370.09
+[2026-08-27 05:07:55 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 471, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 487386.35
+[2026-08-27 05:07:55 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 31311.94
+[2026-08-27 05:07:55 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 31458.64
+[2026-08-27 05:08:03 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8659, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1099.50
+[2026-08-27 05:08:03 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8659, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1082.83
+[2026-08-27 05:08:03 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 467, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 61030.58
+[2026-08-27 05:08:03 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 467, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 440213.50
+[2026-08-27 05:08:03 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33725.14
+[2026-08-27 05:08:03 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32738.01
+[2026-08-27 05:08:10 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8692, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1100.46
+[2026-08-27 05:08:11 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8692, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1085.28
+[2026-08-27 05:08:11 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 500, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 67432.08
+[2026-08-27 05:08:11 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 500, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 480206.29
+[2026-08-27 05:08:11 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32427.60
+[2026-08-27 05:08:11 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 30441.07
+[2026-08-27 05:08:18 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8557, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1101.25
+[2026-08-27 05:08:18 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8557, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1089.66
+[2026-08-27 05:08:18 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 365, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 85531.52
+[2026-08-27 05:08:18 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 365, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 467687.86
+[2026-08-27 05:08:18 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 22103.80
+[2026-08-27 05:08:18 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25225.78
+[2026-08-27 05:08:26 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8683, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1099.06
+[2026-08-27 05:08:26 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8683, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1084.12
+[2026-08-27 05:08:26 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 491, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 68454.69
+[2026-08-27 05:08:26 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 491, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 485556.59
+[2026-08-27 05:08:26 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33944.36
+[2026-08-27 05:08:26 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 30281.29
+[2026-08-27 05:11:45 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.32
+[2026-08-27 05:11:45 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.32
+[2026-08-27 05:14:42 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1088, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6.16
+[2026-08-27 05:14:42 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1088, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6.16
+[2026-08-27 05:17:47 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 44.24
+[2026-08-27 05:17:47 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 44.21
+[2026-08-27 05:17:47 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 74322.16
+[2026-08-27 05:17:47 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 505572.40
+[2026-08-27 05:17:47 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 29824.24
+[2026-08-27 05:17:47 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32304.11
+[2026-08-27 05:17:55 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8673, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1100.05
+[2026-08-27 05:17:55 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8673, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1085.77
+[2026-08-27 05:17:55 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 481, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 72167.09
+[2026-08-27 05:17:55 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 481, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 477186.86
+[2026-08-27 05:17:55 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 30671.49
+[2026-08-27 05:17:55 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 34433.71
+[2026-08-27 05:18:02 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8656, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1101.74
+[2026-08-27 05:18:02 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8656, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1087.80
+[2026-08-27 05:18:02 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 464, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 74681.51
+[2026-08-27 05:18:02 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 464, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 550093.83
+[2026-08-27 05:18:02 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33172.26
+[2026-08-27 05:18:02 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32409.88
+[2026-08-27 05:18:10 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8663, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1101.73
+[2026-08-27 05:18:10 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8663, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1087.58
+[2026-08-27 05:18:10 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 471, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 73073.93
+[2026-08-27 05:18:10 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 471, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 528029.69
+[2026-08-27 05:18:10 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32561.99
+[2026-08-27 05:18:10 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 31690.72
+[2026-08-27 05:18:17 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8659, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1099.69
+[2026-08-27 05:18:17 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8659, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1085.59
+[2026-08-27 05:18:17 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 467, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 72775.07
+[2026-08-27 05:18:17 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 467, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 520101.00
+[2026-08-27 05:18:17 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 30837.63
+[2026-08-27 05:18:18 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32670.85
+[2026-08-27 05:18:25 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8692, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1100.48
+[2026-08-27 05:18:25 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8692, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1086.22
+[2026-08-27 05:18:25 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 500, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 72864.89
+[2026-08-27 05:18:25 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 500, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 524671.87
+[2026-08-27 05:18:25 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 31161.52
+[2026-08-27 05:18:25 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 35156.96
+[2026-08-27 05:18:32 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8557, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1102.45
+[2026-08-27 05:18:33 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8557, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1090.72
+[2026-08-27 05:18:33 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 365, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 85680.60
+[2026-08-27 05:18:33 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 365, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 482538.50
+[2026-08-27 05:18:33 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 24798.12
+[2026-08-27 05:18:33 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 23973.27
+[2026-08-27 05:18:40 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8683, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1102.03
+[2026-08-27 05:18:40 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8683, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1087.43
+[2026-08-27 05:18:40 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 491, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 70828.52
+[2026-08-27 05:18:40 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 491, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 529856.70
+[2026-08-27 05:18:40 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32426.64
+[2026-08-27 05:18:40 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33626.30
+[2026-08-27 05:21:44 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 2, #pending-token: 8652, #bootstrap-req: 2, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 44.47
+[2026-08-27 05:21:45 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 3, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9902.45
+[2026-08-27 05:21:45 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 2, #pending-token: 8652, #bootstrap-req: 2, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 44.27
+[2026-08-27 05:21:46 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 3, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8141.53
+[2026-08-27 05:21:46 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 3, #pending-token: 9185, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7907.49
+[2026-08-27 05:21:47 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 3, #pending-token: 9185, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10345.23
+[2026-08-27 05:21:47 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 2, #pending-token: 993, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9860.59
+[2026-08-27 05:21:48 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 2, #pending-token: 993, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8351.60
+[2026-08-27 05:21:48 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 2, #pending-token: 9680, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8645.66
+[2026-08-27 05:21:49 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 2, #pending-token: 9680, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10382.09
+[2026-08-27 05:21:49 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1488, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10477.17
+[2026-08-27 05:21:50 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1488, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8583.82
+[2026-08-27 05:21:50 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 10199, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8562.77
+[2026-08-27 05:21:51 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 18858, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9978.66
+[2026-08-27 05:21:51 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 10199, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9847.23
+[2026-08-27 05:21:51 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 18858, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9275.72
+[2026-08-27 05:21:51 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 27591, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9156.16
+[2026-08-27 05:21:52 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 53023, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10350.08
+[2026-08-27 05:21:52 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 27591, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9601.22
+[2026-08-27 05:21:53 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 44876, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9145.49
+[2026-08-27 05:21:53 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 53023, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9622.28
+[2026-08-27 05:21:54 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 36684, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9799.45
+[2026-08-27 05:21:54 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 44876, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9677.55
+[2026-08-27 05:21:55 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 28504, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10333.04
+[2026-08-27 05:21:55 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 36684, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9394.13
+[2026-08-27 05:21:56 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 28504, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10860.67
+[2026-08-27 05:21:56 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 20312, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9527.15
+[2026-08-27 05:21:57 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 20312, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9357.72
+[2026-08-27 05:21:57 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 12139, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9361.75
+[2026-08-27 05:21:57 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 12139, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 15465.96
+[2026-08-27 05:21:57 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 3947, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 15452.41
+[2026-08-27 05:21:57 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 3947, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 310446.55
+[2026-08-27 05:21:57 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 3968, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 156702.42
+[2026-08-27 05:21:57 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 3968, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 139513.02
+[2026-08-27 05:22:04 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 0, #pending-token: 8608, #bootstrap-req: 4, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1132.42
+[2026-08-27 05:22:05 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 416, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10805.74
+[2026-08-27 05:22:05 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 0, #pending-token: 8608, #bootstrap-req: 4, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1019.78
+[2026-08-27 05:22:06 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 416, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7895.05
+[2026-08-27 05:22:06 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 9149, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7392.79
+[2026-08-27 05:22:07 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 957, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11340.60
+[2026-08-27 05:22:07 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 9149, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11099.21
+[2026-08-27 05:22:08 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 957, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8063.56
+[2026-08-27 05:22:08 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 9638, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7911.72
+[2026-08-27 05:22:09 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1446, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11076.15
+[2026-08-27 05:22:09 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 9638, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10949.54
+[2026-08-27 05:22:10 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1446, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8408.34
+[2026-08-27 05:22:10 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 10033, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8234.48
+[2026-08-27 05:22:10 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1841, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10913.26
+[2026-08-27 05:22:10 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 10033, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10431.88
+[2026-08-27 05:22:11 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1856, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2259.05
+[2026-08-27 05:22:11 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1841, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9802.26
+[2026-08-27 05:22:12 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8540, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8332.93
+[2026-08-27 05:22:12 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1856, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1872.96
+[2026-08-27 05:22:13 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 17066, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9894.47
+[2026-08-27 05:22:13 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8540, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10158.45
+[2026-08-27 05:22:14 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8910, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8389.00
+[2026-08-27 05:22:14 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 17066, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8419.18
+[2026-08-27 05:22:15 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 17466, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10331.84
+[2026-08-27 05:22:15 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8910, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10077.68
+[2026-08-27 05:22:16 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9324, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9373.73
+[2026-08-27 05:22:16 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 17466, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8567.43
+[2026-08-27 05:22:16 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 17751, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10615.28
+[2026-08-27 05:22:17 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9324, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11315.42
+[2026-08-27 05:22:17 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 17751, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8963.12
+[2026-08-27 05:22:17 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9579, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8324.62
+[2026-08-27 05:22:18 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1387, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 35673.33
+[2026-08-27 05:22:18 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9579, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 35404.74
+[2026-08-27 05:22:18 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1387, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 314704.56
+[2026-08-27 05:22:18 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1408, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 52789.70
+[2026-08-27 05:22:18 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1408, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 13474.13
+[2026-08-27 05:22:25 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8547, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1143.27
+[2026-08-27 05:22:26 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 355, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10109.37
+[2026-08-27 05:22:26 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8547, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1034.06
+[2026-08-27 05:22:27 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 355, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7893.67
+[2026-08-27 05:22:27 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8932, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7504.02
+[2026-08-27 05:22:28 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 740, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11535.13
+[2026-08-27 05:22:28 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8932, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10838.23
+[2026-08-27 05:22:29 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 740, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8155.37
+[2026-08-27 05:22:29 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9259, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7789.29
+[2026-08-27 05:22:29 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1067, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10733.11
+[2026-08-27 05:22:29 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9259, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10388.64
+[2026-08-27 05:22:30 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1067, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8544.20
+[2026-08-27 05:22:30 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9745, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8357.94
+[2026-08-27 05:22:31 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1553, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10706.17
+[2026-08-27 05:22:31 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9745, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10445.99
+[2026-08-27 05:22:32 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1600, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1952.39
+[2026-08-27 05:22:32 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1553, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9931.31
+[2026-08-27 05:22:33 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8602, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8370.54
+[2026-08-27 05:22:33 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1600, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1624.80
+[2026-08-27 05:22:34 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 410, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9785.83
+[2026-08-27 05:22:34 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8602, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9969.67
+[2026-08-27 05:22:35 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8696, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8591.23
+[2026-08-27 05:22:35 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 410, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8607.27
+[2026-08-27 05:22:35 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 504, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10402.38
+[2026-08-27 05:22:35 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8696, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10102.69
+[2026-08-27 05:22:36 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9164, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9308.92
+[2026-08-27 05:22:36 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 504, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8520.94
+[2026-08-27 05:22:37 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 972, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10664.47
+[2026-08-27 05:22:37 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9164, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11299.50
+[2026-08-27 05:22:38 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 972, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8956.53
+[2026-08-27 05:22:38 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9322, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8269.42
+[2026-08-27 05:22:38 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1130, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 43785.83
+[2026-08-27 05:22:38 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9322, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 43114.10
+[2026-08-27 05:22:38 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1130, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 348034.30
+[2026-08-27 05:22:38 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1152, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 43670.18
+[2026-08-27 05:22:38 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1152, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10950.18
+[2026-08-27 05:22:45 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8602, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1146.51
+[2026-08-27 05:22:46 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 410, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10113.13
+[2026-08-27 05:22:46 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8602, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1036.82
+[2026-08-27 05:22:47 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 410, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8059.88
+[2026-08-27 05:22:47 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 9105, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7668.92
+[2026-08-27 05:22:48 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 913, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11406.02
+[2026-08-27 05:22:48 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 9105, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11028.76
+[2026-08-27 05:22:49 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 913, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8034.52
+[2026-08-27 05:22:49 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9375, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7803.47
+[2026-08-27 05:22:50 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1183, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10701.85
+[2026-08-27 05:22:50 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9375, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10338.95
+[2026-08-27 05:22:51 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1183, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8629.50
+[2026-08-27 05:22:51 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9868, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8448.89
+[2026-08-27 05:22:52 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1676, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10696.20
+[2026-08-27 05:22:52 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9868, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10383.50
+[2026-08-27 05:22:52 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2088.70
+[2026-08-27 05:22:52 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1676, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9980.67
+[2026-08-27 05:22:53 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8467, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8474.75
+[2026-08-27 05:22:53 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1765.75
+[2026-08-27 05:22:54 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 275, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10047.97
+[2026-08-27 05:22:54 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8467, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10138.75
+[2026-08-27 05:22:55 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8988, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8471.06
+[2026-08-27 05:22:55 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 275, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8570.54
+[2026-08-27 05:22:56 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 796, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10303.51
+[2026-08-27 05:22:56 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8988, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9711.56
+[2026-08-27 05:22:57 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9363, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9482.47
+[2026-08-27 05:22:57 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 796, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8960.80
+[2026-08-27 05:22:58 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 1171, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10488.36
+[2026-08-27 05:22:58 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9363, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11295.90
+[2026-08-27 05:22:59 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 1171, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8985.86
+[2026-08-27 05:22:59 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9880, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8497.44
+[2026-08-27 05:22:59 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1688, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 30171.97
+[2026-08-27 05:22:59 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9880, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 29961.56
+[2026-08-27 05:22:59 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1688, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 384839.25
+[2026-08-27 05:22:59 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 77829.47
+[2026-08-27 05:22:59 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 17255.50
+[2026-08-27 05:23:06 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8601, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1148.36
+[2026-08-27 05:23:07 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 409, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10091.44
+[2026-08-27 05:23:07 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8601, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1039.48
+[2026-08-27 05:23:08 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 409, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8052.72
+[2026-08-27 05:23:08 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 9104, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7736.57
+[2026-08-27 05:23:09 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 912, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11209.86
+[2026-08-27 05:23:09 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 9104, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10701.52
+[2026-08-27 05:23:10 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 912, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8193.31
+[2026-08-27 05:23:10 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9485, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7925.16
+[2026-08-27 05:23:10 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1293, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10717.69
+[2026-08-27 05:23:10 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9485, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10383.19
+[2026-08-27 05:23:11 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1293, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8591.94
+[2026-08-27 05:23:11 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9900, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8353.36
+[2026-08-27 05:23:12 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1708, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10835.80
+[2026-08-27 05:23:12 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9900, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10393.61
+[2026-08-27 05:23:13 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2099.70
+[2026-08-27 05:23:13 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1708, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9919.38
+[2026-08-27 05:23:14 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8525, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8406.32
+[2026-08-27 05:23:14 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1756.57
+[2026-08-27 05:23:15 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 333, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10057.88
+[2026-08-27 05:23:15 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8525, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9966.29
+[2026-08-27 05:23:16 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8909, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8449.44
+[2026-08-27 05:23:16 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 333, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8768.58
+[2026-08-27 05:23:17 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 717, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10363.27
+[2026-08-27 05:23:17 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8909, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9712.12
+[2026-08-27 05:23:17 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9452, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9349.89
+[2026-08-27 05:23:17 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 717, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8950.85
+[2026-08-27 05:23:18 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 1260, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10634.96
+[2026-08-27 05:23:18 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9452, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11269.38
+[2026-08-27 05:23:19 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 1260, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9067.98
+[2026-08-27 05:23:19 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9830, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8519.49
+[2026-08-27 05:23:19 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1638, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33116.32
+[2026-08-27 05:23:19 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9830, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32958.23
+[2026-08-27 05:23:19 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1638, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 385069.93
+[2026-08-27 05:23:19 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1664, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 72026.72
+[2026-08-27 05:23:19 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1664, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 77715.90
+[2026-08-27 05:32:52 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.11
+[2026-08-27 05:32:52 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.11
+[2026-08-27 05:35:54 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 45.18
+[2026-08-27 05:35:54 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 45.15
+[2026-08-27 05:35:54 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 72669.47
+[2026-08-27 05:35:54 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 516710.98
+[2026-08-27 05:35:54 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32925.10
+[2026-08-27 05:35:54 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 28289.90
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-20260827-1325/logs/manual_p_node2.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-20260827-1325/logs/manual_p_node2.log
new file mode 100644
index 0000000..a5dd065
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-20260827-1325/logs/manual_p_node2.log
@@ -0,0 +1,1060 @@
+/usr/local/lib/python3.12/dist-packages/torchao/quantization/quant_api.py:1731: SyntaxWarning: invalid escape sequence '\.'
+  """Configuration class for applying different quantization configs to modules or parameters based on their fully qualified names (FQNs).
+/sgl-workspace/sglang/python/sglang/launch_server.py:56: UserWarning: 'python -m sglang.launch_server' is still supported, but 'sglang serve' is the recommended entrypoint.
+  Example: sglang serve --model-path <model> [options]
+  warnings.warn(
+[2026-08-27 04:28:43] Breakable CUDA graph is incompatible with MLA attention; disabling prefill CUDA graph.
+[2026-08-27 04:28:43] Attention backend not specified. Use triton backend by default.
+[2026-08-27 04:28:43] Pipeline parallelism is incompatible with overlap schedule.
+[2026-08-27 04:28:45] server_args=ServerArgs(model_path='/data/hf_models/Kimi-K3', tokenizer_path='/data/hf_models/Kimi-K3', tokenizer_mode='auto', tokenizer_backend='huggingface', tokenizer_worker_num=1, detokenizer_worker_num=1, skip_tokenizer_init=False, load_format='auto', model_loader_extra_config='{}', trust_remote_code=True, context_length=None, is_embedding=False, enable_multimodal=None, revision=None, model_impl='auto', model_config_parser='auto', json_model_override_args='{}', dtype='auto', quantization=None, quantization_param_path=None, kv_cache_dtype='auto', enable_fp32_lm_head=False, modelopt_quant=None, modelopt_checkpoint_restore_path=None, modelopt_checkpoint_save_path=None, modelopt_export_path=None, quantize_and_serve=False, rl_quant_profile=None, enable_tf32_matmul=False, mem_fraction_static=0.88, max_running_requests=None, max_queued_requests=None, max_total_tokens=None, chunked_prefill_size=8192, enable_dynamic_chunking=False, max_prefill_tokens=16384, prefill_max_requests=None, schedule_policy='fcfs', enable_priority_scheduling=False, disable_priority_preemption=False, default_priority_value=None, abort_on_priority_when_disabled=False, schedule_low_priority_values_first=False, priority_scheduling_preemption_threshold=10, retraction_policy='length', schedule_conservativeness=1.0, page_size=64, swa_full_tokens_ratio=0.8, disable_hybrid_swa_memory=False, radix_eviction_policy='lru', prefill_only_disable_kv_cache=False, disable_radix_cache=True, enable_page_major_kv_layout=False, enable_unified_memory=False, disable_chunked_prefix_cache=False, disable_overlap_schedule=True, num_continuous_decode_steps=1, scheduler_recv_interval=1, enable_mixed_chunk=False, nccl_port=None, dist_timeout=3600, dist_init_addr='174.1.60.1:20000', nnodes=4, node_rank=2, tp_size=4, dcp_size=1, pp_size=8, pp_max_micro_batch_size=None, pp_async_batch_depth=0, dp_size=1, load_balance_method='follow_bootstrap_room', attn_cp_size=1, moe_dp_size=1, dcp_comm_backend='ag_rs', dcp_replicate_q_proj=None, dwdp_size=1, enable_prefill_cp=False, cp_strategy=None, enable_dsa_cache_layer_split=False, enable_dsa_prefill_context_parallel=False, dsa_prefill_cp_mode='round-robin-split', enable_prefill_context_parallel=False, prefill_cp_mode='in-seq-split', enable_cp_decode_attn_tp=False, enable_dp_attention=False, enable_dp_attention_local_control_broadcast=False, enable_dp_lm_head=False, enable_attn_tp_input_scattered=False, disable_attn_tp_gather=False, enable_p2p_check=False, device='cuda', base_gpu_id=0, gpu_id_step=1, random_seed=336932344, watchdog_timeout=300, soft_watchdog_timeout=None, sleep_on_idle=False, use_ray=False, custom_sigquit_handler=None, numa_node=None, gc_threshold=None, host='0.0.0.0', port=30000, fastapi_root_path='', smg_grpc_mode=False, grpc_mode=False, grpc_port=None, sidecar=None, sidecar_args=None, skip_server_warmup=False, warmups=None, enable_http2=False, ssl_keyfile=None, ssl_certfile=None, ssl_ca_certs=None, ssl_keyfile_password=None, enable_ssl_refresh=False, api_key=None, admin_api_key=None, served_model_name='kimi-k3', weight_version='default', chat_template=None, hf_chat_template_name=None, completion_template=None, file_storage_path='sglang_storage', enable_cache_report=False, reasoning_parser=None, default_chat_template_kwargs=None, strip_thinking_cache=False, enable_strict_thinking=False, tool_call_parser=None, tool_server=None, sampling_defaults='model', asr_max_buffer_seconds=60, asr_max_concurrent_sessions=32, preferred_sampling_params=None, allow_auto_truncate=False, stream_interval=1, batch_notify_size=16, stream_response_default_include_usage=False, incremental_streaming_output=False, enable_streaming_session=False, enable_session_radix_cache=False, log_level='info', log_level_http=None, log_requests=False, log_requests_level=2, log_requests_format='text', log_requests_target=None, uvicorn_access_log_exclude_prefixes=[], crash_dump_folder=None, show_time_cost=False, enable_metrics=False, smg_http_sidecar_port=None, enable_mfu_metrics=False, enable_metrics_for_all_schedulers=False, load_snapshot_publish_interval=15, tokenizer_metrics_custom_labels_header='x-custom-labels', tokenizer_metrics_allowed_custom_labels=None, extra_metric_labels=None, bucket_time_to_first_token=None, bucket_inter_token_latency=None, bucket_e2e_request_latency=None, prompt_tokens_buckets=None, generation_tokens_buckets=None, gc_warning_threshold_secs=0.0, decode_log_interval=40, enable_request_time_stats_logging=False, kv_events_config=None, enable_forward_pass_metrics=False, forward_pass_metrics_worker_id='', forward_pass_metrics_ipc_name=None, enable_trace=False, trace_modules='request', otlp_traces_endpoint='localhost:4317', export_metrics_to_file=False, export_metrics_to_file_dir=None, stat_loggers=None, constrained_json_whitespace_pattern=None, constrained_json_disable_any_whitespace=False, attention_backend='triton', decode_attention_backend=None, prefill_attention_backend=None, sampling_backend='flashinfer', grammar_backend='xgrammar', radix_cache_backend=None, mm_attention_backend=None, fp8_gemm_runner_backend='auto', fp4_gemm_runner_backend='auto', bf16_gemm_backend='auto', dsa_prefill_backend=None, dsa_decode_backend=None, dsa_paged_mqa_logits_backend='auto', dsa_topk_backend='sgl-kernel', disable_flashinfer_autotune=False, flashinfer_autotune_skip_ops=None, mamba_backend='triton', cuda_graph_config=CudaGraphConfig(decode=PhaseConfig(backend='disabled', max_bs=16, bs=[1, 2, 4, 8, 12, 16], tc_compiler='eager', full_prefill_max_req=None), prefill=PhaseConfig(backend='disabled', max_bs=2048, bs=[4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], tc_compiler='eager', full_prefill_max_req=None)), cuda_graph_backend_decode=None, cuda_graph_backend_prefill=None, cuda_graph_max_bs_decode=16, cuda_graph_max_bs_prefill=None, cuda_graph_bs_decode=None, cuda_graph_bs_prefill=None, cuda_graph_tc_compiler=None, disable_prefill_cuda_graph=False, disable_decode_cuda_graph=False, disable_cuda_graph=False, disable_cuda_graph_padding=False, enable_profile_cuda_graph=False, enable_cudagraph_gc=False, debug_cuda_graph=False, enable_layerwise_nvtx_marker=False, enable_nccl_nvls=False, enable_symm_mem=False, triton_attention_reduce_in_fp32=False, triton_attention_num_kv_splits=8, triton_attention_split_tile_size=None, flashinfer_mla_disable_ragged=False, enable_fused_qk_norm_rope=False, enable_precise_embedding_interpolation=False, enable_fused_moe_sum_all_reduce=False, enable_deepseek_v4_fp4_indexer=False, disable_custom_all_reduce=False, enable_mscclpp=False, enable_torch_symm_mem=False, enable_scattered_sconv=False, pre_warm_nccl=False, enable_quant_communications=False, enable_flashinfer_allreduce_fusion=False, enforce_disable_flashinfer_allreduce_fusion=False, flashinfer_allreduce_fusion_backend=None, enable_aiter_allreduce_fusion=False, enable_torch_compile=False, enable_torch_compile_debug_mode=False, torch_compile_max_bs=32, torchao_config='', speculative_algorithm=None, speculative_draft_model_path=None, speculative_draft_model_revision=None, speculative_draft_load_format=None, speculative_num_steps=None, speculative_eagle_topk=None, speculative_num_draft_tokens=None, speculative_dflash_block_size=None, speculative_dspark_block_size=None, speculative_dspark_sps_table_path=None, speculative_dspark_confidence_sts_path=None, speculative_dspark_align_verify_tokens_to_graph_tier=False, speculative_accept_threshold_single=1.0, speculative_accept_threshold_acc=1.0, speculative_use_rejection_sampling=False, speculative_token_map=None, speculative_attention_mode='prefill', speculative_draft_attention_backend=None, speculative_draft_window_size=None, speculative_moe_runner_backend='flashinfer_mxfp4', speculative_moe_a2a_backend=None, speculative_draft_model_quantization=None, speculative_skip_dp_mlp_sync=False, enable_multi_layer_eagle=False, speculative_adaptive=False, speculative_adaptive_config=None, decoupled_spec_bind_endpoint=None, decoupled_spec_connect_endpoints=None, decoupled_spec_rank=None, decoupled_spec_role='null', spec_trace_dir=None, speculative_ngram_min_bfs_breadth=1, speculative_ngram_max_bfs_breadth=10, speculative_ngram_match_type='BFS', speculative_ngram_max_trie_depth=18, speculative_ngram_capacity=10000000, speculative_ngram_external_corpus_path=None, speculative_ngram_external_sam_budget=0, speculative_ngram_external_corpus_max_tokens=10000000, ep_size=4, moe_a2a_backend='none', moe_runner_backend='flashinfer_mxfp4', flashinfer_mxfp4_moe_precision='default', deepep_mode='auto', fuseep_mode=2, deepep_dispatcher_output_dtype='auto', ep_num_redundant_experts=0, ep_dispatch_algorithm=None, init_expert_location='trivial', enable_eplb=False, eplb_algorithm='auto', eplb_rebalance_num_iterations=1000, eplb_rebalance_layers_per_chunk=None, eplb_min_rebalancing_utilization_threshold=1.0, expert_distribution_recorder_mode=None, expert_distribution_recorder_buffer_size=1000, enable_expert_distribution_metrics=False, deepep_config=None, moe_dense_tp_size=None, elastic_ep_backend=None, enable_elastic_expert_backup=False, mooncake_ib_device=None, enable_waterfill=False, ep_join_mode=None, ep_join_rank_offset=0, elastic_ep_initial_size=None, max_ep_size=None, elastic_ep_scale_timeout=600, elastic_ep_rejoin=False, disable_flashinfer_cutlass_moe_fp4_allgather=False, disable_shared_experts_fusion=False, enforce_shared_experts_fusion=False, max_mamba_cache_size=None, mamba_ssm_dtype=None, mamba_max_states_per_path=-1, enable_mamba_cache_stochastic_rounding=False, mamba_cache_philox_rounds=0, mamba_full_memory_ratio=0.36, mamba_radix_cache_strategy='extra_buffer_lazy', uses_mamba_radix_cache=False, mamba_track_interval=256, enable_int8_mamba_checkpoint=False, int8_mamba_ckpt_size=None, linear_attn_backend='triton', linear_attn_decode_backend=None, linear_attn_prefill_backend=None, linear_attn_verify_backend=None, enable_linear_replayssm=False, linear_replayssm_cache_len=16, enable_linear_replayssm_spec=False, enable_hierarchical_cache=False, hicache_ratio=2.0, hicache_size=0, hicache_write_policy='write_through', hicache_io_backend='kernel', hicache_mem_layout='page_first', hicache_storage_backend=None, hicache_storage_prefetch_policy='timeout', hicache_storage_backend_extra_config=None, enable_hisparse=False, hisparse_config=None, enable_broadcast_mm_inputs_process=False, enable_prefix_mm_cache=False, mm_enable_dp_encoder=False, mm_process_config={}, mm_processor_worker_num=0, mm_io_worker_num=0, limit_mm_data_per_request=None, enable_mm_global_cache=False, disable_fast_image_processor=False, mm_feature_transport='cpu', keep_mm_feature_on_device=False, enable_lora=None, enable_lora_overlap_loading=None, max_lora_rank=None, lora_target_modules=None, lora_paths=None, max_loaded_loras=None, max_loras_per_batch=8, lora_eviction_policy='lru', lora_backend='csgmv', max_lora_chunk_size=16, experts_shared_outer_loras=None, lora_use_virtual_experts=False, lora_strict_loading=False, lora_drain_wait_threshold=0.0, enable_two_batch_overlap=False, enable_single_batch_overlap=False, tbo_token_distribution_threshold=0.48, cpu_offload_gb=0, offload_group_size=-1, offload_num_in_group=1, offload_prefetch_step=1, offload_mode='cpu', enable_lmcache=False, lmcache_config_file=None, enable_flexkv=False, flexkv_config_file=None, kt_weight_path=None, kt_method='AMXINT4', kt_cpuinfer=None, kt_threadpool_count=2, kt_num_gpu_experts=None, kt_max_deferred_experts_per_token=None, dllm_algorithm=None, dllm_algorithm_config=None, dllm_fdfo=True, disaggregation_mode='prefill', disaggregation_transfer_backend='mooncake', disaggregation_bootstrap_port=28800, disaggregation_ib_device='mlx5_0,mlx5_1,mlx5_2,mlx5_3', disaggregation_decode_enable_radix_cache=False, disaggregation_decode_enable_offload_kvcache=False, num_reserved_decode_tokens=512, disaggregation_decode_extra_slots=None, disaggregation_decode_polling_interval=1, optimistic_prefill_attempts=0, encoder_only=False, language_only=False, encoder_transfer_backend='zmq_to_tokenizer', encoder_urls=[], encoder_bootstrap_port=8997, encoder_register_urls=[], enable_adaptive_dispatch_to_encoder=False, enable_pdmux=False, pdmux_config_path=None, sm_group_num=8, custom_weight_loader=[], weight_loader_disable_mmap=False, weight_loader_prefetch_checkpoints=False, weight_loader_prefetch_num_threads=4, weight_loader_drop_cache_after_load=False, remote_instance_weight_loader_seed_instance_ip=None, remote_instance_weight_loader_seed_instance_service_port=None, remote_instance_weight_loader_send_weights_group_ports=None, remote_instance_weight_loader_backend='nccl', remote_instance_weight_loader_start_seed_via_transfer_engine=False, engine_info_bootstrap_port=6789, modelexpress_config=None, download_dir=None, model_checksum=None, delete_ckpt_after_loading=False, decrypted_config_file=None, decrypted_draft_config_file=None, checkpoint_engine_wait_weights_before_ready=False, enable_prefill_delayer=False, prefill_delayer_max_delay_passes=30, prefill_delayer_token_usage_low_watermark=None, prefill_delayer_forward_passes_buckets=None, prefill_delayer_wait_seconds_buckets=None, prefill_delayer_queue_min_ratio=None, prefill_delayer_max_delay_ms=None, min_free_slots_delay=None, enable_deterministic_inference=False, rl_on_policy_target=None, kv_canary='none', kv_canary_real_data='none', kv_canary_sweep_interval=0, enable_dynamic_batch_tokenizer=False, dynamic_batch_tokenizer_batch_size=32, dynamic_batch_tokenizer_batch_timeout=0.002, enable_tokenizer_batch_encode=False, disable_tokenizer_batch_decode=False, debug_tensor_dump_output_folder=None, debug_tensor_dump_layers=None, debug_tensor_dump_input_file=None, enable_memory_saver=False, enable_weights_cpu_backup=False, enable_draft_weights_cpu_backup=False, enable_custom_logit_processor=False, enable_return_hidden_states=False, enable_return_routed_experts=False, enable_return_indexer_topk=False, disable_outlines_disk_cache=False, enable_mis=False, weight_cache_mode='off', weight_cache_socket=None, weight_cache_timeout=1800, forward_hooks=None, msprobe_dump_config=None)
+[2026-08-27 04:28:45] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:45] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:45] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:45] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:45] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:45] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:45] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:45] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:01 PP4 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:01 PP4 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:01 PP5 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:01 PP5 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:01 PP4 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:01 PP4 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:01 PP5 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:01 PP5 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:01 PP4 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:01 PP4 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:01 PP4 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:01 PP4 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:02 PP5 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:02 PP5 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:02 PP5 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:02 PP5 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:03 PP5 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:03 PP5 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:03 PP4 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:03 PP4 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:03 PP5 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:03 PP5 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:03 PP4 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:03 PP4 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:05.195727   240 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:05.195828   240 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:05.195843   240 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.3 port: 12001
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:05.195847   236 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:05.195899   240 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.3:15652
+I0827 04:29:05.195935   236 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:05.195948   236 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.3 port: 12001
+I0827 04:29:05.195989   236 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.3:15032
+I0827 04:29:05.196039   240 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:05.196110   236 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:05.206050   241 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:05.206164   241 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:05.206182   241 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.3 port: 12001
+I0827 04:29:05.206233   241 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.3:16487
+I0827 04:29:05.206414   241 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:05.206894   238 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:05.207010   238 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:05.207029   238 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.3 port: 12001
+I0827 04:29:05.207082   238 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.3:15213
+I0827 04:29:05.207248   238 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:05.213657   238 topology.cpp:127] Device mlx5_0 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:05.215921   237 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:05.216045   237 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:05.216063   237 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.3 port: 12001
+I0827 04:29:05.216125   237 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.3:16450
+I0827 04:29:05.216308   237 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:05.219630   235 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:05.219745   235 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:05.219763   235 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.3 port: 12001
+I0827 04:29:05.219815   235 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.3:15658
+I0827 04:29:05.219810   236 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:05.220000   235 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:05.222787   237 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:05.223301   240 topology.cpp:127] Device mlx5_0 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:05.223796   239 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+W0827 04:29:05.223798   242 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:05.223948   242 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:05.223948   239 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:05.223969   242 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.3 port: 12001
+I0827 04:29:05.223973   239 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.3 port: 12001
+I0827 04:29:05.224030   242 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.3:16182
+I0827 04:29:05.224030   239 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.3:15581
+I0827 04:29:05.224237   242 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:05.224242   239 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:05.226227   235 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:05.227162   236 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:05.228888   237 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:05.229261   240 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:05.232141   235 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:05.232787   239 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:05.232861   242 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:05.233434   241 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:05.233938   236 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:05.236400   237 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:05.236641   240 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:05.238986   235 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:05.241036   236 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:05.241355   238 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:05.241638   241 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:05.242061   239 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:05.242151   242 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:05.244179   236 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:05.244223   236 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:05.244287   237 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:05.244297   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.244899   240 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:05.245494   235 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:05.249099   237 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:05.249130   237 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:05.249187   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.251125   235 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:05.251137   240 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:05.251168   235 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:05.251168   240 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:05.251241   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.251241   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.251624   239 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:05.251729   242 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:05.251823   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:05.253651   236 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:03
+I0827 04:29:05.253687   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.256857   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:05.257697   239 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:05.258617   237 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:03
+I0827 04:29:05.258652   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.258757   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:05.258822   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:05.260108   239 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:05.260136   239 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:05.260191   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.260488   235 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:03
+I0827 04:29:05.260502   240 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:03
+I0827 04:29:05.260530   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.260533   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.262187   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:05.264094   236 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:03
+I0827 04:29:05.264137   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.266639   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:05.268234   237 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:03
+I0827 04:29:05.268261   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.268549   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:05.269187   241 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:05.269232   238 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:05.270207   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:05.270285   240 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:03
+I0827 04:29:05.270316   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.271827   235 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:03
+I0827 04:29:05.271884   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.275274   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:05.276836   238 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:05.276907   242 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:05.276996   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:05.277058   236 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:03
+I0827 04:29:05.277101   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.278084   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:05.278632   237 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:03
+I0827 04:29:05.278667   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.279789   240 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:03
+I0827 04:29:05.279817   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.282152   238 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:05.282193   238 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:05.282260   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.283128   242 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:05.283200   242 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:05.283263   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.286593   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:05.287469   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:05.288283   239 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:03
+I0827 04:29:05.288322   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.289064   240 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:03
+I0827 04:29:05.289194   240 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:05 PP5 TP1 EP1] Init torch distributed begin.
+I0827 04:29:05.290841   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:05.292457   242 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:03
+I0827 04:29:05.292488   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.292999   241 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:05.296203   241 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:05.296272   241 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:05.296341   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.300206   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:05.301280   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:05.301775   242 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:03
+I0827 04:29:05.301801   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.303004   235 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:03
+I0827 04:29:05.303042   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.303999   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:05.305142   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:05.305728   241 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:03
+I0827 04:29:05.305768   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.305974   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:05.306758   237 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:03
+I0827 04:29:05.306921   237 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:05 PP4 TP2 EP2] Init torch distributed begin.
+I0827 04:29:05.307497   236 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:03
+I0827 04:29:05.307627   236 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:05 PP4 TP1 EP1] Init torch distributed begin.
+I0827 04:29:05.309624   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:05.309773   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:05.311504   238 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:03
+I0827 04:29:05.311542   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.311590   242 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:03
+I0827 04:29:05.311633   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.313076   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:05.314762   239 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:03
+I0827 04:29:05.314813   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.319522   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:05.319566   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:05.321249   242 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:03
+I0827 04:29:05.321398   242 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:05 PP5 TP3 EP3] Init torch distributed begin.
+I0827 04:29:05.326727   238 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:03
+I0827 04:29:05.326759   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.329339   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:05.331007   235 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:03
+I0827 04:29:05.331142   235 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:05 PP4 TP0 EP0] Init torch distributed begin.
+I0827 04:29:05.334666   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:05.336285   238 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:03
+I0827 04:29:05.336313   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.337296   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:05.338828   241 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:03
+I0827 04:29:05.338877   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.344906   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:05.346731   239 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:03
+I0827 04:29:05.346776   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.365299   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:05.366928   238 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:03
+I0827 04:29:05.367062   238 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:05 PP4 TP3 EP3] Init torch distributed begin.
+I0827 04:29:05.368464   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:05.369644   241 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:03
+I0827 04:29:05.369663   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.377357   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:05.379019   239 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:03
+I0827 04:29:05.379146   239 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:05 PP5 TP0 EP0] Init torch distributed begin.
+I0827 04:29:05.396831   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:05.398531   241 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:03
+I0827 04:29:05.398662   241 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:05 PP5 TP2 EP2] Init torch distributed begin.
+[2026-08-27 04:29:26 PP5 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:26 PP4 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:27 PP4 TP1 EP1] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP4 TP3 EP3] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP4 TP2 EP2] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP4 TP0 EP0] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP5 TP2 EP2] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP5 TP3 EP3] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP5 TP0 EP0] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP5 TP1 EP1] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP5 TP0 EP0] Init torch distributed ends. elapsed=22.37 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP5 TP3 EP3] Init torch distributed ends. elapsed=22.42 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP5 TP2 EP2] Init torch distributed ends. elapsed=22.35 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP4 TP1 EP1] Init torch distributed ends. elapsed=22.44 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP5 TP1 EP1] Init torch distributed ends. elapsed=22.46 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP4 TP3 EP3] Init torch distributed ends. elapsed=22.38 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP4 TP2 EP2] Init torch distributed ends. elapsed=22.44 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP4 TP0 EP0] Init torch distributed ends. elapsed=22.41 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP5 TP0 EP0] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP4 TP1 EP1] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP4 TP3 EP3] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP4 TP2 EP2] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP5 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP4 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP4 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP5 TP2 EP2] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP4 TP0 EP0] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP4 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP5 TP3 EP3] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP5 TP1 EP1] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP4 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP5 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP5 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP5 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP4 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP5 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP4 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP5 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP5 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP4 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP4 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP5 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP5 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP4 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP4 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP4 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP5 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP5 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP4 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP5 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:30:33 PP4 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:33 PP4 TP2 EP2] Load weight end. elapsed=65.50 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:34 PP4 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:34 PP4 TP3 EP3] Load weight end. elapsed=66.61 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP4 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:36 PP4 TP1 EP1] Load weight end. elapsed=68.45 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:36 PP5 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:36 PP5 TP0 EP0] Load weight end. elapsed=69.00 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:37 PP5 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:37 PP5 TP3 EP3] Load weight end. elapsed=69.63 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:37 PP5 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:37 PP5 TP1 EP1] Load weight end. elapsed=69.72 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:38 PP5 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:38 PP5 TP2 EP2] Load weight end. elapsed=70.08 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:38 PP5 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:38 PP5 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:38 PP5 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:38 PP5 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:38 PP5 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:38 PP5 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:38 PP5 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:38 PP5 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:47 PP4 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:47 PP4 TP0 EP0] Load weight end. elapsed=79.31 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:47 PP4 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:47 PP4 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:47 PP4 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:47 PP4 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:47 PP4 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:47 PP4 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:47 PP4 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:47 PP4 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:54 PP5 TP3 EP3] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP5 TP1 EP1] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP5 TP2 EP2] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP5 TP0 EP0] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP4 TP3 EP3] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP4 TP2 EP2] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP4 TP1 EP1] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP4 TP0 EP0] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP5 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP5 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP4 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP4 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP4 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP5 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP4 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP5 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP5 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP5 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP4 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP5 TP3 EP3] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP4 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP4 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP4 TP1 EP1] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP5 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP5 TP2 EP2] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP4 TP3 EP3] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP4 TP0 EP0] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP5 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP5 TP1 EP1] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP5 TP0 EP0] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP4 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP4 TP2 EP2] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP4 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:30:54 PP4 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:30:54 PP4 TP0 EP0] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp0_pp4_dp0.json
+[2026-08-27 04:30:54 PP4 TP1 EP1] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp1_pp4_dp0.json
+[2026-08-27 04:30:54 PP5 TP2 EP2] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp2_pp5_dp0.json
+[2026-08-27 04:30:54 PP5 TP3 EP3] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp3_pp5_dp0.json
+[2026-08-27 04:30:54 PP5 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:30:54 PP5 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO[TensorRT-LLM][INFO] Set logger level to INFO
+
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[2026-08-27 04:30:54 PP5 TP0 EP0] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp0_pp5_dp0.json
+[2026-08-27 04:30:54 PP4 TP2 EP2] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp2_pp4_dp0.json
+[2026-08-27 04:30:54 PP4 TP3 EP3] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp3_pp4_dp0.json
+[2026-08-27 04:30:54 PP5 TP1 EP1] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp1_pp5_dp0.json
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.37profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.12profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.87profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.32profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.91profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.77profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.94profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.88profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.20profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.87profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  7.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 10.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.46profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.67profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.91profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.29profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.49profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  7.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.30profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.94profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.47profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.12profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.06profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.35profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  9.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:01<00:02,  2.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.89profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.58profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.32profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.70profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.20profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.27profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  6.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  7.03profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.56profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.73profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.88profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.37profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:05,  1.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:01<00:02,  2.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.05profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.19profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.77profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.18profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.25profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  3.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.18profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.20profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.37profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  9.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.27profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  8.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.43profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.89profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.35profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.96profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.90profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.30profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.72profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.18profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.33profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.51profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  2.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.03profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.68profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.16profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.01profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.24profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 10.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:02,  2.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.06profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.83profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.35profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.87profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.19profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  3.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.06profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.86profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 13.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 13.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.13profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:01,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:01,  3.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.47profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  6.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.47profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.57profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.22profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.17profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.50profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.71profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.19profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.55profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.90profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:05,  1.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:01<00:06,  1.00s/profile]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.33profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.19profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.03profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.76profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.28profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.32profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.15profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.79profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.31profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.18profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 11.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:01,  3.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.06profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:01,  3.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  6.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  3.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.05profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.60profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.50profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.92profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.00profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.29profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 11.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.09profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.43profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.11profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.91profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  7.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:02,  2.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.18profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.78profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.75profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.89profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.64profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.05profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.47profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.47profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.65profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:05,  1.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.85profile/s]
+[2026-08-27 04:31:28 PP5 TP1 EP1] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.21profile/s]
+[2026-08-27 04:31:28 PP5 TP0 EP0] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.74profile/s]
+[2026-08-27 04:31:28 PP5 TP2 EP2] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.77profile/s]
+[2026-08-27 04:31:28 PP5 TP3 EP3] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.19profile/s][2026-08-27 04:31:28 PP5 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:28 PP5 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:28 PP5 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:28 PP5 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:28 PP5 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:28 PP5 TP0 EP0] max_total_num_tokens=3801408, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=54, context_len=1048576, available_gpu_mem=16.30 GB
+[2026-08-27 04:31:28 PP5 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:28 PP5 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:28 PP5 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.87profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.90profile/s]
+[2026-08-27 04:31:29 PP4 TP3 EP3] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.87profile/s]
+[2026-08-27 04:31:29 PP4 TP2 EP2] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.93profile/s]
+[2026-08-27 04:31:29 PP4 TP0 EP0] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.43profile/s]
+[2026-08-27 04:31:29 PP4 TP1 EP1] FlashInfer autotune completed.
+[2026-08-27 04:31:29 PP4 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:29 PP4 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:29 PP4 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:29 PP4 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:29 PP4 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:29 PP4 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:29 PP4 TP0 EP0] max_total_num_tokens=3801408, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=54, context_len=1048576, available_gpu_mem=16.30 GB
+[2026-08-27 04:31:29 PP4 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:29 PP4 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+W0827 04:31:30.513926  2527 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50a93b80, len: 6912: Operation not permitted [1]
+W0827 04:31:30.514082  2528 memory_location.cpp:73] Failed to get NUMA node, addr: 0x507df700, len: 6912: Operation not permitted [1]
+W0827 04:31:30.514106  2529 memory_location.cpp:73] Failed to get NUMA node, addr: 0x508aa940, len: 6912: Operation not permitted [1]
+W0827 04:31:30.514214  2530 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50f67bc0, len: 6912: Operation not permitted [1]
+W0827 04:31:30.514272  2531 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50a2f680, len: 55296: Operation not permitted [1]
+W0827 04:31:30.514393  2532 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50afecc0, len: 55296: Operation not permitted [1]
+W0827 04:31:30.514506  2533 memory_location.cpp:73] Failed to get NUMA node, addr: 0x502be580, len: 6912: Operation not permitted [1]
+W0827 04:31:30.514607  2534 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5054d3c0, len: 13824: Operation not permitted [1]
+W0827 04:31:30.514652  2535 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50ad4580, len: 6912: Operation not permitted [1]
+W0827 04:31:30.514743  2536 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50eb9880, len: 6912: Operation not permitted [1]
+W0827 04:31:31.680995  2556 memory_location.cpp:73] Failed to get NUMA node, addr: 0x501c3300, len: 6912: Operation not permitted [1]
+W0827 04:31:31.680985  2557 memory_location.cpp:73] Failed to get NUMA node, addr: 0x504ecec0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.680985  2555 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd4bac0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681169  2558 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50706900, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681272  2559 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5006d700, len: 55296: Operation not permitted [1]
+W0827 04:31:31.681360  2560 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50041cc0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.681468  2561 memory_location.cpp:73] Failed to get NUMA node, addr: 0x506f5a40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681562  2562 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fdfdf40, len: 13824: Operation not permitted [1]
+W0827 04:31:31.681607  2563 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd16080, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681685  2564 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fcfa300, len: 6912: Operation not permitted [1]
+W0827 04:31:31.733497  2583 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51554040, len: 6912: Operation not permitted [1]
+W0827 04:31:31.733552  2584 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5110a8c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.733618  2585 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51ca8e80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.733795  2586 memory_location.cpp:73] Failed to get NUMA node, addr: 0x52017680, len: 6912: Operation not permitted [1]
+W0827 04:31:31.733909  2587 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5174af80, len: 55296: Operation not permitted [1]
+W0827 04:31:31.734004  2588 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51770440, len: 55296: Operation not permitted [1]
+W0827 04:31:31.734061  2589 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51118000, len: 6912: Operation not permitted [1]
+W0827 04:31:31.734179  2590 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51564f40, len: 13824: Operation not permitted [1]
+W0827 04:31:31.734277  2591 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51f69480, len: 6912: Operation not permitted [1]
+W0827 04:31:31.734400  2592 memory_location.cpp:73] Failed to get NUMA node, addr: 0x517c2940, len: 6912: Operation not permitted [1]
+W0827 04:31:31.741586  2611 memory_location.cpp:73] Failed to get NUMA node, addr: 0x519520c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.741609  2612 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50ff8c40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.741679  2613 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51a2ad80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.741792  2614 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51b5b300, len: 6912: Operation not permitted [1]
+W0827 04:31:31.741938  2615 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50fe5b80, len: 55296: Operation not permitted [1]
+W0827 04:31:31.742043  2616 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51625140, len: 55296: Operation not permitted [1]
+W0827 04:31:31.742168  2617 memory_location.cpp:73] Failed to get NUMA node, addr: 0x516bf640, len: 6912: Operation not permitted [1]
+W0827 04:31:31.742278  2618 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51b73e80, len: 13824: Operation not permitted [1]
+W0827 04:31:31.742414  2619 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51ef60c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.742429  2620 memory_location.cpp:73] Failed to get NUMA node, addr: 0x513e95c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.900663  2641 memory_location.cpp:73] Failed to get NUMA node, addr: 0x517e8c80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.900679  2640 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5170b840, len: 6912: Operation not permitted [1]
+W0827 04:31:31.900738  2642 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5131d0c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.900669  2639 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51372400, len: 6912: Operation not permitted [1]
+W0827 04:31:31.900923  2643 memory_location.cpp:73] Failed to get NUMA node, addr: 0x512b7e00, len: 55296: Operation not permitted [1]
+W0827 04:31:31.900976  2644 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51380200, len: 55296: Operation not permitted [1]
+W0827 04:31:31.901000  2646 memory_location.cpp:73] Failed to get NUMA node, addr: 0x513151c0, len: 13824: Operation not permitted [1]
+W0827 04:31:31.901011  2645 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51194f40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.901038  2647 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51520740, len: 6912: Operation not permitted [1]
+W0827 04:31:31.901073  2648 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51724980, len: 6912: Operation not permitted [1]
+W0827 04:31:31.931138  2668 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51061000, len: 6912: Operation not permitted [1]
+W0827 04:31:31.931178  2669 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50d74c00, len: 6912: Operation not permitted [1]
+W0827 04:31:31.931098  2667 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50b2fb40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.931272  2670 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50de9000, len: 6912: Operation not permitted [1]
+W0827 04:31:31.931349  2671 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5097dc80, len: 55296: Operation not permitted [1]
+W0827 04:31:31.931519  2672 memory_location.cpp:73] Failed to get NUMA node, addr: 0x515aa000, len: 55296: Operation not permitted [1]
+W0827 04:31:31.931557  2675 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51269700, len: 6912: Operation not permitted [1]
+W0827 04:31:31.931556  2673 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5097bd00, len: 6912: Operation not permitted [1]
+W0827 04:31:31.931576  2674 memory_location.cpp:73] Failed to get NUMA node, addr: 0x514ee800, len: 13824: Operation not permitted [1]
+W0827 04:31:31.931759  2676 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50deac80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.983201  2696 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50819380, len: 6912: Operation not permitted [1]
+W0827 04:31:31.983201  2695 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50600a80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.983378  2697 memory_location.cpp:73] Failed to get NUMA node, addr: 0x506ddf40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.983438  2698 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50716800, len: 6912: Operation not permitted [1]
+W0827 04:31:31.983490  2699 memory_location.cpp:73] Failed to get NUMA node, addr: 0x501575c0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.983624  2700 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50863340, len: 55296: Operation not permitted [1]
+W0827 04:31:31.983737  2701 memory_location.cpp:73] Failed to get NUMA node, addr: 0x508d7a40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.983930  2702 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50668c40, len: 13824: Operation not permitted [1]
+W0827 04:31:31.984035  2703 memory_location.cpp:73] Failed to get NUMA node, addr: 0x506841c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.984107  2704 memory_location.cpp:73] Failed to get NUMA node, addr: 0x508f16c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.001853  2725 memory_location.cpp:73] Failed to get NUMA node, addr: 0x507c8d40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.001852  2724 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50b39240, len: 6912: Operation not permitted [1]
+W0827 04:31:32.002048  2727 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50a8ecc0, len: 55296: Operation not permitted [1]
+W0827 04:31:32.001888  2726 memory_location.cpp:73] Failed to get NUMA node, addr: 0x512a8140, len: 6912: Operation not permitted [1]
+W0827 04:31:32.002044  2728 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50a68880, len: 55296: Operation not permitted [1]
+W0827 04:31:32.002257  2729 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5087b880, len: 6912: Operation not permitted [1]
+W0827 04:31:32.002264  2731 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50fc6980, len: 6912: Operation not permitted [1]
+W0827 04:31:32.002266  2730 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51331100, len: 13824: Operation not permitted [1]
+W0827 04:31:32.002470  2732 memory_location.cpp:73] Failed to get NUMA node, addr: 0x513384c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.002780  2723 memory_location.cpp:73] Failed to get NUMA node, addr: 0x512c6940, len: 6912: Operation not permitted [1]
+[2026-08-27 04:31:32] Dummy health check server started in background thread at 0.0.0.0:30000
+[rank16]:[W827 04:33:25.972222579 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank17]:[W827 04:33:25.980800100 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank19]:[W827 04:33:25.981413562 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank18]:[W827 04:33:26.644294178 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank16]:[W827 04:34:10.467297472 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank18]:[W827 04:34:10.467837454 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank17]:[W827 04:34:10.468151853 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank19]:[W827 04:34:10.468427857 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank22]:[W827 04:34:10.495396864 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank23]:[W827 04:34:10.495456053 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank21]:[W827 04:34:10.495795969 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank20]:[W827 04:34:10.495814123 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank22]:[W827 04:34:16.703081720 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank20]:[W827 04:34:16.703915664 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank21]:[W827 04:34:16.704033167 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank23]:[W827 04:34:16.704314444 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[2026-08-27 04:35:10 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.29
+[2026-08-27 04:35:10 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.29
+[2026-08-27 04:35:29 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.39
+[2026-08-27 04:35:29 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.39
+[2026-08-27 04:35:31 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25.89
+[2026-08-27 04:35:31 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25.91
+[2026-08-27 04:43:23 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.14
+[2026-08-27 04:43:23 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.14
+[2026-08-27 04:48:13 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.22
+[2026-08-27 04:48:14 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.22
+[2026-08-27 04:48:14 PP4 TP1 EP1] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP4 TP3 EP3] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP4 TP2 EP2] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP4 TP0 EP0] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP5 TP1 EP1] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP5 TP2 EP2] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP5 TP0 EP0] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP5 TP3 EP3] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:57:14 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.12
+[2026-08-27 04:57:14 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.12
+[2026-08-27 05:04:21 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1088, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2.55
+[2026-08-27 05:04:21 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1088, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2.55
+[2026-08-27 05:04:41 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.19
+[2026-08-27 05:04:41 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.19
+[2026-08-27 05:05:39 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1.10
+[2026-08-27 05:05:39 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1.10
+[2026-08-27 05:07:33 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 72.17
+[2026-08-27 05:07:33 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 72.17
+[2026-08-27 05:07:33 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 437051.18
+[2026-08-27 05:07:33 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 375850.16
+[2026-08-27 05:07:33 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25700.77
+[2026-08-27 05:07:33 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33225.98
+[2026-08-27 05:07:40 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8673, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1083.39
+[2026-08-27 05:07:40 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8673, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1083.24
+[2026-08-27 05:07:40 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 481, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 429004.90
+[2026-08-27 05:07:40 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 481, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 414364.03
+[2026-08-27 05:07:40 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 30845.47
+[2026-08-27 05:07:40 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 34555.92
+[2026-08-27 05:07:48 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8656, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1082.66
+[2026-08-27 05:07:48 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8656, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1082.26
+[2026-08-27 05:07:48 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 464, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 397827.41
+[2026-08-27 05:07:48 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 464, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 437968.14
+[2026-08-27 05:07:48 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32533.07
+[2026-08-27 05:07:48 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 31758.27
+[2026-08-27 05:07:55 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8663, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1087.59
+[2026-08-27 05:07:55 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8663, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1087.59
+[2026-08-27 05:07:55 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 471, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 394787.42
+[2026-08-27 05:07:55 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 471, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 405862.61
+[2026-08-27 05:07:55 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32866.57
+[2026-08-27 05:07:55 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32717.90
+[2026-08-27 05:08:03 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8659, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1082.97
+[2026-08-27 05:08:03 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8659, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1083.02
+[2026-08-27 05:08:03 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 467, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 419865.10
+[2026-08-27 05:08:03 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 467, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 410377.15
+[2026-08-27 05:08:03 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33296.73
+[2026-08-27 05:08:03 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 28803.86
+[2026-08-27 05:08:11 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8692, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1085.66
+[2026-08-27 05:08:11 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8692, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1085.97
+[2026-08-27 05:08:11 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 500, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 394417.45
+[2026-08-27 05:08:11 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 500, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 404424.58
+[2026-08-27 05:08:11 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33136.89
+[2026-08-27 05:08:11 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 31089.88
+[2026-08-27 05:08:18 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8557, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1090.03
+[2026-08-27 05:08:18 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8557, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1089.95
+[2026-08-27 05:08:18 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 365, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 409432.91
+[2026-08-27 05:08:18 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 365, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 521781.11
+[2026-08-27 05:08:18 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 30469.13
+[2026-08-27 05:08:18 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 22116.29
+[2026-08-27 05:08:26 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8683, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1083.95
+[2026-08-27 05:08:26 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8683, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1084.25
+[2026-08-27 05:08:26 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 491, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 406971.55
+[2026-08-27 05:08:26 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 491, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 413225.93
+[2026-08-27 05:08:26 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33768.16
+[2026-08-27 05:08:26 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32002.71
+[2026-08-27 05:11:45 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.32
+[2026-08-27 05:11:45 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.32
+[2026-08-27 05:14:42 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1088, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6.16
+[2026-08-27 05:14:42 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1088, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6.16
+[2026-08-27 05:17:47 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 44.22
+[2026-08-27 05:17:47 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 44.22
+[2026-08-27 05:17:47 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 414289.62
+[2026-08-27 05:17:47 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 418259.05
+[2026-08-27 05:17:47 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 31189.29
+[2026-08-27 05:17:47 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 34799.42
+[2026-08-27 05:17:55 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8673, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1086.30
+[2026-08-27 05:17:55 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8673, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1086.07
+[2026-08-27 05:17:55 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 481, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 412548.65
+[2026-08-27 05:17:55 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 481, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 453654.88
+[2026-08-27 05:17:55 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33203.44
+[2026-08-27 05:17:55 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32631.53
+[2026-08-27 05:18:02 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8656, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1088.44
+[2026-08-27 05:18:02 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8656, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1088.37
+[2026-08-27 05:18:02 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 464, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 420591.31
+[2026-08-27 05:18:02 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 464, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 421477.03
+[2026-08-27 05:18:02 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33878.95
+[2026-08-27 05:18:02 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 35064.03
+[2026-08-27 05:18:10 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8663, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1088.01
+[2026-08-27 05:18:10 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8663, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1087.84
+[2026-08-27 05:18:10 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 471, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 429396.76
+[2026-08-27 05:18:10 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 471, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 403808.84
+[2026-08-27 05:18:10 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 30291.78
+[2026-08-27 05:18:10 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 35867.18
+[2026-08-27 05:18:18 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8659, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1086.23
+[2026-08-27 05:18:18 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8659, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1086.07
+[2026-08-27 05:18:18 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 467, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 416829.87
+[2026-08-27 05:18:18 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 467, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 367035.35
+[2026-08-27 05:18:18 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 27946.41
+[2026-08-27 05:18:18 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32311.83
+[2026-08-27 05:18:25 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8692, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1087.22
+[2026-08-27 05:18:25 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8692, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1087.19
+[2026-08-27 05:18:25 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 500, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 454585.15
+[2026-08-27 05:18:25 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 500, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 478466.41
+[2026-08-27 05:18:25 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 34596.06
+[2026-08-27 05:18:25 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 36601.41
+[2026-08-27 05:18:33 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8557, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1090.87
+[2026-08-27 05:18:33 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8557, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1090.67
+[2026-08-27 05:18:33 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 365, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 423632.77
+[2026-08-27 05:18:33 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 365, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 427798.84
+[2026-08-27 05:18:33 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 24938.13
+[2026-08-27 05:18:33 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 26814.84
+[2026-08-27 05:18:40 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8683, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1087.87
+[2026-08-27 05:18:40 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8683, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1087.73
+[2026-08-27 05:18:40 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 491, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 435790.11
+[2026-08-27 05:18:40 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 491, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 437235.07
+[2026-08-27 05:18:40 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33178.64
+[2026-08-27 05:18:40 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 30137.19
+[2026-08-27 05:21:46 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 2, #pending-token: 8652, #bootstrap-req: 2, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 44.04
+[2026-08-27 05:21:47 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 3, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9796.50
+[2026-08-27 05:21:47 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 2, #pending-token: 8652, #bootstrap-req: 2, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 43.85
+[2026-08-27 05:21:48 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 3, #pending-token: 9185, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8786.35
+[2026-08-27 05:21:48 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 3, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8753.58
+[2026-08-27 05:21:49 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 2, #pending-token: 993, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9748.57
+[2026-08-27 05:21:49 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 3, #pending-token: 9185, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9863.29
+[2026-08-27 05:21:50 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 2, #pending-token: 9680, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8707.37
+[2026-08-27 05:21:50 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 2, #pending-token: 993, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8549.77
+[2026-08-27 05:21:51 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1488, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9553.86
+[2026-08-27 05:21:51 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 2, #pending-token: 9680, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9316.43
+[2026-08-27 05:21:52 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1488, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9810.72
+[2026-08-27 05:21:52 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 10199, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8877.35
+[2026-08-27 05:21:52 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 18858, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10683.27
+[2026-08-27 05:21:52 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 10199, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9914.60
+[2026-08-27 05:21:53 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 27591, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9125.55
+[2026-08-27 05:21:53 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 18858, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9121.64
+[2026-08-27 05:21:54 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 53023, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9612.99
+[2026-08-27 05:21:54 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 27591, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9653.85
+[2026-08-27 05:21:55 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 44876, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9352.82
+[2026-08-27 05:21:55 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 53023, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9371.66
+[2026-08-27 05:21:56 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 36684, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9652.74
+[2026-08-27 05:21:56 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 44876, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9597.62
+[2026-08-27 05:21:57 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 28504, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10734.61
+[2026-08-27 05:21:57 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 36684, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9728.96
+[2026-08-27 05:21:57 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 20312, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 15406.33
+[2026-08-27 05:21:57 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 28504, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 18481.89
+[2026-08-27 05:21:57 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 20312, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 294820.71
+[2026-08-27 05:21:57 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 12139, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 258255.44
+[2026-08-27 05:21:57 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 3947, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 296280.34
+[2026-08-27 05:21:57 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 12139, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 273258.91
+[2026-08-27 05:21:57 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 3968, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 41900.16
+[2026-08-27 05:21:57 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 3947, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 83041.18
+[2026-08-27 05:21:58 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 3968, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7314.63
+[2026-08-27 05:22:06 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 0, #pending-token: 8608, #bootstrap-req: 4, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 922.15
+[2026-08-27 05:22:07 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 416, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9857.85
+[2026-08-27 05:22:07 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 0, #pending-token: 8608, #bootstrap-req: 4, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 892.94
+[2026-08-27 05:22:08 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 9149, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8818.94
+[2026-08-27 05:22:08 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 416, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8761.69
+[2026-08-27 05:22:09 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 957, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10110.54
+[2026-08-27 05:22:09 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 9149, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10200.43
+[2026-08-27 05:22:10 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 9638, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8782.22
+[2026-08-27 05:22:10 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 957, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8750.32
+[2026-08-27 05:22:10 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1446, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9843.04
+[2026-08-27 05:22:10 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 9638, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9909.13
+[2026-08-27 05:22:11 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1446, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9535.57
+[2026-08-27 05:22:11 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 10033, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9304.25
+[2026-08-27 05:22:12 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1841, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8648.37
+[2026-08-27 05:22:12 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 10033, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8117.84
+[2026-08-27 05:22:13 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1841, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10198.16
+[2026-08-27 05:22:13 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1856, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2161.72
+[2026-08-27 05:22:14 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8540, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8634.86
+[2026-08-27 05:22:14 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1856, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1928.73
+[2026-08-27 05:22:15 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8540, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9967.52
+[2026-08-27 05:22:15 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 17066, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9942.20
+[2026-08-27 05:22:16 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8910, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8633.09
+[2026-08-27 05:22:16 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 17066, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8614.10
+[2026-08-27 05:22:17 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 17466, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10316.02
+[2026-08-27 05:22:17 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8910, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10031.51
+[2026-08-27 05:22:18 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9324, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9877.92
+[2026-08-27 05:22:18 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 17466, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9424.86
+[2026-08-27 05:22:18 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 17751, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 36929.95
+[2026-08-27 05:22:18 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9324, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 50710.16
+[2026-08-27 05:22:18 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 17751, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 310670.60
+[2026-08-27 05:22:18 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9579, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 230464.10
+[2026-08-27 05:22:18 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1387, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 75748.54
+[2026-08-27 05:22:18 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9579, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 70984.49
+[2026-08-27 05:22:18 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1408, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2600.96
+[2026-08-27 05:22:18 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1387, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 15177.10
+[2026-08-27 05:22:19 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1408, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1879.05
+[2026-08-27 05:22:27 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8547, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 982.50
+[2026-08-27 05:22:28 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 355, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9870.85
+[2026-08-27 05:22:28 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8547, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 973.15
+[2026-08-27 05:22:29 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8932, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8633.06
+[2026-08-27 05:22:29 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 355, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8626.74
+[2026-08-27 05:22:29 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 740, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10223.25
+[2026-08-27 05:22:29 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8932, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10140.49
+[2026-08-27 05:22:30 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 740, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8655.21
+[2026-08-27 05:22:30 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9259, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8412.98
+[2026-08-27 05:22:31 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1067, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10205.64
+[2026-08-27 05:22:31 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9259, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9957.18
+[2026-08-27 05:22:32 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1067, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9648.18
+[2026-08-27 05:22:32 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9745, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9472.33
+[2026-08-27 05:22:33 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1553, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8576.88
+[2026-08-27 05:22:33 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9745, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8156.56
+[2026-08-27 05:22:34 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1600, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1870.62
+[2026-08-27 05:22:34 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1553, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9817.59
+[2026-08-27 05:22:35 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8602, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8886.75
+[2026-08-27 05:22:35 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1600, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1744.17
+[2026-08-27 05:22:36 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 410, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9856.51
+[2026-08-27 05:22:36 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8602, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9924.95
+[2026-08-27 05:22:36 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8696, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8614.79
+[2026-08-27 05:22:36 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 410, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8588.26
+[2026-08-27 05:22:37 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 504, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10212.49
+[2026-08-27 05:22:37 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8696, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10054.15
+[2026-08-27 05:22:38 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9164, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9916.25
+[2026-08-27 05:22:38 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 504, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9469.58
+[2026-08-27 05:22:38 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 972, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 45807.47
+[2026-08-27 05:22:38 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9164, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 65745.12
+[2026-08-27 05:22:38 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 972, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 271786.52
+[2026-08-27 05:22:38 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9322, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 228784.17
+[2026-08-27 05:22:38 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1130, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 76741.41
+[2026-08-27 05:22:38 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9322, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 73462.62
+[2026-08-27 05:22:39 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1152, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2130.27
+[2026-08-27 05:22:39 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1130, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 15175.73
+[2026-08-27 05:22:40 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1152, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1541.83
+[2026-08-27 05:22:47 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8602, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 986.08
+[2026-08-27 05:22:48 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 410, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9910.97
+[2026-08-27 05:22:48 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8602, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 976.33
+[2026-08-27 05:22:49 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 9105, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8736.80
+[2026-08-27 05:22:49 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 410, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8755.48
+[2026-08-27 05:22:50 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 913, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10163.94
+[2026-08-27 05:22:50 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 9105, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10126.04
+[2026-08-27 05:22:51 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9375, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8789.11
+[2026-08-27 05:22:51 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 913, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8802.92
+[2026-08-27 05:22:52 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1183, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9769.42
+[2026-08-27 05:22:52 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9375, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9789.67
+[2026-08-27 05:22:52 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9868, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9680.86
+[2026-08-27 05:22:52 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1183, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9551.84
+[2026-08-27 05:22:53 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1676, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8479.61
+[2026-08-27 05:22:53 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9868, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8580.43
+[2026-08-27 05:22:54 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1676, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9905.15
+[2026-08-27 05:22:54 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2025.45
+[2026-08-27 05:22:55 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8467, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8799.98
+[2026-08-27 05:22:55 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1790.28
+[2026-08-27 05:22:56 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8467, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9998.59
+[2026-08-27 05:22:56 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 275, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9869.57
+[2026-08-27 05:22:57 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8988, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8731.62
+[2026-08-27 05:22:57 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 275, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8714.49
+[2026-08-27 05:22:58 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 796, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10327.91
+[2026-08-27 05:22:58 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8988, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10150.84
+[2026-08-27 05:22:59 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9363, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9995.05
+[2026-08-27 05:22:59 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 796, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9408.40
+[2026-08-27 05:22:59 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 1171, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 30814.55
+[2026-08-27 05:22:59 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9363, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 40670.98
+[2026-08-27 05:22:59 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 1171, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 312400.93
+[2026-08-27 05:22:59 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9880, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 273542.20
+[2026-08-27 05:22:59 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1688, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 81674.04
+[2026-08-27 05:22:59 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9880, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 78852.53
+[2026-08-27 05:23:00 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3180.06
+[2026-08-27 05:23:00 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1688, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 15111.86
+[2026-08-27 05:23:00 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2313.52
+[2026-08-27 05:23:08 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8601, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 987.24
+[2026-08-27 05:23:09 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 409, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9965.41
+[2026-08-27 05:23:09 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8601, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 978.45
+[2026-08-27 05:23:10 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 9104, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8686.72
+[2026-08-27 05:23:10 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 409, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8681.63
+[2026-08-27 05:23:10 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 912, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10219.36
+[2026-08-27 05:23:10 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 9104, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10129.55
+[2026-08-27 05:23:11 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9485, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8726.43
+[2026-08-27 05:23:11 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 912, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8797.95
+[2026-08-27 05:23:12 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1293, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9968.13
+[2026-08-27 05:23:12 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9485, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9970.37
+[2026-08-27 05:23:13 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1293, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9561.18
+[2026-08-27 05:23:13 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9900, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9322.88
+[2026-08-27 05:23:14 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1708, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8660.45
+[2026-08-27 05:23:14 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9900, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8445.20
+[2026-08-27 05:23:15 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1708, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9863.39
+[2026-08-27 05:23:15 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2040.28
+[2026-08-27 05:23:16 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8525, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8723.80
+[2026-08-27 05:23:16 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1809.05
+[2026-08-27 05:23:17 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8525, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9978.38
+[2026-08-27 05:23:17 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 333, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9952.63
+[2026-08-27 05:23:18 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 333, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8701.62
+[2026-08-27 05:23:18 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8909, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8696.74
+[2026-08-27 05:23:18 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 717, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10354.85
+[2026-08-27 05:23:18 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8909, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10184.48
+[2026-08-27 05:23:19 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9452, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10071.93
+[2026-08-27 05:23:19 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 717, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9491.81
+[2026-08-27 05:23:19 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 1260, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33416.21
+[2026-08-27 05:23:19 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9452, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 44390.95
+[2026-08-27 05:23:19 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 1260, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 436736.09
+[2026-08-27 05:23:19 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9830, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 336391.66
+[2026-08-27 05:23:19 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9830, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 339855.11
+[2026-08-27 05:23:19 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1638, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 394477.69
+[2026-08-27 05:23:19 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1638, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 408438.83
+[2026-08-27 05:23:19 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1664, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 78605.25
+[2026-08-27 05:23:19 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1664, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 99592.51
+[2026-08-27 05:32:52 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.11
+[2026-08-27 05:32:52 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.11
+[2026-08-27 05:35:54 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 45.15
+[2026-08-27 05:35:54 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 45.15
+[2026-08-27 05:35:54 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 377056.19
+[2026-08-27 05:35:54 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 398020.47
+[2026-08-27 05:35:54 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 31553.00
+[2026-08-27 05:35:54 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 31134.23
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-20260827-1325/logs/manual_p_node3.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-20260827-1325/logs/manual_p_node3.log
new file mode 100644
index 0000000..e286bf5
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-20260827-1325/logs/manual_p_node3.log
@@ -0,0 +1,1063 @@
+/usr/local/lib/python3.12/dist-packages/torchao/quantization/quant_api.py:1731: SyntaxWarning: invalid escape sequence '\.'
+  """Configuration class for applying different quantization configs to modules or parameters based on their fully qualified names (FQNs).
+/sgl-workspace/sglang/python/sglang/launch_server.py:56: UserWarning: 'python -m sglang.launch_server' is still supported, but 'sglang serve' is the recommended entrypoint.
+  Example: sglang serve --model-path <model> [options]
+  warnings.warn(
+[2026-08-27 04:28:46] Breakable CUDA graph is incompatible with MLA attention; disabling prefill CUDA graph.
+[2026-08-27 04:28:47] Attention backend not specified. Use triton backend by default.
+[2026-08-27 04:28:47] Pipeline parallelism is incompatible with overlap schedule.
+[2026-08-27 04:28:49] server_args=ServerArgs(model_path='/data/hf_models/Kimi-K3', tokenizer_path='/data/hf_models/Kimi-K3', tokenizer_mode='auto', tokenizer_backend='huggingface', tokenizer_worker_num=1, detokenizer_worker_num=1, skip_tokenizer_init=False, load_format='auto', model_loader_extra_config='{}', trust_remote_code=True, context_length=None, is_embedding=False, enable_multimodal=None, revision=None, model_impl='auto', model_config_parser='auto', json_model_override_args='{}', dtype='auto', quantization=None, quantization_param_path=None, kv_cache_dtype='auto', enable_fp32_lm_head=False, modelopt_quant=None, modelopt_checkpoint_restore_path=None, modelopt_checkpoint_save_path=None, modelopt_export_path=None, quantize_and_serve=False, rl_quant_profile=None, enable_tf32_matmul=False, mem_fraction_static=0.88, max_running_requests=None, max_queued_requests=None, max_total_tokens=None, chunked_prefill_size=8192, enable_dynamic_chunking=False, max_prefill_tokens=16384, prefill_max_requests=None, schedule_policy='fcfs', enable_priority_scheduling=False, disable_priority_preemption=False, default_priority_value=None, abort_on_priority_when_disabled=False, schedule_low_priority_values_first=False, priority_scheduling_preemption_threshold=10, retraction_policy='length', schedule_conservativeness=1.0, page_size=64, swa_full_tokens_ratio=0.8, disable_hybrid_swa_memory=False, radix_eviction_policy='lru', prefill_only_disable_kv_cache=False, disable_radix_cache=True, enable_page_major_kv_layout=False, enable_unified_memory=False, disable_chunked_prefix_cache=False, disable_overlap_schedule=True, num_continuous_decode_steps=1, scheduler_recv_interval=1, enable_mixed_chunk=False, nccl_port=None, dist_timeout=3600, dist_init_addr='174.1.60.1:20000', nnodes=4, node_rank=3, tp_size=4, dcp_size=1, pp_size=8, pp_max_micro_batch_size=None, pp_async_batch_depth=0, dp_size=1, load_balance_method='follow_bootstrap_room', attn_cp_size=1, moe_dp_size=1, dcp_comm_backend='ag_rs', dcp_replicate_q_proj=None, dwdp_size=1, enable_prefill_cp=False, cp_strategy=None, enable_dsa_cache_layer_split=False, enable_dsa_prefill_context_parallel=False, dsa_prefill_cp_mode='round-robin-split', enable_prefill_context_parallel=False, prefill_cp_mode='in-seq-split', enable_cp_decode_attn_tp=False, enable_dp_attention=False, enable_dp_attention_local_control_broadcast=False, enable_dp_lm_head=False, enable_attn_tp_input_scattered=False, disable_attn_tp_gather=False, enable_p2p_check=False, device='cuda', base_gpu_id=0, gpu_id_step=1, random_seed=927276655, watchdog_timeout=300, soft_watchdog_timeout=None, sleep_on_idle=False, use_ray=False, custom_sigquit_handler=None, numa_node=None, gc_threshold=None, host='0.0.0.0', port=30000, fastapi_root_path='', smg_grpc_mode=False, grpc_mode=False, grpc_port=None, sidecar=None, sidecar_args=None, skip_server_warmup=False, warmups=None, enable_http2=False, ssl_keyfile=None, ssl_certfile=None, ssl_ca_certs=None, ssl_keyfile_password=None, enable_ssl_refresh=False, api_key=None, admin_api_key=None, served_model_name='kimi-k3', weight_version='default', chat_template=None, hf_chat_template_name=None, completion_template=None, file_storage_path='sglang_storage', enable_cache_report=False, reasoning_parser=None, default_chat_template_kwargs=None, strip_thinking_cache=False, enable_strict_thinking=False, tool_call_parser=None, tool_server=None, sampling_defaults='model', asr_max_buffer_seconds=60, asr_max_concurrent_sessions=32, preferred_sampling_params=None, allow_auto_truncate=False, stream_interval=1, batch_notify_size=16, stream_response_default_include_usage=False, incremental_streaming_output=False, enable_streaming_session=False, enable_session_radix_cache=False, log_level='info', log_level_http=None, log_requests=False, log_requests_level=2, log_requests_format='text', log_requests_target=None, uvicorn_access_log_exclude_prefixes=[], crash_dump_folder=None, show_time_cost=False, enable_metrics=False, smg_http_sidecar_port=None, enable_mfu_metrics=False, enable_metrics_for_all_schedulers=False, load_snapshot_publish_interval=15, tokenizer_metrics_custom_labels_header='x-custom-labels', tokenizer_metrics_allowed_custom_labels=None, extra_metric_labels=None, bucket_time_to_first_token=None, bucket_inter_token_latency=None, bucket_e2e_request_latency=None, prompt_tokens_buckets=None, generation_tokens_buckets=None, gc_warning_threshold_secs=0.0, decode_log_interval=40, enable_request_time_stats_logging=False, kv_events_config=None, enable_forward_pass_metrics=False, forward_pass_metrics_worker_id='', forward_pass_metrics_ipc_name=None, enable_trace=False, trace_modules='request', otlp_traces_endpoint='localhost:4317', export_metrics_to_file=False, export_metrics_to_file_dir=None, stat_loggers=None, constrained_json_whitespace_pattern=None, constrained_json_disable_any_whitespace=False, attention_backend='triton', decode_attention_backend=None, prefill_attention_backend=None, sampling_backend='flashinfer', grammar_backend='xgrammar', radix_cache_backend=None, mm_attention_backend=None, fp8_gemm_runner_backend='auto', fp4_gemm_runner_backend='auto', bf16_gemm_backend='auto', dsa_prefill_backend=None, dsa_decode_backend=None, dsa_paged_mqa_logits_backend='auto', dsa_topk_backend='sgl-kernel', disable_flashinfer_autotune=False, flashinfer_autotune_skip_ops=None, mamba_backend='triton', cuda_graph_config=CudaGraphConfig(decode=PhaseConfig(backend='disabled', max_bs=16, bs=[1, 2, 4, 8, 12, 16], tc_compiler='eager', full_prefill_max_req=None), prefill=PhaseConfig(backend='disabled', max_bs=2048, bs=[4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], tc_compiler='eager', full_prefill_max_req=None)), cuda_graph_backend_decode=None, cuda_graph_backend_prefill=None, cuda_graph_max_bs_decode=16, cuda_graph_max_bs_prefill=None, cuda_graph_bs_decode=None, cuda_graph_bs_prefill=None, cuda_graph_tc_compiler=None, disable_prefill_cuda_graph=False, disable_decode_cuda_graph=False, disable_cuda_graph=False, disable_cuda_graph_padding=False, enable_profile_cuda_graph=False, enable_cudagraph_gc=False, debug_cuda_graph=False, enable_layerwise_nvtx_marker=False, enable_nccl_nvls=False, enable_symm_mem=False, triton_attention_reduce_in_fp32=False, triton_attention_num_kv_splits=8, triton_attention_split_tile_size=None, flashinfer_mla_disable_ragged=False, enable_fused_qk_norm_rope=False, enable_precise_embedding_interpolation=False, enable_fused_moe_sum_all_reduce=False, enable_deepseek_v4_fp4_indexer=False, disable_custom_all_reduce=False, enable_mscclpp=False, enable_torch_symm_mem=False, enable_scattered_sconv=False, pre_warm_nccl=False, enable_quant_communications=False, enable_flashinfer_allreduce_fusion=False, enforce_disable_flashinfer_allreduce_fusion=False, flashinfer_allreduce_fusion_backend=None, enable_aiter_allreduce_fusion=False, enable_torch_compile=False, enable_torch_compile_debug_mode=False, torch_compile_max_bs=32, torchao_config='', speculative_algorithm=None, speculative_draft_model_path=None, speculative_draft_model_revision=None, speculative_draft_load_format=None, speculative_num_steps=None, speculative_eagle_topk=None, speculative_num_draft_tokens=None, speculative_dflash_block_size=None, speculative_dspark_block_size=None, speculative_dspark_sps_table_path=None, speculative_dspark_confidence_sts_path=None, speculative_dspark_align_verify_tokens_to_graph_tier=False, speculative_accept_threshold_single=1.0, speculative_accept_threshold_acc=1.0, speculative_use_rejection_sampling=False, speculative_token_map=None, speculative_attention_mode='prefill', speculative_draft_attention_backend=None, speculative_draft_window_size=None, speculative_moe_runner_backend='flashinfer_mxfp4', speculative_moe_a2a_backend=None, speculative_draft_model_quantization=None, speculative_skip_dp_mlp_sync=False, enable_multi_layer_eagle=False, speculative_adaptive=False, speculative_adaptive_config=None, decoupled_spec_bind_endpoint=None, decoupled_spec_connect_endpoints=None, decoupled_spec_rank=None, decoupled_spec_role='null', spec_trace_dir=None, speculative_ngram_min_bfs_breadth=1, speculative_ngram_max_bfs_breadth=10, speculative_ngram_match_type='BFS', speculative_ngram_max_trie_depth=18, speculative_ngram_capacity=10000000, speculative_ngram_external_corpus_path=None, speculative_ngram_external_sam_budget=0, speculative_ngram_external_corpus_max_tokens=10000000, ep_size=4, moe_a2a_backend='none', moe_runner_backend='flashinfer_mxfp4', flashinfer_mxfp4_moe_precision='default', deepep_mode='auto', fuseep_mode=2, deepep_dispatcher_output_dtype='auto', ep_num_redundant_experts=0, ep_dispatch_algorithm=None, init_expert_location='trivial', enable_eplb=False, eplb_algorithm='auto', eplb_rebalance_num_iterations=1000, eplb_rebalance_layers_per_chunk=None, eplb_min_rebalancing_utilization_threshold=1.0, expert_distribution_recorder_mode=None, expert_distribution_recorder_buffer_size=1000, enable_expert_distribution_metrics=False, deepep_config=None, moe_dense_tp_size=None, elastic_ep_backend=None, enable_elastic_expert_backup=False, mooncake_ib_device=None, enable_waterfill=False, ep_join_mode=None, ep_join_rank_offset=0, elastic_ep_initial_size=None, max_ep_size=None, elastic_ep_scale_timeout=600, elastic_ep_rejoin=False, disable_flashinfer_cutlass_moe_fp4_allgather=False, disable_shared_experts_fusion=False, enforce_shared_experts_fusion=False, max_mamba_cache_size=None, mamba_ssm_dtype=None, mamba_max_states_per_path=-1, enable_mamba_cache_stochastic_rounding=False, mamba_cache_philox_rounds=0, mamba_full_memory_ratio=0.36, mamba_radix_cache_strategy='extra_buffer_lazy', uses_mamba_radix_cache=False, mamba_track_interval=256, enable_int8_mamba_checkpoint=False, int8_mamba_ckpt_size=None, linear_attn_backend='triton', linear_attn_decode_backend=None, linear_attn_prefill_backend=None, linear_attn_verify_backend=None, enable_linear_replayssm=False, linear_replayssm_cache_len=16, enable_linear_replayssm_spec=False, enable_hierarchical_cache=False, hicache_ratio=2.0, hicache_size=0, hicache_write_policy='write_through', hicache_io_backend='kernel', hicache_mem_layout='page_first', hicache_storage_backend=None, hicache_storage_prefetch_policy='timeout', hicache_storage_backend_extra_config=None, enable_hisparse=False, hisparse_config=None, enable_broadcast_mm_inputs_process=False, enable_prefix_mm_cache=False, mm_enable_dp_encoder=False, mm_process_config={}, mm_processor_worker_num=0, mm_io_worker_num=0, limit_mm_data_per_request=None, enable_mm_global_cache=False, disable_fast_image_processor=False, mm_feature_transport='cpu', keep_mm_feature_on_device=False, enable_lora=None, enable_lora_overlap_loading=None, max_lora_rank=None, lora_target_modules=None, lora_paths=None, max_loaded_loras=None, max_loras_per_batch=8, lora_eviction_policy='lru', lora_backend='csgmv', max_lora_chunk_size=16, experts_shared_outer_loras=None, lora_use_virtual_experts=False, lora_strict_loading=False, lora_drain_wait_threshold=0.0, enable_two_batch_overlap=False, enable_single_batch_overlap=False, tbo_token_distribution_threshold=0.48, cpu_offload_gb=0, offload_group_size=-1, offload_num_in_group=1, offload_prefetch_step=1, offload_mode='cpu', enable_lmcache=False, lmcache_config_file=None, enable_flexkv=False, flexkv_config_file=None, kt_weight_path=None, kt_method='AMXINT4', kt_cpuinfer=None, kt_threadpool_count=2, kt_num_gpu_experts=None, kt_max_deferred_experts_per_token=None, dllm_algorithm=None, dllm_algorithm_config=None, dllm_fdfo=True, disaggregation_mode='prefill', disaggregation_transfer_backend='mooncake', disaggregation_bootstrap_port=28800, disaggregation_ib_device='mlx5_0,mlx5_1,mlx5_2,mlx5_3', disaggregation_decode_enable_radix_cache=False, disaggregation_decode_enable_offload_kvcache=False, num_reserved_decode_tokens=512, disaggregation_decode_extra_slots=None, disaggregation_decode_polling_interval=1, optimistic_prefill_attempts=0, encoder_only=False, language_only=False, encoder_transfer_backend='zmq_to_tokenizer', encoder_urls=[], encoder_bootstrap_port=8997, encoder_register_urls=[], enable_adaptive_dispatch_to_encoder=False, enable_pdmux=False, pdmux_config_path=None, sm_group_num=8, custom_weight_loader=[], weight_loader_disable_mmap=False, weight_loader_prefetch_checkpoints=False, weight_loader_prefetch_num_threads=4, weight_loader_drop_cache_after_load=False, remote_instance_weight_loader_seed_instance_ip=None, remote_instance_weight_loader_seed_instance_service_port=None, remote_instance_weight_loader_send_weights_group_ports=None, remote_instance_weight_loader_backend='nccl', remote_instance_weight_loader_start_seed_via_transfer_engine=False, engine_info_bootstrap_port=6789, modelexpress_config=None, download_dir=None, model_checksum=None, delete_ckpt_after_loading=False, decrypted_config_file=None, decrypted_draft_config_file=None, checkpoint_engine_wait_weights_before_ready=False, enable_prefill_delayer=False, prefill_delayer_max_delay_passes=30, prefill_delayer_token_usage_low_watermark=None, prefill_delayer_forward_passes_buckets=None, prefill_delayer_wait_seconds_buckets=None, prefill_delayer_queue_min_ratio=None, prefill_delayer_max_delay_ms=None, min_free_slots_delay=None, enable_deterministic_inference=False, rl_on_policy_target=None, kv_canary='none', kv_canary_real_data='none', kv_canary_sweep_interval=0, enable_dynamic_batch_tokenizer=False, dynamic_batch_tokenizer_batch_size=32, dynamic_batch_tokenizer_batch_timeout=0.002, enable_tokenizer_batch_encode=False, disable_tokenizer_batch_decode=False, debug_tensor_dump_output_folder=None, debug_tensor_dump_layers=None, debug_tensor_dump_input_file=None, enable_memory_saver=False, enable_weights_cpu_backup=False, enable_draft_weights_cpu_backup=False, enable_custom_logit_processor=False, enable_return_hidden_states=False, enable_return_routed_experts=False, enable_return_indexer_topk=False, disable_outlines_disk_cache=False, enable_mis=False, weight_cache_mode='off', weight_cache_socket=None, weight_cache_timeout=1800, forward_hooks=None, msprobe_dump_config=None)
+[2026-08-27 04:28:49] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:49] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:49] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:49] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:49] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:49] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:49] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:49] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:05 PP6 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:05 PP6 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:07 PP6 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:07 PP6 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:07 PP6 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:07 PP7 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:07 PP7 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:07 PP7 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:07 PP7 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:07 PP6 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:07 PP6 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:07.662256   238 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:07.662379   238 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:07.662392   238 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.4 port: 12001
+I0827 04:29:07.662429   238 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.4:15914
+I0827 04:29:07.662514   238 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:07.685189   238 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:07.689950   238 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:07.695539   238 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:07.701294   238 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:07.703691   238 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:07.703715   238 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:07.703760   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:07.710492   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:07.711953   238 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:04
+I0827 04:29:07.711989   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:07.735716   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:07.737195   238 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:04
+I0827 04:29:07.737223   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:07.763653   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:07.765146   238 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:04
+I0827 04:29:07.765172   238 rdma_context.cpp:205] Using SIEVE endpoint store
+[2026-08-27 04:29:07 PP7 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:07 PP7 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:07 PP7 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:07 PP7 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+I0827 04:29:07.791589   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:07.793025   238 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:04
+I0827 04:29:07.793139   238 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:07 PP6 TP3 EP3] Init torch distributed begin.
+[2026-08-27 04:29:07 PP6 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:07 PP6 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:08 PP6 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:08.916008   235 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:08.916072   235 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:08.916082   235 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.4 port: 12001
+I0827 04:29:08.916121   235 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.4:15737
+I0827 04:29:08.916215   235 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:08.941840   235 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:08.950320   235 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:08.979278   235 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:08.988204   235 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:08.991855   235 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:08.991879   235 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:08.991928   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:08.999985   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:09.001416   235 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:04
+I0827 04:29:09.001446   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:09.027477   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:09.028954   235 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:04
+I0827 04:29:09.028980   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:09.055524   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:09.056981   235 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:04
+I0827 04:29:09.057020   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:09.083544   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:09.084911   235 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:04
+I0827 04:29:09.085031   235 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:09 PP6 TP0 EP0] Init torch distributed begin.
+[2026-08-27 04:29:09 PP7 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:09 PP6 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:09 PP7 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:09 PP6 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:09 PP7 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:09 PP7 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:10.000486   242 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:10.000609   242 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:10.000628   242 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.4 port: 12001
+I0827 04:29:10.000689   242 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.4:16890
+I0827 04:29:10.000851   242 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:10.006592   236 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:10.006711   236 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:10.006728   236 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.4 port: 12001
+I0827 04:29:10.006786   236 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.4:15893
+I0827 04:29:10.006963   236 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:10.022238   241 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:10.022354   241 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:10.022370   241 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.4 port: 12001
+I0827 04:29:10.022423   241 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.4:16277
+I0827 04:29:10.022593   241 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:10.032052   242 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:10.032101   236 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:10.039772   236 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:10.039858   242 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:10.047292   236 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:10.047341   242 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:10.047785   241 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:10.054806   242 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:10.054838   236 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:10.055280   241 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:10.059140   236 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:10.059171   236 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:10.059175   242 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:10.059201   242 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:10.059227   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.059260   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.062610   241 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:10.066787   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:10.066826   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:10.068500   242 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:04
+I0827 04:29:10.068532   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.068558   236 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:04
+I0827 04:29:10.068590   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.069866   241 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:10.072566   241 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:10.072592   241 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:10.072649   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.080149   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:10.081809   241 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:04
+I0827 04:29:10.081843   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.095701   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:10.097368   236 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:04
+I0827 04:29:10.097400   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.099974   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:10.101672   242 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:04
+I0827 04:29:10.101717   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.107705   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:10.109316   241 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:04
+I0827 04:29:10.109347   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.123868   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:10.125504   236 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:04
+I0827 04:29:10.125536   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.127957   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:10.129673   242 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:04
+I0827 04:29:10.129703   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.139813   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:10.141404   241 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:04
+I0827 04:29:10.141431   241 rdma_context.cpp:205] Using SIEVE endpoint store
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:10.150050   237 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:10.150159   237 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:10.150173   237 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.4 port: 12001
+I0827 04:29:10.150219   237 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.4:15413
+I0827 04:29:10.150358   237 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:10.151880   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:10.153475   236 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:04
+I0827 04:29:10.153601   236 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:10 PP6 TP1 EP1] Init torch distributed begin.
+I0827 04:29:10.155849   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:10.155864   237 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:10.157462   242 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:04
+I0827 04:29:10.157585   242 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:10 PP7 TP3 EP3] Init torch distributed begin.
+I0827 04:29:10.171923   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:10.172516   240 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:10.172667   240 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:10.172688   240 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.4 port: 12001
+I0827 04:29:10.172762   240 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.4:15705
+I0827 04:29:10.172977   240 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:10.173573   241 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:04
+I0827 04:29:10.173692   241 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:10 PP7 TP2 EP2] Init torch distributed begin.
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:10.177717   239 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:10.177865   239 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:10.177886   239 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.4 port: 12001
+I0827 04:29:10.177956   239 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.4:16192
+I0827 04:29:10.178146   239 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:10.183638   237 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:10.187090   239 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:10.191001   237 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:10.195258   239 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:10.198271   237 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:10.200984   237 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:10.201025   237 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:10.201079   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.203014   239 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:10.203156   240 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:10.209589   240 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:10.210305   239 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:10.212793   239 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:10.212821   239 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:10.212878   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.216092   240 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:10.219946   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:10.221297   239 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:04
+I0827 04:29:10.221323   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.222187   240 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:10.224284   240 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:10.224304   240 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:10.224344   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.230640   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:10.231778   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:10.231937   240 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:04
+I0827 04:29:10.231962   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.233362   237 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:04
+I0827 04:29:10.233397   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.236478   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:10.237651   240 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:04
+I0827 04:29:10.237670   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.250197   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:10.251711   239 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:04
+I0827 04:29:10.251737   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.263743   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:10.265311   237 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:04
+I0827 04:29:10.265339   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.267139   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:10.268607   240 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:04
+I0827 04:29:10.268632   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.273075   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:10.274682   237 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:04
+I0827 04:29:10.274710   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.276602   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:10.278275   239 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:04
+I0827 04:29:10.278314   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.295709   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:10.297385   240 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:04
+I0827 04:29:10.297510   240 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:10 PP7 TP1 EP1] Init torch distributed begin.
+I0827 04:29:10.299893   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:10.301563   237 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:04
+I0827 04:29:10.301689   237 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:10 PP6 TP2 EP2] Init torch distributed begin.
+I0827 04:29:10.303737   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:10.305307   239 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:04
+I0827 04:29:10.305433   239 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:10 PP7 TP0 EP0] Init torch distributed begin.
+[2026-08-27 04:29:26 PP7 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:26 PP6 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:27 PP6 TP0 EP0] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP6 TP3 EP3] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP6 TP2 EP2] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP6 TP1 EP1] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP7 TP1 EP1] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP7 TP0 EP0] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP7 TP3 EP3] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP7 TP2 EP2] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP7 TP1 EP1] Init torch distributed ends. elapsed=17.45 s, mem usage=0.46 GB
+[2026-08-27 04:29:27 PP7 TP3 EP3] Init torch distributed ends. elapsed=17.59 s, mem usage=0.46 GB
+[2026-08-27 04:29:27 PP7 TP2 EP2] Init torch distributed ends. elapsed=17.57 s, mem usage=0.46 GB
+[2026-08-27 04:29:27 PP7 TP0 EP0] Init torch distributed ends. elapsed=17.44 s, mem usage=0.46 GB
+[2026-08-27 04:29:27 PP6 TP1 EP1] Init torch distributed ends. elapsed=17.59 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP6 TP3 EP3] Init torch distributed ends. elapsed=19.95 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP6 TP0 EP0] Init torch distributed ends. elapsed=18.66 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP6 TP2 EP2] Init torch distributed ends. elapsed=17.44 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP6 TP1 EP1] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP7 TP1 EP1] Load weight begin. avail mem=82.12 GB
+[2026-08-27 04:29:27 PP6 TP0 EP0] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP6 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP7 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP7 TP3 EP3] Load weight begin. avail mem=82.12 GB
+[2026-08-27 04:29:27 PP6 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP7 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP6 TP2 EP2] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP6 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP7 TP2 EP2] Load weight begin. avail mem=82.12 GB
+[2026-08-27 04:29:28 PP7 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP6 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP7 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP6 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP7 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP7 TP0 EP0] Load weight begin. avail mem=82.12 GB
+[2026-08-27 04:29:28 PP6 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP7 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP7 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP6 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP7 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP6 TP3 EP3] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP6 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP6 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP6 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP7 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP7 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP7 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP6 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP7 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP6 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:30:40 PP7 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:40 PP7 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:40 PP7 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:40 PP7 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:40 PP7 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:40 PP7 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:40 PP7 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:40 PP7 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:41 PP7 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:41 PP7 TP1 EP1] Load weight end. elapsed=73.17 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=31.91 GB, mem usage=50.21 GB.
+[2026-08-27 04:30:41 PP7 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:41 PP7 TP0 EP0] Load weight end. elapsed=73.80 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=31.91 GB, mem usage=50.21 GB.
+[2026-08-27 04:30:42 PP6 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:42 PP6 TP1 EP1] Load weight end. elapsed=74.47 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:45 PP6 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:45 PP6 TP2 EP2] Load weight end. elapsed=77.20 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:50 PP7 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:50 PP7 TP3 EP3] Load weight end. elapsed=82.79 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=31.91 GB, mem usage=50.21 GB.
+[2026-08-27 04:30:51 PP6 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:51 PP6 TP3 EP3] Load weight end. elapsed=83.19 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:51 PP7 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:51 PP7 TP2 EP2] Load weight end. elapsed=83.68 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=31.91 GB, mem usage=50.21 GB.
+[2026-08-27 04:30:51 PP7 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:51 PP7 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:52 PP7 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:52 PP7 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:52 PP7 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:52 PP7 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:52 PP7 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:52 PP7 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:53 PP6 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:53 PP6 TP0 EP0] Load weight end. elapsed=85.47 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:53 PP6 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:53 PP6 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:53 PP6 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:53 PP6 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:53 PP6 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:53 PP6 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:53 PP6 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:53 PP6 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:54 PP7 TP3 EP3] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP7 TP2 EP2] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP7 TP1 EP1] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP7 TP0 EP0] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP6 TP3 EP3] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP6 TP2 EP2] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP6 TP1 EP1] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP6 TP0 EP0] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP6 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP6 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP7 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:54 PP6 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP7 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:54 PP7 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:54 PP7 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:54 PP6 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP6 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP6 TP3 EP3] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP6 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP6 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP6 TP2 EP2] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP6 TP1 EP1] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP7 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 16.31 GB
+[2026-08-27 04:30:54 PP7 TP2 EP2] Memory pool end. avail mem=14.95 GB
+[2026-08-27 04:30:54 PP6 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP6 TP0 EP0] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP7 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 16.31 GB
+[2026-08-27 04:30:54 PP7 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 16.31 GB
+[2026-08-27 04:30:54 PP7 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 16.31 GB
+[2026-08-27 04:30:54 PP7 TP3 EP3] Memory pool end. avail mem=14.95 GB
+[2026-08-27 04:30:54 PP7 TP0 EP0] Memory pool end. avail mem=14.95 GB
+[2026-08-27 04:30:54 PP7 TP1 EP1] Memory pool end. avail mem=14.95 GB
+[2026-08-27 04:30:54 PP7 TP2 EP2] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp2_pp7_dp0.json
+[2026-08-27 04:30:54 PP6 TP3 EP3] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp3_pp6_dp0.json
+[2026-08-27 04:30:54 PP6 TP1 EP1] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp1_pp6_dp0.json
+[2026-08-27 04:30:54 PP7 TP3 EP3] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp3_pp7_dp0.json
+[2026-08-27 04:30:54 PP6 TP2 EP2] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp2_pp6_dp0.json
+[2026-08-27 04:30:54 PP6 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:30:54 PP6 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:30:54 PP6 TP0 EP0] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp0_pp6_dp0.json
+[2026-08-27 04:30:54 PP7 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:30:54 PP7 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:30:54 PP7 TP0 EP0] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp0_pp7_dp0.json
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[2026-08-27 04:30:54 PP7 TP1 EP1] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp1_pp7_dp0.json
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 11.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 13.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.76profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.62profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.66profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.72profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.68profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.55profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.12profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  6.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  7.16profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.92profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.15profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.59profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.56profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 19.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  8.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.67profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.40profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.88profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  8.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  6.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.83profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.50profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.32profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.92profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.77profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.94profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.57profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 12.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 12.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 13.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.36profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.66profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  6.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.72profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  6.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.22profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.65profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.13profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.03profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.46profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.59profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.16profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  6.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.96profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.76profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.67profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  8.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.27profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.88profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.01profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.57profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.12profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 11.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  2.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.68profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.90profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.43profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.92profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  8.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.66profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  6.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  7.13profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.32profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.71profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:05,  1.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.16profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.31profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.77profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.28profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.48profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.87profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.40profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 13.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.35profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.98profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.88profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.73profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  3.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.88profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.09profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.20profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.97profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.83profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.47profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.40profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.60profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  8.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  7.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.97profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.05profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.05profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  6.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.49profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.72profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.19profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 10.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.91profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.64profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.58profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  3.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.35profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  9.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 12.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.09profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:02,  2.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 12.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.15profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.76profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.49profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.03profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.79profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.96profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.47profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.65profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.55profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.03profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  9.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.63profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.98profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:05,  1.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:01,  3.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.50profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  6.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.94profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.13profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.78profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.53profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.35profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.56profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.36profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:01<00:06,  1.06s/profile]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  8.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.91profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.66profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  8.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  7.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:02,  2.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.83profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.01profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.66profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.13profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.59profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  8.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  8.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.71profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.29profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.97profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.88profile/s]
+[2026-08-27 04:31:27 PP6 TP3 EP3] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.86profile/s]
+[2026-08-27 04:31:27 PP6 TP1 EP1] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.24profile/s]
+[2026-08-27 04:31:27 PP6 TP2 EP2] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.33profile/s]
+[2026-08-27 04:31:27 PP6 TP0 EP0] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.56profile/s][2026-08-27 04:31:27 PP6 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:27 PP6 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:27 PP6 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:27 PP6 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:27 PP6 TP0 EP0] max_total_num_tokens=3801408, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=54, context_len=1048576, available_gpu_mem=16.52 GB
+[2026-08-27 04:31:27 PP6 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:27 PP6 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:27 PP6 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:27 PP6 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
+[2026-08-27 04:31:28 PP7 TP2 EP2] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.32profile/s]
+[2026-08-27 04:31:28 PP7 TP3 EP3] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.60profile/s]
+[2026-08-27 04:31:28 PP7 TP0 EP0] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.60profile/s]
+[2026-08-27 04:31:28 PP7 TP1 EP1] FlashInfer autotune completed.
+[2026-08-27 04:31:28 PP7 TP2 EP2] multimem all-gather disabled (no multicast for world_size=4)
+[2026-08-27 04:31:28 PP7 TP3 EP3] multimem all-gather disabled (no multicast for world_size=4)
+[2026-08-27 04:31:28 PP7 TP1 EP1] multimem all-gather disabled (no multicast for world_size=4)
+[2026-08-27 04:31:28 PP7 TP0 EP0] multimem all-gather disabled (no multicast for world_size=4)
+[2026-08-27 04:31:28 PP7 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:28 PP7 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:28 PP7 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:28 PP7 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:28 PP7 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:28 PP7 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:28 PP7 TP0 EP0] max_total_num_tokens=3801408, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=54, context_len=1048576, available_gpu_mem=12.14 GB
+[2026-08-27 04:31:28 PP7 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:28 PP7 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+W0827 04:31:31.959518  2596 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5efc8ec0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.959637  2598 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ee654c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.959640  2597 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5eda3c80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.959542  2595 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5efc6f80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.959741  2599 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5eeaf500, len: 55296: Operation not permitted [1]
+W0827 04:31:31.959794  2600 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ef4e380, len: 55296: Operation not permitted [1]
+W0827 04:31:31.959827  2601 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5f548500, len: 6912: Operation not permitted [1]
+W0827 04:31:31.959918  2602 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5eec3bc0, len: 13824: Operation not permitted [1]
+W0827 04:31:31.960085  2603 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5f01bcc0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.960222  2604 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5f4c8ac0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.980162  2623 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60ba35c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.980152  2624 memory_location.cpp:73] Failed to get NUMA node, addr: 0x609ccf40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.980242  2625 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60d38600, len: 6912: Operation not permitted [1]
+W0827 04:31:31.980361  2626 memory_location.cpp:73] Failed to get NUMA node, addr: 0x609546c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.980516  2628 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60c49a00, len: 55296: Operation not permitted [1]
+W0827 04:31:31.980537  2627 memory_location.cpp:73] Failed to get NUMA node, addr: 0x607c6800, len: 55296: Operation not permitted [1]
+W0827 04:31:31.980547  2629 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60d11800, len: 6912: Operation not permitted [1]
+W0827 04:31:31.980834  2630 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60a1b740, len: 13824: Operation not permitted [1]
+W0827 04:31:31.981053  2632 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60ce5000, len: 6912: Operation not permitted [1]
+W0827 04:31:31.981062  2631 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60aef300, len: 6912: Operation not permitted [1]
+W0827 04:31:32.117022  2651 memory_location.cpp:73] Failed to get NUMA node, addr: 0x74c79d40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.117132  2652 memory_location.cpp:73] Failed to get NUMA node, addr: 0x74f99700, len: 6912: Operation not permitted [1]
+W0827 04:31:32.117317  2653 memory_location.cpp:73] Failed to get NUMA node, addr: 0x7546c740, len: 6912: Operation not permitted [1]
+W0827 04:31:32.117331  2654 memory_location.cpp:73] Failed to get NUMA node, addr: 0x74f4f140, len: 6912: Operation not permitted [1]
+W0827 04:31:32.117414  2656 memory_location.cpp:73] Failed to get NUMA node, addr: 0x750037c0, len: 55296: Operation not permitted [1]
+W0827 04:31:32.117445  2655 memory_location.cpp:73] Failed to get NUMA node, addr: 0x74bf1200, len: 55296: Operation not permitted [1]
+W0827 04:31:32.117705  2657 memory_location.cpp:73] Failed to get NUMA node, addr: 0x74f80080, len: 6912: Operation not permitted [1]
+W0827 04:31:32.117738  2658 memory_location.cpp:73] Failed to get NUMA node, addr: 0x74fbd2c0, len: 13824: Operation not permitted [1]
+W0827 04:31:32.117790  2659 memory_location.cpp:73] Failed to get NUMA node, addr: 0x74fa5e40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.118012  2660 memory_location.cpp:73] Failed to get NUMA node, addr: 0x74c24880, len: 6912: Operation not permitted [1]
+W0827 04:31:32.137427  2677 memory_location.cpp:73] Failed to get NUMA node, addr: 0x685bea40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.137485  2678 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6865c240, len: 6912: Operation not permitted [1]
+W0827 04:31:32.137612  2680 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6832f280, len: 6912: Operation not permitted [1]
+W0827 04:31:32.137683  2679 memory_location.cpp:73] Failed to get NUMA node, addr: 0x68511ec0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.137791  2681 memory_location.cpp:73] Failed to get NUMA node, addr: 0x687ae280, len: 55296: Operation not permitted [1]
+W0827 04:31:32.137800  2682 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6863aec0, len: 55296: Operation not permitted [1]
+W0827 04:31:32.137928  2683 memory_location.cpp:73] Failed to get NUMA node, addr: 0x685f4540, len: 6912: Operation not permitted [1]
+W0827 04:31:32.138145  2685 memory_location.cpp:73] Failed to get NUMA node, addr: 0x68b70f80, len: 6912: Operation not permitted [1]
+W0827 04:31:32.138144  2684 memory_location.cpp:73] Failed to get NUMA node, addr: 0x686e87c0, len: 13824: Operation not permitted [1]
+W0827 04:31:32.138219  2686 memory_location.cpp:73] Failed to get NUMA node, addr: 0x685491c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.187127  2706 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8eaeff40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.187187  2707 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8eaa1740, len: 6912: Operation not permitted [1]
+W0827 04:31:32.187136  2705 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8e5f2600, len: 6912: Operation not permitted [1]
+W0827 04:31:32.187413  2709 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ec4eb00, len: 55296: Operation not permitted [1]
+W0827 04:31:32.187405  2708 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ea3b800, len: 6912: Operation not permitted [1]
+W0827 04:31:32.187651  2710 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ed92ec0, len: 55296: Operation not permitted [1]
+W0827 04:31:32.187799  2711 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8eb3fbc0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.188133  2712 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ed0e1c0, len: 13824: Operation not permitted [1]
+W0827 04:31:32.188231  2713 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8e9d20c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.188272  2714 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8f2c7a80, len: 6912: Operation not permitted [1]
+W0827 04:31:32.252096  2732 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72b6b180, len: 6912: Operation not permitted [1]
+W0827 04:31:32.252050  2731 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72be0600, len: 6912: Operation not permitted [1]
+W0827 04:31:32.252147  2733 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72716c00, len: 6912: Operation not permitted [1]
+W0827 04:31:32.252276  2734 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72739940, len: 6912: Operation not permitted [1]
+W0827 04:31:32.252379  2735 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72da8a80, len: 55296: Operation not permitted [1]
+W0827 04:31:32.252439  2736 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72d15800, len: 55296: Operation not permitted [1]
+W0827 04:31:32.252678  2737 memory_location.cpp:73] Failed to get NUMA node, addr: 0x7315db40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.252701  2738 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72e0e700, len: 13824: Operation not permitted [1]
+W0827 04:31:32.252745  2739 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72f5f480, len: 6912: Operation not permitted [1]
+W0827 04:31:32.252832  2740 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72b295c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.268508  2760 memory_location.cpp:73] Failed to get NUMA node, addr: 0x69eefc00, len: 6912: Operation not permitted [1]
+W0827 04:31:32.268532  2759 memory_location.cpp:73] Failed to get NUMA node, addr: 0x69ec8680, len: 6912: Operation not permitted [1]
+W0827 04:31:32.268577  2761 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6a02e280, len: 6912: Operation not permitted [1]
+W0827 04:31:32.268708  2762 memory_location.cpp:73] Failed to get NUMA node, addr: 0x69d75dc0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.268766  2763 memory_location.cpp:73] Failed to get NUMA node, addr: 0x69fc4880, len: 55296: Operation not permitted [1]
+W0827 04:31:32.268886  2764 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6a0c8540, len: 55296: Operation not permitted [1]
+W0827 04:31:32.268975  2765 memory_location.cpp:73] Failed to get NUMA node, addr: 0x69c75940, len: 6912: Operation not permitted [1]
+W0827 04:31:32.269125  2766 memory_location.cpp:73] Failed to get NUMA node, addr: 0x698ae1c0, len: 13824: Operation not permitted [1]
+W0827 04:31:32.269232  2767 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6a5457c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.269354  2768 memory_location.cpp:73] Failed to get NUMA node, addr: 0x69c8fe80, len: 6912: Operation not permitted [1]
+W0827 04:31:32.281724  2785 memory_location.cpp:73] Failed to get NUMA node, addr: 0x678e47c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.281751  2786 memory_location.cpp:73] Failed to get NUMA node, addr: 0x67a39b80, len: 6912: Operation not permitted [1]
+W0827 04:31:32.281852  2787 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6798d9c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.281994  2788 memory_location.cpp:73] Failed to get NUMA node, addr: 0x67981dc0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.282085  2789 memory_location.cpp:73] Failed to get NUMA node, addr: 0x67823300, len: 55296: Operation not permitted [1]
+W0827 04:31:32.282146  2790 memory_location.cpp:73] Failed to get NUMA node, addr: 0x67a3bb00, len: 55296: Operation not permitted [1]
+W0827 04:31:32.282214  2791 memory_location.cpp:73] Failed to get NUMA node, addr: 0x67cdad40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.282285  2792 memory_location.cpp:73] Failed to get NUMA node, addr: 0x68028e40, len: 13824: Operation not permitted [1]
+W0827 04:31:32.282544  2794 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6796cf40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.282545  2793 memory_location.cpp:73] Failed to get NUMA node, addr: 0x67fa6600, len: 6912: Operation not permitted [1]
+[2026-08-27 04:31:32] Dummy health check server started in background thread at 0.0.0.0:30000
+[rank27]:[W827 04:34:16.127713860 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank25]:[W827 04:34:16.128748517 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank24]:[W827 04:34:16.129831915 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank26]:[W827 04:34:16.136239904 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank26]:[W827 04:35:04.964905528 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank25]:[W827 04:35:04.966225881 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank27]:[W827 04:35:04.966239242 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank24]:[W827 04:35:04.966301664 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank31]:[W827 04:35:04.990701389 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank28]:[W827 04:35:04.991914270 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank29]:[W827 04:35:04.994546241 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank30]:[W827 04:35:04.999020079 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank31]:[W827 04:35:10.873647776 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank30]:[W827 04:35:10.873901815 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank28]:[W827 04:35:10.874227564 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank29]:[W827 04:35:10.885318244 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[2026-08-27 04:35:10 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.29
+[2026-08-27 04:35:10 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.29
+[2026-08-27 04:35:29 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.39
+[2026-08-27 04:35:29 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.39
+[2026-08-27 04:35:31 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25.93
+[2026-08-27 04:35:31 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25.95
+[2026-08-27 04:43:23 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.14
+[2026-08-27 04:43:23 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.14
+[2026-08-27 04:48:14 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.22
+[2026-08-27 04:48:14 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.22
+[2026-08-27 04:48:14 PP6 TP1 EP1] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP6 TP2 EP2] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP7 TP2 EP2] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP6 TP0 EP0] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP7 TP0 EP0] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP7 TP1 EP1] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP6 TP3 EP3] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP7 TP3 EP3] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:57:14 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.12
+[2026-08-27 04:57:14 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.12
+[2026-08-27 05:04:21 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1088, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2.55
+[2026-08-27 05:04:21 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1088, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2.55
+[2026-08-27 05:04:41 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.19
+[2026-08-27 05:04:41 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.18
+[2026-08-27 05:05:39 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1.10
+[2026-08-27 05:05:39 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1.10
+[2026-08-27 05:07:33 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 72.18
+[2026-08-27 05:07:33 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 72.18
+[2026-08-27 05:07:33 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 389087.41
+[2026-08-27 05:07:33 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 459604.95
+[2026-08-27 05:07:33 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32068.03
+[2026-08-27 05:07:33 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 36484.58
+[2026-08-27 05:07:40 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8673, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1083.15
+[2026-08-27 05:07:40 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8673, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1082.87
+[2026-08-27 05:07:40 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 481, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 423255.51
+[2026-08-27 05:07:40 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 481, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 454364.97
+[2026-08-27 05:07:40 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 34494.75
+[2026-08-27 05:07:40 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33090.16
+[2026-08-27 05:07:48 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8656, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1082.19
+[2026-08-27 05:07:48 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8656, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1082.17
+[2026-08-27 05:07:48 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 464, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 443415.27
+[2026-08-27 05:07:48 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 464, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 422320.55
+[2026-08-27 05:07:48 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 30852.82
+[2026-08-27 05:07:48 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33407.06
+[2026-08-27 05:07:55 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8663, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1087.54
+[2026-08-27 05:07:55 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8663, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1087.42
+[2026-08-27 05:07:55 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 471, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 437681.61
+[2026-08-27 05:07:55 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 471, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 442367.54
+[2026-08-27 05:07:55 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 31561.40
+[2026-08-27 05:07:55 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 31433.36
+[2026-08-27 05:08:03 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8659, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1083.03
+[2026-08-27 05:08:03 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8659, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1082.87
+[2026-08-27 05:08:03 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 467, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 376910.00
+[2026-08-27 05:08:03 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 467, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 408825.56
+[2026-08-27 05:08:03 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32102.57
+[2026-08-27 05:08:03 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33296.68
+[2026-08-27 05:08:11 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8692, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1085.73
+[2026-08-27 05:08:11 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8692, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1085.61
+[2026-08-27 05:08:11 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 500, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 434356.57
+[2026-08-27 05:08:11 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 500, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 383780.71
+[2026-08-27 05:08:11 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 27941.29
+[2026-08-27 05:08:11 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32876.54
+[2026-08-27 05:08:18 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8557, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1090.32
+[2026-08-27 05:08:18 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8557, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1090.65
+[2026-08-27 05:08:18 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 365, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 477756.06
+[2026-08-27 05:08:18 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 365, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 457694.49
+[2026-08-27 05:08:18 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 27080.22
+[2026-08-27 05:08:18 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 27388.57
+[2026-08-27 05:08:26 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8683, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1083.79
+[2026-08-27 05:08:26 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8683, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1083.51
+[2026-08-27 05:08:26 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 491, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 434750.14
+[2026-08-27 05:08:26 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 491, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 447315.13
+[2026-08-27 05:08:26 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33020.44
+[2026-08-27 05:08:26 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33110.34
+[2026-08-27 05:11:45 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.32
+[2026-08-27 05:11:45 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.32
+[2026-08-27 05:14:42 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1088, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6.16
+[2026-08-27 05:14:42 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1088, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6.16
+[2026-08-27 05:17:47 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 44.21
+[2026-08-27 05:17:47 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 44.21
+[2026-08-27 05:17:47 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 461237.58
+[2026-08-27 05:17:47 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 443473.87
+[2026-08-27 05:17:47 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 30106.68
+[2026-08-27 05:17:47 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32700.87
+[2026-08-27 05:17:55 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8673, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1086.24
+[2026-08-27 05:17:55 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8673, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1086.14
+[2026-08-27 05:17:55 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 481, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 438855.61
+[2026-08-27 05:17:55 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 481, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 448853.01
+[2026-08-27 05:17:55 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32917.40
+[2026-08-27 05:17:55 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 35669.59
+[2026-08-27 05:18:02 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8656, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1088.38
+[2026-08-27 05:18:02 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8656, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1088.12
+[2026-08-27 05:18:02 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 464, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 432657.94
+[2026-08-27 05:18:02 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 464, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 441680.19
+[2026-08-27 05:18:02 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32420.83
+[2026-08-27 05:18:02 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33219.83
+[2026-08-27 05:18:10 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8663, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1087.85
+[2026-08-27 05:18:10 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8663, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1087.68
+[2026-08-27 05:18:10 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 471, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 433552.65
+[2026-08-27 05:18:10 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 471, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 448022.33
+[2026-08-27 05:18:10 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32265.40
+[2026-08-27 05:18:10 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32427.35
+[2026-08-27 05:18:18 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8659, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1086.07
+[2026-08-27 05:18:18 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8659, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1085.68
+[2026-08-27 05:18:18 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 467, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 365245.86
+[2026-08-27 05:18:18 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 467, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 360696.15
+[2026-08-27 05:18:18 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 27416.04
+[2026-08-27 05:18:18 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 35182.90
+[2026-08-27 05:18:25 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8692, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1087.81
+[2026-08-27 05:18:25 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8692, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1087.87
+[2026-08-27 05:18:25 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 500, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 494405.53
+[2026-08-27 05:18:25 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 500, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 406159.77
+[2026-08-27 05:18:25 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 28692.54
+[2026-08-27 05:18:25 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 35731.06
+[2026-08-27 05:18:33 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8557, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1090.88
+[2026-08-27 05:18:33 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8557, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1090.72
+[2026-08-27 05:18:33 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 365, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 476480.55
+[2026-08-27 05:18:33 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 365, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 465689.14
+[2026-08-27 05:18:33 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 24743.50
+[2026-08-27 05:18:33 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 26986.87
+[2026-08-27 05:18:40 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8683, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1087.65
+[2026-08-27 05:18:40 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8683, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1087.47
+[2026-08-27 05:18:40 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 491, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 444510.88
+[2026-08-27 05:18:40 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 491, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 381334.71
+[2026-08-27 05:18:40 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 27475.75
+[2026-08-27 05:18:40 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 35546.91
+[2026-08-27 05:21:48 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 2, #pending-token: 8652, #bootstrap-req: 2, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 43.63
+[2026-08-27 05:21:49 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 3, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9876.33
+[2026-08-27 05:21:50 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 2, #pending-token: 8652, #bootstrap-req: 2, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 43.26
+[2026-08-27 05:21:50 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 3, #pending-token: 9185, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8052.60
+[2026-08-27 05:21:50 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 3, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9996.10
+[2026-08-27 05:21:51 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 2, #pending-token: 993, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9980.46
+[2026-08-27 05:21:51 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 3, #pending-token: 9185, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8828.21
+[2026-08-27 05:21:52 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 2, #pending-token: 9680, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9798.99
+[2026-08-27 05:21:52 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 2, #pending-token: 993, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9906.89
+[2026-08-27 05:21:52 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1488, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9933.78
+[2026-08-27 05:21:53 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 2, #pending-token: 9680, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9164.38
+[2026-08-27 05:21:53 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 10199, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9100.99
+[2026-08-27 05:21:54 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1488, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9595.39
+[2026-08-27 05:21:54 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 18858, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9669.02
+[2026-08-27 05:21:55 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 10199, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9348.80
+[2026-08-27 05:21:55 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 27591, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9301.63
+[2026-08-27 05:21:56 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 18858, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9457.16
+[2026-08-27 05:21:56 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 53023, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9526.90
+[2026-08-27 05:21:57 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 27591, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9435.09
+[2026-08-27 05:21:57 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 44876, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9774.31
+[2026-08-27 05:21:57 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 53023, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 15230.20
+[2026-08-27 05:21:57 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 36684, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 18786.53
+[2026-08-27 05:21:57 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 44876, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 521404.18
+[2026-08-27 05:21:57 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 28504, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 355637.74
+[2026-08-27 05:21:57 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 36684, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 388735.22
+[2026-08-27 05:21:57 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 20312, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 74956.78
+[2026-08-27 05:21:57 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 28504, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 73931.58
+[2026-08-27 05:21:58 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 12139, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 16102.75
+[2026-08-27 05:21:58 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 20312, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 14970.35
+[2026-08-27 05:21:58 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 3947, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 12109.59
+[2026-08-27 05:21:59 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 12139, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10923.81
+[2026-08-27 05:21:59 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 3968, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 5070.20
+[2026-08-27 05:21:59 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 3947, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11542.46
+[2026-08-27 05:22:00 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 3968, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4785.41
+[2026-08-27 05:22:08 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 0, #pending-token: 8608, #bootstrap-req: 4, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 940.83
+[2026-08-27 05:22:09 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 416, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10204.68
+[2026-08-27 05:22:09 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 0, #pending-token: 8608, #bootstrap-req: 4, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 868.76
+[2026-08-27 05:22:10 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 9149, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8740.29
+[2026-08-27 05:22:10 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 416, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 29834.98
+[2026-08-27 05:22:10 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 957, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9976.76
+[2026-08-27 05:22:11 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 9149, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7962.19
+[2026-08-27 05:22:11 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 9638, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9468.71
+[2026-08-27 05:22:12 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 957, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6057.13
+[2026-08-27 05:22:12 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1446, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8241.57
+[2026-08-27 05:22:13 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 9638, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10874.51
+[2026-08-27 05:22:13 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 10033, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10016.49
+[2026-08-27 05:22:14 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1446, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7794.33
+[2026-08-27 05:22:14 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1841, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8525.06
+[2026-08-27 05:22:15 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 10033, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10735.67
+[2026-08-27 05:22:15 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1856, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2257.01
+[2026-08-27 05:22:16 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1841, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8114.90
+[2026-08-27 05:22:16 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8540, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8613.20
+[2026-08-27 05:22:16 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1856, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2421.25
+[2026-08-27 05:22:17 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 17066, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10036.34
+[2026-08-27 05:22:17 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8540, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8278.93
+[2026-08-27 05:22:18 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8910, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9399.94
+[2026-08-27 05:22:18 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 17066, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 36997.59
+[2026-08-27 05:22:18 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 17466, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 51312.40
+[2026-08-27 05:22:18 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8910, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 411464.53
+[2026-08-27 05:22:18 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9324, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 66779.98
+[2026-08-27 05:22:18 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 17466, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 68447.00
+[2026-08-27 05:22:18 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 17751, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 16323.38
+[2026-08-27 05:22:18 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9324, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 15086.19
+[2026-08-27 05:22:19 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9579, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 12098.16
+[2026-08-27 05:22:19 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 17751, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10943.79
+[2026-08-27 05:22:20 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1387, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10522.42
+[2026-08-27 05:22:20 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9579, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11560.43
+[2026-08-27 05:22:20 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1408, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2179.97
+[2026-08-27 05:22:21 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1387, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9951.54
+[2026-08-27 05:22:21 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1408, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1996.17
+[2026-08-27 05:22:29 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8547, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1014.77
+[2026-08-27 05:22:29 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 355, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10133.87
+[2026-08-27 05:22:30 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8547, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 936.03
+[2026-08-27 05:22:30 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8932, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8653.72
+[2026-08-27 05:22:30 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 355, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 34268.89
+[2026-08-27 05:22:31 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 740, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10007.94
+[2026-08-27 05:22:31 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8932, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7704.93
+[2026-08-27 05:22:32 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9259, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9600.28
+[2026-08-27 05:22:33 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 740, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6120.45
+[2026-08-27 05:22:33 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1067, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8368.71
+[2026-08-27 05:22:34 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9259, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10965.05
+[2026-08-27 05:22:34 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9745, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9275.86
+[2026-08-27 05:22:35 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1067, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7846.08
+[2026-08-27 05:22:35 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1553, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9172.31
+[2026-08-27 05:22:35 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9745, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10985.13
+[2026-08-27 05:22:36 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1600, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1929.98
+[2026-08-27 05:22:36 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1553, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7933.67
+[2026-08-27 05:22:36 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8602, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8619.09
+[2026-08-27 05:22:37 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1600, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2101.55
+[2026-08-27 05:22:37 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 410, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10046.41
+[2026-08-27 05:22:38 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8602, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8223.98
+[2026-08-27 05:22:38 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8696, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9435.25
+[2026-08-27 05:22:38 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 410, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 45497.16
+[2026-08-27 05:22:38 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 504, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 66281.64
+[2026-08-27 05:22:38 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8696, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 425064.62
+[2026-08-27 05:22:38 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9164, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 67906.51
+[2026-08-27 05:22:38 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 504, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 69370.77
+[2026-08-27 05:22:39 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 972, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 16327.79
+[2026-08-27 05:22:39 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9164, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 15103.13
+[2026-08-27 05:22:40 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9322, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 12119.18
+[2026-08-27 05:22:40 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 972, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10991.29
+[2026-08-27 05:22:40 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1130, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10534.55
+[2026-08-27 05:22:40 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9322, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11588.54
+[2026-08-27 05:22:41 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1152, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1784.29
+[2026-08-27 05:22:41 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1130, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10014.78
+[2026-08-27 05:22:42 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1152, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1638.04
+[2026-08-27 05:22:49 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8602, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1020.06
+[2026-08-27 05:22:50 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 410, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10100.36
+[2026-08-27 05:22:51 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8602, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 939.88
+[2026-08-27 05:22:51 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 9105, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8814.04
+[2026-08-27 05:22:51 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 410, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 31402.72
+[2026-08-27 05:22:52 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 913, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9823.70
+[2026-08-27 05:22:52 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 9105, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7950.01
+[2026-08-27 05:22:52 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9375, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9520.99
+[2026-08-27 05:22:53 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 913, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6088.53
+[2026-08-27 05:22:53 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1183, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8593.45
+[2026-08-27 05:22:54 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9375, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11034.51
+[2026-08-27 05:22:54 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9868, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9888.23
+[2026-08-27 05:22:55 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1183, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7825.73
+[2026-08-27 05:22:55 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1676, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8507.30
+[2026-08-27 05:22:56 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9868, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10702.83
+[2026-08-27 05:22:56 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2096.94
+[2026-08-27 05:22:57 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1676, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8175.44
+[2026-08-27 05:22:57 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8467, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8745.10
+[2026-08-27 05:22:58 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2235.70
+[2026-08-27 05:22:58 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 275, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9728.14
+[2026-08-27 05:22:59 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8467, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8423.31
+[2026-08-27 05:22:59 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8988, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9773.55
+[2026-08-27 05:22:59 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 275, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 30877.72
+[2026-08-27 05:22:59 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 796, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 41016.53
+[2026-08-27 05:22:59 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8988, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 458929.33
+[2026-08-27 05:22:59 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9363, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 73474.43
+[2026-08-27 05:22:59 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 796, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 74700.60
+[2026-08-27 05:22:59 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 1171, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 16297.37
+[2026-08-27 05:23:00 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9363, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 15107.75
+[2026-08-27 05:23:00 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9880, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 12178.33
+[2026-08-27 05:23:00 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 1171, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11020.95
+[2026-08-27 05:23:01 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1688, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10567.86
+[2026-08-27 05:23:01 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9880, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11615.07
+[2026-08-27 05:23:02 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2677.95
+[2026-08-27 05:23:02 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1688, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10008.16
+[2026-08-27 05:23:02 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2457.28
+[2026-08-27 05:23:10 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8601, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1020.50
+[2026-08-27 05:23:10 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 409, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10128.83
+[2026-08-27 05:23:11 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8601, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 941.05
+[2026-08-27 05:23:11 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 9104, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8792.53
+[2026-08-27 05:23:11 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 409, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 31651.22
+[2026-08-27 05:23:12 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 912, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9996.90
+[2026-08-27 05:23:12 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 9104, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7936.01
+[2026-08-27 05:23:13 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9485, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9519.44
+[2026-08-27 05:23:14 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 912, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6079.95
+[2026-08-27 05:23:14 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1293, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8474.47
+[2026-08-27 05:23:15 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9485, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10878.89
+[2026-08-27 05:23:15 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9900, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9816.19
+[2026-08-27 05:23:16 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1293, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7901.51
+[2026-08-27 05:23:16 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1708, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8613.59
+[2026-08-27 05:23:16 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9900, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10766.20
+[2026-08-27 05:23:17 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2099.82
+[2026-08-27 05:23:17 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1708, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8087.47
+[2026-08-27 05:23:18 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8525, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8707.95
+[2026-08-27 05:23:18 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2244.91
+[2026-08-27 05:23:18 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 333, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10166.80
+[2026-08-27 05:23:19 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8525, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8506.15
+[2026-08-27 05:23:19 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8909, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9459.25
+[2026-08-27 05:23:19 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 333, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33322.73
+[2026-08-27 05:23:19 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 717, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 45670.17
+[2026-08-27 05:23:19 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8909, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 667975.55
+[2026-08-27 05:23:19 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9452, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 458625.25
+[2026-08-27 05:23:19 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 717, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 444740.14
+[2026-08-27 05:23:19 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9452, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 446180.29
+[2026-08-27 05:23:19 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 1260, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 337352.56
+[2026-08-27 05:23:19 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 1260, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 382567.64
+[2026-08-27 05:23:19 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9830, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 429427.21
+[2026-08-27 05:23:19 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1638, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 448443.92
+[2026-08-27 05:23:19 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9830, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 396809.23
+[2026-08-27 05:23:19 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1664, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 93514.06
+[2026-08-27 05:23:19 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1638, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 464006.76
+[2026-08-27 05:23:20 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1664, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 108432.29
+[2026-08-27 05:32:52 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.11
+[2026-08-27 05:32:52 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.11
+[2026-08-27 05:35:54 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 45.15
+[2026-08-27 05:35:54 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 45.15
+[2026-08-27 05:35:54 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 412328.11
+[2026-08-27 05:35:54 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 421944.54
+[2026-08-27 05:35:54 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 31302.30
+[2026-08-27 05:35:54 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32994.85
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-20260827-1325/logs/manual_router.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-20260827-1325/logs/manual_router.log
new file mode 100644
index 0000000..3755f8c
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-20260827-1325/logs/manual_router.log
@@ -0,0 +1,133 @@
+MiniLB is only for debugging purposes, it only supports random policy!
+[MiniLB] Overriding policy to random
+INFO:     Started server process [1]
+INFO:     Waiting for application startup.
+INFO:     Application startup complete.
+INFO:     Uvicorn running on http://0.0.0.0:31000 (Press CTRL+C to quit)
+INFO:     174.1.60.1:51774 - "GET /health HTTP/1.1" 200 OK
+INFO:     174.1.60.1:39128 - "GET /health HTTP/1.1" 200 OK
+INFO:     174.1.60.1:43568 - "GET /health HTTP/1.1" 200 OK
+INFO:     174.1.60.1:58114 - "GET /v1/models HTTP/1.1" 200 OK
+INFO:     174.1.60.1:58122 - "GET /v1/models HTTP/1.1" 200 OK
+INFO:     174.1.60.1:34030 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:52970 - "GET /server_info HTTP/1.1" 200 OK
+INFO:     174.1.60.1:52972 - "GET /server_info HTTP/1.1" 200 OK
+INFO:     174.1.60.1:43934 - "GET /health HTTP/1.1" 200 OK
+INFO:     174.1.60.1:57120 - "GET /v1/models HTTP/1.1" 200 OK
+/usr/local/lib/python3.12/dist-packages/fastapi/routing.py:350: FastAPIDeprecationWarning: ORJSONResponse is deprecated, FastAPI now serializes data directly to JSON bytes via Pydantic when a return type or response model is set, which is faster and doesn't need a custom response class. Read more in the FastAPI docs: https://fastapi.tiangolo.com/advanced/custom-response/#orjson-or-response-model and https://fastapi.tiangolo.com/tutorial/response-model/
+  return await dependant.call(**values)
+INFO:     174.1.60.1:57126 - "GET /v1/models HTTP/1.1" 200 OK
+INFO:     174.1.60.1:55592 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:55598 - "GET /server_info HTTP/1.1" 200 OK
+INFO:     174.1.60.1:55600 - "GET /server_info HTTP/1.1" 200 OK
+INFO:     174.1.60.1:39092 - "GET /v1/models HTTP/1.1" 200 OK
+INFO:     174.1.60.1:39100 - "GET /v1/models HTTP/1.1" 200 OK
+INFO:     174.1.60.1:42280 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:51506 - "GET /server_info HTTP/1.1" 200 OK
+INFO:     174.1.60.1:51522 - "GET /server_info HTTP/1.1" 200 OK
+/usr/local/lib/python3.12/dist-packages/sglang_router/mini_lb.py:390: FastAPIDeprecationWarning: ORJSONResponse is deprecated, FastAPI now serializes data directly to JSON bytes via Pydantic when a return type or response model is set, which is faster and doesn't need a custom response class. Read more in the FastAPI docs: https://fastapi.tiangolo.com/advanced/custom-response/#orjson-or-response-model and https://fastapi.tiangolo.com/tutorial/response-model/
+  return await lb.generate(
+INFO:     174.1.60.1:59130 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:54978 - "GET /v1/models HTTP/1.1" 200 OK
+INFO:     174.1.60.1:54992 - "GET /v1/models HTTP/1.1" 200 OK
+INFO:     174.1.60.1:45422 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:49916 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:52084 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:52096 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:45122 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:35498 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:50726 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:52480 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:52490 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:40770 - "GET /server_info HTTP/1.1" 200 OK
+INFO:     174.1.60.1:40776 - "GET /server_info HTTP/1.1" 200 OK
+INFO:     174.1.60.1:41252 - "GET /health HTTP/1.1" 200 OK
+INFO:     174.1.60.1:53822 - "GET /v1/models HTTP/1.1" 200 OK
+INFO:     174.1.60.1:53832 - "GET /v1/models HTTP/1.1" 200 OK
+INFO:     174.1.60.1:36760 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:36764 - "GET /server_info HTTP/1.1" 200 OK
+INFO:     174.1.60.1:36772 - "GET /server_info HTTP/1.1" 200 OK
+INFO:     174.1.60.1:47514 - "GET /v1/models HTTP/1.1" 200 OK
+INFO:     174.1.60.1:47520 - "GET /v1/models HTTP/1.1" 200 OK
+INFO:     174.1.60.1:35562 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:35578 - "GET /server_info HTTP/1.1" 200 OK
+INFO:     174.1.60.1:35586 - "GET /server_info HTTP/1.1" 200 OK
+INFO:     174.1.60.1:40864 - "GET /v1/models HTTP/1.1" 200 OK
+INFO:     174.1.60.1:40878 - "GET /v1/models HTTP/1.1" 200 OK
+INFO:     174.1.60.1:35998 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:47822 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:41780 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:49022 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:49034 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:36558 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:49056 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:50776 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:50790 - "GET /server_info HTTP/1.1" 200 OK
+INFO:     174.1.60.1:50802 - "GET /server_info HTTP/1.1" 200 OK
+INFO:     174.1.60.1:45642 - "GET /v1/models HTTP/1.1" 200 OK
+INFO:     174.1.60.1:45656 - "GET /v1/models HTTP/1.1" 200 OK
+INFO:     174.1.60.1:47786 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:47794 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:47804 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:47806 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:47812 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:47816 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:47822 - "POST /generate HTTP/1.1" 200 OK
+Unclosed connection
+client_connection: Connection<ConnectionKey(host='174.1.60.1', port=30000, is_ssl=False, ssl=True, proxy=None, proxy_auth=None, proxy_headers_hash=None, server_hostname=None)>
+Unclosed connection
+client_connection: Connection<ConnectionKey(host='174.1.60.1', port=30000, is_ssl=False, ssl=True, proxy=None, proxy_auth=None, proxy_headers_hash=None, server_hostname=None)>
+INFO:     174.1.60.1:47836 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:47582 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:47594 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:47602 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:47604 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:47618 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:47626 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:47640 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:47644 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:50356 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:50366 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:50374 - "POST /generate HTTP/1.1" 200 OK
+Unclosed connection
+client_connection: Connection<ConnectionKey(host='174.1.60.1', port=30000, is_ssl=False, ssl=True, proxy=None, proxy_auth=None, proxy_headers_hash=None, server_hostname=None)>
+Unclosed connection
+client_connection: Connection<ConnectionKey(host='174.1.60.1', port=30000, is_ssl=False, ssl=True, proxy=None, proxy_auth=None, proxy_headers_hash=None, server_hostname=None)>
+Unclosed connection
+client_connection: Connection<ConnectionKey(host='174.1.60.1', port=30000, is_ssl=False, ssl=True, proxy=None, proxy_auth=None, proxy_headers_hash=None, server_hostname=None)>
+INFO:     174.1.60.1:50376 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:50388 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:50402 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:50414 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:46792 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:53680 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:53696 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:53708 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:53714 - "POST /generate HTTP/1.1" 200 OK
+Unclosed connection
+client_connection: Connection<ConnectionKey(host='174.1.60.1', port=30000, is_ssl=False, ssl=True, proxy=None, proxy_auth=None, proxy_headers_hash=None, server_hostname=None)>
+Unclosed connection
+client_connection: Connection<ConnectionKey(host='174.1.60.1', port=30000, is_ssl=False, ssl=True, proxy=None, proxy_auth=None, proxy_headers_hash=None, server_hostname=None)>
+INFO:     174.1.60.1:53718 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:53728 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:53742 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:57586 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:53888 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:53898 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:53900 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:53906 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:53920 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:53934 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:47682 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:47698 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:41500 - "GET /server_info HTTP/1.1" 200 OK
+INFO:     174.1.60.1:41510 - "GET /server_info HTTP/1.1" 200 OK
+INFO:     174.1.60.1:51428 - "GET /health HTTP/1.1" 200 OK
+INFO:     174.1.60.1:51434 - "GET /health HTTP/1.1" 200 OK
+INFO:     174.1.60.1:41938 - "GET /v1/models HTTP/1.1" 200 OK
+INFO:     174.1.60.1:41948 - "GET /v1/models HTTP/1.1" 200 OK
+INFO:     174.1.60.1:43850 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:52950 - "GET /server_info HTTP/1.1" 200 OK
+INFO:     174.1.60.1:52964 - "GET /server_info HTTP/1.1" 200 OK
+Unclosed connection
+client_connection: Connection<ConnectionKey(host='174.1.60.1', port=30000, is_ssl=False, ssl=True, proxy=None, proxy_auth=None, proxy_headers_hash=None, server_hostname=None)>
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-20260827-1325/metadata/fatal_pattern_scan.txt b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-20260827-1325/metadata/fatal_pattern_scan.txt
new file mode 100644
index 0000000..e69de29
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-20260827-1325/metadata/router_server_info.json b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-20260827-1325/metadata/router_server_info.json
new file mode 100644
index 0000000..1825fa4
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-20260827-1325/metadata/router_server_info.json
@@ -0,0 +1 @@
+{"internal_states":[{"model_path":"/data/hf_models/Kimi-K3","tokenizer_path":"/data/hf_models/Kimi-K3","tokenizer_mode":"auto","tokenizer_backend":"huggingface","tokenizer_worker_num":1,"detokenizer_worker_num":1,"skip_tokenizer_init":false,"load_format":"auto","model_loader_extra_config":"{}","trust_remote_code":true,"context_length":null,"is_embedding":false,"enable_multimodal":null,"revision":null,"model_impl":"auto","model_config_parser":"auto","json_model_override_args":"{}","dtype":"auto","quantization":null,"quantization_param_path":null,"kv_cache_dtype":"auto","enable_fp32_lm_head":false,"modelopt_quant":null,"modelopt_checkpoint_restore_path":null,"modelopt_checkpoint_save_path":null,"modelopt_export_path":null,"quantize_and_serve":false,"rl_quant_profile":null,"enable_tf32_matmul":false,"mem_fraction_static":0.88,"max_running_requests":null,"max_queued_requests":null,"max_total_tokens":null,"chunked_prefill_size":8192,"enable_dynamic_chunking":false,"max_prefill_tokens":16384,"prefill_max_requests":null,"schedule_policy":"fcfs","enable_priority_scheduling":false,"disable_priority_preemption":false,"default_priority_value":null,"abort_on_priority_when_disabled":false,"schedule_low_priority_values_first":false,"priority_scheduling_preemption_threshold":10,"retraction_policy":"length","schedule_conservativeness":1.0,"page_size":64,"swa_full_tokens_ratio":0.8,"disable_hybrid_swa_memory":false,"radix_eviction_policy":"lru","prefill_only_disable_kv_cache":false,"disable_radix_cache":true,"enable_page_major_kv_layout":false,"enable_unified_memory":false,"disable_chunked_prefix_cache":true,"disable_overlap_schedule":false,"num_continuous_decode_steps":1,"scheduler_recv_interval":1,"enable_mixed_chunk":false,"nccl_port":null,"dist_timeout":3600,"dist_init_addr":"174.1.60.5:20000","nnodes":4,"node_rank":0,"tp_size":32,"dcp_size":1,"pp_size":1,"pp_max_micro_batch_size":181,"pp_async_batch_depth":0,"dp_size":1,"load_balance_method":"round_robin","attn_cp_size":1,"moe_dp_size":1,"dcp_comm_backend":"ag_rs","dcp_replicate_q_proj":null,"dwdp_size":1,"enable_prefill_cp":false,"cp_strategy":null,"enable_dsa_cache_layer_split":false,"enable_dsa_prefill_context_parallel":false,"dsa_prefill_cp_mode":"round-robin-split","enable_prefill_context_parallel":false,"prefill_cp_mode":"in-seq-split","enable_cp_decode_attn_tp":false,"enable_dp_attention":false,"enable_dp_attention_local_control_broadcast":false,"enable_dp_lm_head":false,"enable_attn_tp_input_scattered":false,"disable_attn_tp_gather":false,"enable_p2p_check":false,"device":"cuda","base_gpu_id":0,"gpu_id_step":1,"random_seed":369235288,"watchdog_timeout":300,"soft_watchdog_timeout":null,"sleep_on_idle":false,"use_ray":false,"numa_node":null,"gc_threshold":null,"host":"0.0.0.0","port":30000,"fastapi_root_path":"","smg_grpc_mode":false,"grpc_mode":false,"grpc_port":null,"sidecar":null,"sidecar_args":null,"skip_server_warmup":false,"warmups":null,"enable_http2":false,"ssl_keyfile":null,"ssl_certfile":null,"ssl_ca_certs":null,"ssl_keyfile_password":null,"enable_ssl_refresh":false,"api_key":null,"admin_api_key":null,"served_model_name":"kimi-k3","weight_version":"default","chat_template":null,"hf_chat_template_name":null,"completion_template":null,"file_storage_path":"sglang_storage","enable_cache_report":false,"reasoning_parser":null,"default_chat_template_kwargs":null,"strip_thinking_cache":false,"enable_strict_thinking":false,"tool_call_parser":null,"tool_server":null,"sampling_defaults":"model","asr_max_buffer_seconds":60,"asr_max_concurrent_sessions":32,"preferred_sampling_params":null,"allow_auto_truncate":false,"stream_interval":1,"batch_notify_size":16,"stream_response_default_include_usage":false,"incremental_streaming_output":false,"enable_streaming_session":false,"enable_session_radix_cache":false,"log_level":"info","log_level_http":null,"log_requests":false,"log_requests_level":2,"log_requests_format":"text","log_requests_target":null,"uvicorn_access_log_exclude_prefixes":[],"crash_dump_folder":null,"show_time_cost":false,"enable_metrics":false,"smg_http_sidecar_port":null,"enable_mfu_metrics":false,"enable_metrics_for_all_schedulers":false,"load_snapshot_publish_interval":15,"tokenizer_metrics_custom_labels_header":"x-custom-labels","tokenizer_metrics_allowed_custom_labels":null,"extra_metric_labels":null,"bucket_time_to_first_token":null,"bucket_inter_token_latency":null,"bucket_e2e_request_latency":null,"prompt_tokens_buckets":null,"generation_tokens_buckets":null,"gc_warning_threshold_secs":0.0,"decode_log_interval":40,"enable_request_time_stats_logging":false,"kv_events_config":null,"enable_forward_pass_metrics":false,"forward_pass_metrics_worker_id":"","forward_pass_metrics_ipc_name":null,"enable_trace":false,"trace_modules":"request","otlp_traces_endpoint":"localhost:4317","export_metrics_to_file":false,"export_metrics_to_file_dir":null,"stat_loggers":null,"constrained_json_whitespace_pattern":null,"constrained_json_disable_any_whitespace":false,"attention_backend":"triton","decode_attention_backend":null,"prefill_attention_backend":null,"sampling_backend":"flashinfer","grammar_backend":"xgrammar","radix_cache_backend":null,"mm_attention_backend":null,"fp8_gemm_runner_backend":"auto","fp4_gemm_runner_backend":"auto","bf16_gemm_backend":"auto","dsa_prefill_backend":null,"dsa_decode_backend":null,"dsa_paged_mqa_logits_backend":"auto","dsa_topk_backend":"sgl-kernel","disable_flashinfer_autotune":false,"flashinfer_autotune_skip_ops":null,"mamba_backend":"triton","cuda_graph_config":{"decode":{"backend":"full","max_bs":16,"bs":[1,2,4,8,12,16],"tc_compiler":"eager","full_prefill_max_req":null},"prefill":{"backend":"disabled","max_bs":2048,"bs":[4,8,12,16,20,24,28,32,48,64,80,96,112,128,144,160,176,192,208,224,240,256,288,320,352,384,416,448,480,512,576,640,704,768,832,896,960,1024,1280,1536,1792,2048],"tc_compiler":"eager","full_prefill_max_req":null}},"cuda_graph_backend_decode":null,"cuda_graph_backend_prefill":null,"cuda_graph_max_bs_decode":16,"cuda_graph_max_bs_prefill":null,"cuda_graph_bs_decode":null,"cuda_graph_bs_prefill":null,"cuda_graph_tc_compiler":null,"disable_prefill_cuda_graph":false,"disable_decode_cuda_graph":false,"disable_cuda_graph":false,"disable_cuda_graph_padding":false,"enable_profile_cuda_graph":false,"enable_cudagraph_gc":false,"debug_cuda_graph":false,"enable_layerwise_nvtx_marker":false,"enable_nccl_nvls":false,"enable_symm_mem":false,"triton_attention_reduce_in_fp32":false,"triton_attention_num_kv_splits":8,"triton_attention_split_tile_size":null,"flashinfer_mla_disable_ragged":false,"enable_fused_qk_norm_rope":false,"enable_precise_embedding_interpolation":false,"enable_fused_moe_sum_all_reduce":false,"enable_deepseek_v4_fp4_indexer":false,"disable_custom_all_reduce":false,"enable_mscclpp":false,"enable_torch_symm_mem":false,"enable_scattered_sconv":false,"pre_warm_nccl":false,"enable_quant_communications":false,"enable_flashinfer_allreduce_fusion":false,"enforce_disable_flashinfer_allreduce_fusion":false,"flashinfer_allreduce_fusion_backend":null,"enable_aiter_allreduce_fusion":false,"enable_torch_compile":false,"enable_torch_compile_debug_mode":false,"torch_compile_max_bs":32,"torchao_config":"","speculative_algorithm":null,"speculative_draft_model_path":null,"speculative_draft_model_revision":null,"speculative_draft_load_format":null,"speculative_num_steps":null,"speculative_eagle_topk":null,"speculative_num_draft_tokens":null,"speculative_dflash_block_size":null,"speculative_dspark_block_size":null,"speculative_dspark_sps_table_path":null,"speculative_dspark_confidence_sts_path":null,"speculative_dspark_align_verify_tokens_to_graph_tier":false,"speculative_accept_threshold_single":1.0,"speculative_accept_threshold_acc":1.0,"speculative_use_rejection_sampling":false,"speculative_token_map":null,"speculative_attention_mode":"prefill","speculative_draft_attention_backend":null,"speculative_draft_window_size":null,"speculative_moe_runner_backend":"marlin","speculative_moe_a2a_backend":null,"speculative_draft_model_quantization":null,"speculative_skip_dp_mlp_sync":false,"enable_multi_layer_eagle":false,"speculative_adaptive":false,"speculative_adaptive_config":null,"decoupled_spec_bind_endpoint":null,"decoupled_spec_connect_endpoints":null,"decoupled_spec_rank":null,"decoupled_spec_role":"null","spec_trace_dir":null,"speculative_ngram_min_bfs_breadth":1,"speculative_ngram_max_bfs_breadth":10,"speculative_ngram_match_type":"BFS","speculative_ngram_max_trie_depth":18,"speculative_ngram_capacity":10000000,"speculative_ngram_external_corpus_path":null,"speculative_ngram_external_sam_budget":0,"speculative_ngram_external_corpus_max_tokens":10000000,"ep_size":32,"moe_a2a_backend":"none","moe_runner_backend":"marlin","flashinfer_mxfp4_moe_precision":"default","deepep_mode":"auto","fuseep_mode":2,"deepep_dispatcher_output_dtype":"auto","ep_num_redundant_experts":0,"ep_dispatch_algorithm":null,"init_expert_location":"trivial","enable_eplb":false,"eplb_algorithm":"auto","eplb_rebalance_num_iterations":1000,"eplb_rebalance_layers_per_chunk":null,"eplb_min_rebalancing_utilization_threshold":1.0,"expert_distribution_recorder_mode":null,"expert_distribution_recorder_buffer_size":1000,"enable_expert_distribution_metrics":false,"deepep_config":null,"moe_dense_tp_size":null,"elastic_ep_backend":null,"enable_elastic_expert_backup":false,"mooncake_ib_device":null,"enable_waterfill":false,"ep_join_mode":null,"ep_join_rank_offset":0,"elastic_ep_initial_size":null,"max_ep_size":null,"elastic_ep_scale_timeout":600,"elastic_ep_rejoin":false,"disable_flashinfer_cutlass_moe_fp4_allgather":false,"disable_shared_experts_fusion":false,"enforce_shared_experts_fusion":false,"max_mamba_cache_size":181,"mamba_ssm_dtype":null,"mamba_max_states_per_path":-1,"enable_mamba_cache_stochastic_rounding":false,"mamba_cache_philox_rounds":0,"mamba_full_memory_ratio":0.21,"mamba_radix_cache_strategy":"extra_buffer_lazy","uses_mamba_radix_cache":false,"mamba_track_interval":256,"enable_int8_mamba_checkpoint":false,"int8_mamba_ckpt_size":null,"linear_attn_backend":"triton","linear_attn_decode_backend":null,"linear_attn_prefill_backend":null,"linear_attn_verify_backend":null,"enable_linear_replayssm":false,"linear_replayssm_cache_len":16,"enable_linear_replayssm_spec":false,"enable_hierarchical_cache":false,"hicache_ratio":2.0,"hicache_size":0,"hicache_write_policy":"write_through","hicache_io_backend":"kernel","hicache_mem_layout":"page_first","hicache_storage_backend":null,"hicache_storage_prefetch_policy":"timeout","hicache_storage_backend_extra_config":null,"enable_hisparse":false,"hisparse_config":null,"enable_broadcast_mm_inputs_process":false,"enable_prefix_mm_cache":false,"mm_enable_dp_encoder":false,"mm_process_config":{},"mm_processor_worker_num":0,"mm_io_worker_num":0,"limit_mm_data_per_request":null,"enable_mm_global_cache":false,"disable_fast_image_processor":false,"mm_feature_transport":"cpu","keep_mm_feature_on_device":false,"enable_lora":null,"enable_lora_overlap_loading":null,"max_lora_rank":null,"lora_target_modules":null,"lora_paths":null,"max_loaded_loras":null,"max_loras_per_batch":8,"lora_eviction_policy":"lru","lora_backend":"csgmv","max_lora_chunk_size":16,"experts_shared_outer_loras":null,"lora_use_virtual_experts":false,"lora_strict_loading":false,"lora_drain_wait_threshold":0.0,"enable_two_batch_overlap":false,"enable_single_batch_overlap":false,"tbo_token_distribution_threshold":0.48,"cpu_offload_gb":0,"offload_group_size":-1,"offload_num_in_group":1,"offload_prefetch_step":1,"offload_mode":"cpu","enable_lmcache":false,"lmcache_config_file":null,"enable_flexkv":false,"flexkv_config_file":null,"kt_weight_path":null,"kt_method":"AMXINT4","kt_cpuinfer":null,"kt_threadpool_count":2,"kt_num_gpu_experts":null,"kt_max_deferred_experts_per_token":null,"dllm_algorithm":null,"dllm_algorithm_config":null,"dllm_fdfo":true,"disaggregation_mode":"decode","disaggregation_transfer_backend":"mooncake","disaggregation_bootstrap_port":28800,"disaggregation_ib_device":"mlx5_0,mlx5_1,mlx5_2,mlx5_3","disaggregation_decode_enable_radix_cache":false,"disaggregation_decode_enable_offload_kvcache":false,"num_reserved_decode_tokens":512,"disaggregation_decode_extra_slots":0,"disaggregation_decode_polling_interval":1,"optimistic_prefill_attempts":0,"encoder_only":false,"language_only":false,"encoder_transfer_backend":"zmq_to_tokenizer","encoder_urls":[],"encoder_bootstrap_port":8997,"encoder_register_urls":[],"enable_adaptive_dispatch_to_encoder":false,"enable_pdmux":false,"pdmux_config_path":null,"sm_group_num":8,"custom_weight_loader":[],"weight_loader_disable_mmap":false,"weight_loader_prefetch_checkpoints":false,"weight_loader_prefetch_num_threads":4,"weight_loader_drop_cache_after_load":false,"remote_instance_weight_loader_seed_instance_ip":null,"remote_instance_weight_loader_seed_instance_service_port":null,"remote_instance_weight_loader_send_weights_group_ports":null,"remote_instance_weight_loader_backend":"nccl","remote_instance_weight_loader_start_seed_via_transfer_engine":false,"engine_info_bootstrap_port":6789,"modelexpress_config":null,"download_dir":null,"model_checksum":null,"delete_ckpt_after_loading":false,"decrypted_config_file":null,"decrypted_draft_config_file":null,"checkpoint_engine_wait_weights_before_ready":false,"enable_prefill_delayer":false,"prefill_delayer_max_delay_passes":30,"prefill_delayer_token_usage_low_watermark":null,"prefill_delayer_forward_passes_buckets":null,"prefill_delayer_wait_seconds_buckets":null,"prefill_delayer_queue_min_ratio":null,"prefill_delayer_max_delay_ms":null,"min_free_slots_delay":null,"enable_deterministic_inference":false,"rl_on_policy_target":null,"kv_canary":"none","kv_canary_real_data":"none","kv_canary_sweep_interval":0,"enable_dynamic_batch_tokenizer":false,"dynamic_batch_tokenizer_batch_size":32,"dynamic_batch_tokenizer_batch_timeout":0.002,"enable_tokenizer_batch_encode":false,"disable_tokenizer_batch_decode":false,"debug_tensor_dump_output_folder":null,"debug_tensor_dump_layers":null,"debug_tensor_dump_input_file":null,"enable_memory_saver":false,"enable_weights_cpu_backup":false,"enable_draft_weights_cpu_backup":false,"enable_custom_logit_processor":false,"enable_return_hidden_states":false,"enable_return_routed_experts":false,"enable_return_indexer_topk":false,"disable_outlines_disk_cache":false,"enable_mis":false,"weight_cache_mode":"off","weight_cache_socket":null,"weight_cache_timeout":1800,"forward_hooks":null,"msprobe_dump_config":null,"_resolved_overrides":[["_sampling_backend_default",{"sampling_backend":"flashinfer"}],["_attention_backend_default",{"attention_backend":"triton"}],["_data_parallelism_defaults",{"enable_dp_attention":false,"enable_dp_lm_head":false}],["_speculative_moe_runner_default",{"speculative_moe_runner_backend":"marlin"}]],"grpc_worker_threads":4,"_quantization_explicitly_unset":false,"_cuda_graph_config_locked":[["decode","max_bs"]],"_declarations_materialized":true,"_runtime_mutations":[["model_runner.chunked_prefix_cache_gate",{"disable_chunked_prefix_cache":true}],["mamba_pool.memory_budget",{"max_mamba_cache_size":181}]],"_in_override":false,"_mx_config_cache":{},"max_speculative_num_draft_tokens":null,"last_gen_throughput":26.412623015816244,"memory_usage":{"weight":58.57,"kvcache":11.34,"token_capacity":440256,"graph":0.6},"effective_max_running_requests_per_dp":181}],"prefill":[{"model_path":"/data/hf_models/Kimi-K3","tokenizer_path":"/data/hf_models/Kimi-K3","tokenizer_mode":"auto","tokenizer_backend":"huggingface","tokenizer_worker_num":1,"detokenizer_worker_num":1,"skip_tokenizer_init":false,"load_format":"auto","model_loader_extra_config":"{}","trust_remote_code":true,"context_length":null,"is_embedding":false,"enable_multimodal":null,"revision":null,"model_impl":"auto","model_config_parser":"auto","json_model_override_args":"{}","dtype":"auto","quantization":null,"quantization_param_path":null,"kv_cache_dtype":"auto","enable_fp32_lm_head":false,"modelopt_quant":null,"modelopt_checkpoint_restore_path":null,"modelopt_checkpoint_save_path":null,"modelopt_export_path":null,"quantize_and_serve":false,"rl_quant_profile":null,"enable_tf32_matmul":false,"mem_fraction_static":0.88,"max_running_requests":null,"max_queued_requests":null,"max_total_tokens":null,"chunked_prefill_size":8192,"enable_dynamic_chunking":false,"max_prefill_tokens":16384,"prefill_max_requests":null,"schedule_policy":"fcfs","enable_priority_scheduling":false,"disable_priority_preemption":false,"default_priority_value":null,"abort_on_priority_when_disabled":false,"schedule_low_priority_values_first":false,"priority_scheduling_preemption_threshold":10,"retraction_policy":"length","schedule_conservativeness":1.0,"page_size":64,"swa_full_tokens_ratio":0.8,"disable_hybrid_swa_memory":false,"radix_eviction_policy":"lru","prefill_only_disable_kv_cache":false,"disable_radix_cache":true,"enable_page_major_kv_layout":false,"enable_unified_memory":false,"disable_chunked_prefix_cache":false,"disable_overlap_schedule":true,"num_continuous_decode_steps":1,"scheduler_recv_interval":1,"enable_mixed_chunk":false,"nccl_port":null,"dist_timeout":3600,"dist_init_addr":"174.1.60.1:20000","nnodes":4,"node_rank":0,"tp_size":4,"dcp_size":1,"pp_size":8,"pp_max_micro_batch_size":null,"pp_async_batch_depth":0,"dp_size":1,"load_balance_method":"follow_bootstrap_room","attn_cp_size":1,"moe_dp_size":1,"dcp_comm_backend":"ag_rs","dcp_replicate_q_proj":null,"dwdp_size":1,"enable_prefill_cp":false,"cp_strategy":null,"enable_dsa_cache_layer_split":false,"enable_dsa_prefill_context_parallel":false,"dsa_prefill_cp_mode":"round-robin-split","enable_prefill_context_parallel":false,"prefill_cp_mode":"in-seq-split","enable_cp_decode_attn_tp":false,"enable_dp_attention":false,"enable_dp_attention_local_control_broadcast":false,"enable_dp_lm_head":false,"enable_attn_tp_input_scattered":false,"disable_attn_tp_gather":false,"enable_p2p_check":false,"device":"cuda","base_gpu_id":0,"gpu_id_step":1,"random_seed":984607766,"watchdog_timeout":300,"soft_watchdog_timeout":null,"sleep_on_idle":false,"use_ray":false,"custom_sigquit_handler":null,"numa_node":null,"gc_threshold":null,"host":"0.0.0.0","port":30000,"fastapi_root_path":"","smg_grpc_mode":false,"grpc_mode":false,"grpc_port":null,"sidecar":null,"sidecar_args":null,"skip_server_warmup":false,"warmups":null,"enable_http2":false,"ssl_keyfile":null,"ssl_certfile":null,"ssl_ca_certs":null,"ssl_keyfile_password":null,"enable_ssl_refresh":false,"api_key":null,"admin_api_key":null,"served_model_name":"kimi-k3","weight_version":"default","chat_template":null,"hf_chat_template_name":null,"completion_template":null,"file_storage_path":"sglang_storage","enable_cache_report":false,"reasoning_parser":null,"default_chat_template_kwargs":null,"strip_thinking_cache":false,"enable_strict_thinking":false,"tool_call_parser":null,"tool_server":null,"sampling_defaults":"model","asr_max_buffer_seconds":60,"asr_max_concurrent_sessions":32,"preferred_sampling_params":null,"allow_auto_truncate":false,"stream_interval":1,"batch_notify_size":16,"stream_response_default_include_usage":false,"incremental_streaming_output":false,"enable_streaming_session":false,"enable_session_radix_cache":false,"log_level":"info","log_level_http":null,"log_requests":false,"log_requests_level":2,"log_requests_format":"text","log_requests_target":null,"uvicorn_access_log_exclude_prefixes":[],"crash_dump_folder":null,"show_time_cost":false,"enable_metrics":false,"smg_http_sidecar_port":null,"enable_mfu_metrics":false,"enable_metrics_for_all_schedulers":false,"load_snapshot_publish_interval":15,"tokenizer_metrics_custom_labels_header":"x-custom-labels","tokenizer_metrics_allowed_custom_labels":null,"extra_metric_labels":null,"bucket_time_to_first_token":null,"bucket_inter_token_latency":null,"bucket_e2e_request_latency":null,"prompt_tokens_buckets":null,"generation_tokens_buckets":null,"gc_warning_threshold_secs":0.0,"decode_log_interval":40,"enable_request_time_stats_logging":false,"kv_events_config":null,"enable_forward_pass_metrics":false,"forward_pass_metrics_worker_id":"","forward_pass_metrics_ipc_name":null,"enable_trace":false,"trace_modules":"request","otlp_traces_endpoint":"localhost:4317","export_metrics_to_file":false,"export_metrics_to_file_dir":null,"stat_loggers":null,"constrained_json_whitespace_pattern":null,"constrained_json_disable_any_whitespace":false,"attention_backend":"triton","decode_attention_backend":null,"prefill_attention_backend":null,"sampling_backend":"flashinfer","grammar_backend":"xgrammar","radix_cache_backend":null,"mm_attention_backend":null,"fp8_gemm_runner_backend":"auto","fp4_gemm_runner_backend":"auto","bf16_gemm_backend":"auto","dsa_prefill_backend":null,"dsa_decode_backend":null,"dsa_paged_mqa_logits_backend":"auto","dsa_topk_backend":"sgl-kernel","disable_flashinfer_autotune":false,"flashinfer_autotune_skip_ops":null,"mamba_backend":"triton","cuda_graph_config":{"decode":{"backend":"disabled","max_bs":16,"bs":[1,2,4,8,12,16],"tc_compiler":"eager","full_prefill_max_req":null},"prefill":{"backend":"disabled","max_bs":2048,"bs":[4,8,12,16,20,24,28,32,48,64,80,96,112,128,144,160,176,192,208,224,240,256,288,320,352,384,416,448,480,512,576,640,704,768,832,896,960,1024,1280,1536,1792,2048],"tc_compiler":"eager","full_prefill_max_req":null}},"cuda_graph_backend_decode":null,"cuda_graph_backend_prefill":null,"cuda_graph_max_bs_decode":16,"cuda_graph_max_bs_prefill":null,"cuda_graph_bs_decode":null,"cuda_graph_bs_prefill":null,"cuda_graph_tc_compiler":null,"disable_prefill_cuda_graph":false,"disable_decode_cuda_graph":false,"disable_cuda_graph":false,"disable_cuda_graph_padding":false,"enable_profile_cuda_graph":false,"enable_cudagraph_gc":false,"debug_cuda_graph":false,"enable_layerwise_nvtx_marker":false,"enable_nccl_nvls":false,"enable_symm_mem":false,"triton_attention_reduce_in_fp32":false,"triton_attention_num_kv_splits":8,"triton_attention_split_tile_size":null,"flashinfer_mla_disable_ragged":false,"enable_fused_qk_norm_rope":false,"enable_precise_embedding_interpolation":false,"enable_fused_moe_sum_all_reduce":false,"enable_deepseek_v4_fp4_indexer":false,"disable_custom_all_reduce":false,"enable_mscclpp":false,"enable_torch_symm_mem":false,"enable_scattered_sconv":false,"pre_warm_nccl":false,"enable_quant_communications":false,"enable_flashinfer_allreduce_fusion":false,"enforce_disable_flashinfer_allreduce_fusion":false,"flashinfer_allreduce_fusion_backend":null,"enable_aiter_allreduce_fusion":false,"enable_torch_compile":false,"enable_torch_compile_debug_mode":false,"torch_compile_max_bs":32,"torchao_config":"","speculative_algorithm":null,"speculative_draft_model_path":null,"speculative_draft_model_revision":null,"speculative_draft_load_format":null,"speculative_num_steps":null,"speculative_eagle_topk":null,"speculative_num_draft_tokens":null,"speculative_dflash_block_size":null,"speculative_dspark_block_size":null,"speculative_dspark_sps_table_path":null,"speculative_dspark_confidence_sts_path":null,"speculative_dspark_align_verify_tokens_to_graph_tier":false,"speculative_accept_threshold_single":1.0,"speculative_accept_threshold_acc":1.0,"speculative_use_rejection_sampling":false,"speculative_token_map":null,"speculative_attention_mode":"prefill","speculative_draft_attention_backend":null,"speculative_draft_window_size":null,"speculative_moe_runner_backend":"flashinfer_mxfp4","speculative_moe_a2a_backend":null,"speculative_draft_model_quantization":null,"speculative_skip_dp_mlp_sync":false,"enable_multi_layer_eagle":false,"speculative_adaptive":false,"speculative_adaptive_config":null,"decoupled_spec_bind_endpoint":null,"decoupled_spec_connect_endpoints":null,"decoupled_spec_rank":null,"decoupled_spec_role":"null","spec_trace_dir":null,"speculative_ngram_min_bfs_breadth":1,"speculative_ngram_max_bfs_breadth":10,"speculative_ngram_match_type":"BFS","speculative_ngram_max_trie_depth":18,"speculative_ngram_capacity":10000000,"speculative_ngram_external_corpus_path":null,"speculative_ngram_external_sam_budget":0,"speculative_ngram_external_corpus_max_tokens":10000000,"ep_size":4,"moe_a2a_backend":"none","moe_runner_backend":"flashinfer_mxfp4","flashinfer_mxfp4_moe_precision":"default","deepep_mode":"auto","fuseep_mode":2,"deepep_dispatcher_output_dtype":"auto","ep_num_redundant_experts":0,"ep_dispatch_algorithm":null,"init_expert_location":"trivial","enable_eplb":false,"eplb_algorithm":"auto","eplb_rebalance_num_iterations":1000,"eplb_rebalance_layers_per_chunk":null,"eplb_min_rebalancing_utilization_threshold":1.0,"expert_distribution_recorder_mode":null,"expert_distribution_recorder_buffer_size":1000,"enable_expert_distribution_metrics":false,"deepep_config":null,"moe_dense_tp_size":null,"elastic_ep_backend":null,"enable_elastic_expert_backup":false,"mooncake_ib_device":null,"enable_waterfill":false,"ep_join_mode":null,"ep_join_rank_offset":0,"elastic_ep_initial_size":null,"max_ep_size":null,"elastic_ep_scale_timeout":600,"elastic_ep_rejoin":false,"disable_flashinfer_cutlass_moe_fp4_allgather":false,"disable_shared_experts_fusion":false,"enforce_shared_experts_fusion":false,"max_mamba_cache_size":null,"mamba_ssm_dtype":null,"mamba_max_states_per_path":-1,"enable_mamba_cache_stochastic_rounding":false,"mamba_cache_philox_rounds":0,"mamba_full_memory_ratio":0.36,"mamba_radix_cache_strategy":"extra_buffer_lazy","uses_mamba_radix_cache":false,"mamba_track_interval":256,"enable_int8_mamba_checkpoint":false,"int8_mamba_ckpt_size":null,"linear_attn_backend":"triton","linear_attn_decode_backend":null,"linear_attn_prefill_backend":null,"linear_attn_verify_backend":null,"enable_linear_replayssm":false,"linear_replayssm_cache_len":16,"enable_linear_replayssm_spec":false,"enable_hierarchical_cache":false,"hicache_ratio":2.0,"hicache_size":0,"hicache_write_policy":"write_through","hicache_io_backend":"kernel","hicache_mem_layout":"page_first","hicache_storage_backend":null,"hicache_storage_prefetch_policy":"timeout","hicache_storage_backend_extra_config":null,"enable_hisparse":false,"hisparse_config":null,"enable_broadcast_mm_inputs_process":false,"enable_prefix_mm_cache":false,"mm_enable_dp_encoder":false,"mm_process_config":{},"mm_processor_worker_num":0,"mm_io_worker_num":0,"limit_mm_data_per_request":null,"enable_mm_global_cache":false,"disable_fast_image_processor":false,"mm_feature_transport":"cpu","keep_mm_feature_on_device":false,"enable_lora":null,"enable_lora_overlap_loading":null,"max_lora_rank":null,"lora_target_modules":null,"lora_paths":null,"max_loaded_loras":null,"max_loras_per_batch":8,"lora_eviction_policy":"lru","lora_backend":"csgmv","max_lora_chunk_size":16,"experts_shared_outer_loras":null,"lora_use_virtual_experts":false,"lora_strict_loading":false,"lora_drain_wait_threshold":0.0,"enable_two_batch_overlap":false,"enable_single_batch_overlap":false,"tbo_token_distribution_threshold":0.48,"cpu_offload_gb":0,"offload_group_size":-1,"offload_num_in_group":1,"offload_prefetch_step":1,"offload_mode":"cpu","enable_lmcache":false,"lmcache_config_file":null,"enable_flexkv":false,"flexkv_config_file":null,"kt_weight_path":null,"kt_method":"AMXINT4","kt_cpuinfer":null,"kt_threadpool_count":2,"kt_num_gpu_experts":null,"kt_max_deferred_experts_per_token":null,"dllm_algorithm":null,"dllm_algorithm_config":null,"dllm_fdfo":true,"disaggregation_mode":"prefill","disaggregation_transfer_backend":"mooncake","disaggregation_bootstrap_port":28800,"disaggregation_ib_device":"mlx5_0,mlx5_1,mlx5_2,mlx5_3","disaggregation_decode_enable_radix_cache":false,"disaggregation_decode_enable_offload_kvcache":false,"num_reserved_decode_tokens":512,"disaggregation_decode_extra_slots":null,"disaggregation_decode_polling_interval":1,"optimistic_prefill_attempts":0,"encoder_only":false,"language_only":false,"encoder_transfer_backend":"zmq_to_tokenizer","encoder_urls":[],"encoder_bootstrap_port":8997,"encoder_register_urls":[],"enable_adaptive_dispatch_to_encoder":false,"enable_pdmux":false,"pdmux_config_path":null,"sm_group_num":8,"custom_weight_loader":[],"weight_loader_disable_mmap":false,"weight_loader_prefetch_checkpoints":false,"weight_loader_prefetch_num_threads":4,"weight_loader_drop_cache_after_load":false,"remote_instance_weight_loader_seed_instance_ip":null,"remote_instance_weight_loader_seed_instance_service_port":null,"remote_instance_weight_loader_send_weights_group_ports":null,"remote_instance_weight_loader_backend":"nccl","remote_instance_weight_loader_start_seed_via_transfer_engine":false,"engine_info_bootstrap_port":6789,"modelexpress_config":null,"download_dir":null,"model_checksum":null,"delete_ckpt_after_loading":false,"decrypted_config_file":null,"decrypted_draft_config_file":null,"checkpoint_engine_wait_weights_before_ready":false,"enable_prefill_delayer":false,"prefill_delayer_max_delay_passes":30,"prefill_delayer_token_usage_low_watermark":null,"prefill_delayer_forward_passes_buckets":null,"prefill_delayer_wait_seconds_buckets":null,"prefill_delayer_queue_min_ratio":null,"prefill_delayer_max_delay_ms":null,"min_free_slots_delay":null,"enable_deterministic_inference":false,"rl_on_policy_target":null,"kv_canary":"none","kv_canary_real_data":"none","kv_canary_sweep_interval":0,"enable_dynamic_batch_tokenizer":false,"dynamic_batch_tokenizer_batch_size":32,"dynamic_batch_tokenizer_batch_timeout":0.002,"enable_tokenizer_batch_encode":false,"disable_tokenizer_batch_decode":false,"debug_tensor_dump_output_folder":null,"debug_tensor_dump_layers":null,"debug_tensor_dump_input_file":null,"enable_memory_saver":false,"enable_weights_cpu_backup":false,"enable_draft_weights_cpu_backup":false,"enable_custom_logit_processor":false,"enable_return_hidden_states":false,"enable_return_routed_experts":false,"enable_return_indexer_topk":false,"disable_outlines_disk_cache":false,"enable_mis":false,"weight_cache_mode":"off","weight_cache_socket":null,"weight_cache_timeout":1800,"forward_hooks":null,"msprobe_dump_config":null,"status":"ready","max_total_num_tokens":3801408,"max_req_input_len":1048570,"internal_states":[{"model_path":"/data/hf_models/Kimi-K3","tokenizer_path":"/data/hf_models/Kimi-K3","tokenizer_mode":"auto","tokenizer_backend":"huggingface","tokenizer_worker_num":1,"detokenizer_worker_num":1,"skip_tokenizer_init":false,"load_format":"auto","model_loader_extra_config":"{}","trust_remote_code":true,"context_length":null,"is_embedding":false,"enable_multimodal":null,"revision":null,"model_impl":"auto","model_config_parser":"auto","json_model_override_args":"{}","dtype":"auto","quantization":null,"quantization_param_path":null,"kv_cache_dtype":"auto","enable_fp32_lm_head":false,"modelopt_quant":null,"modelopt_checkpoint_restore_path":null,"modelopt_checkpoint_save_path":null,"modelopt_export_path":null,"quantize_and_serve":false,"rl_quant_profile":null,"enable_tf32_matmul":false,"mem_fraction_static":0.88,"max_running_requests":null,"max_queued_requests":null,"max_total_tokens":null,"chunked_prefill_size":8192,"enable_dynamic_chunking":false,"max_prefill_tokens":16384,"prefill_max_requests":null,"schedule_policy":"fcfs","enable_priority_scheduling":false,"disable_priority_preemption":false,"default_priority_value":null,"abort_on_priority_when_disabled":false,"schedule_low_priority_values_first":false,"priority_scheduling_preemption_threshold":10,"retraction_policy":"length","schedule_conservativeness":1.0,"page_size":64,"swa_full_tokens_ratio":0.8,"disable_hybrid_swa_memory":false,"radix_eviction_policy":"lru","prefill_only_disable_kv_cache":false,"disable_radix_cache":true,"enable_page_major_kv_layout":false,"enable_unified_memory":false,"disable_chunked_prefix_cache":true,"disable_overlap_schedule":true,"num_continuous_decode_steps":1,"scheduler_recv_interval":1,"enable_mixed_chunk":false,"nccl_port":null,"dist_timeout":3600,"dist_init_addr":"174.1.60.1:20000","nnodes":4,"node_rank":0,"tp_size":4,"dcp_size":1,"pp_size":8,"pp_max_micro_batch_size":6,"pp_async_batch_depth":0,"dp_size":1,"load_balance_method":"follow_bootstrap_room","attn_cp_size":1,"moe_dp_size":1,"dcp_comm_backend":"ag_rs","dcp_replicate_q_proj":null,"dwdp_size":1,"enable_prefill_cp":false,"cp_strategy":null,"enable_dsa_cache_layer_split":false,"enable_dsa_prefill_context_parallel":false,"dsa_prefill_cp_mode":"round-robin-split","enable_prefill_context_parallel":false,"prefill_cp_mode":"in-seq-split","enable_cp_decode_attn_tp":false,"enable_dp_attention":false,"enable_dp_attention_local_control_broadcast":false,"enable_dp_lm_head":false,"enable_attn_tp_input_scattered":false,"disable_attn_tp_gather":false,"enable_p2p_check":false,"device":"cuda","base_gpu_id":0,"gpu_id_step":1,"random_seed":984607766,"watchdog_timeout":300,"soft_watchdog_timeout":null,"sleep_on_idle":false,"use_ray":false,"numa_node":null,"gc_threshold":null,"host":"0.0.0.0","port":30000,"fastapi_root_path":"","smg_grpc_mode":false,"grpc_mode":false,"grpc_port":null,"sidecar":null,"sidecar_args":null,"skip_server_warmup":false,"warmups":null,"enable_http2":false,"ssl_keyfile":null,"ssl_certfile":null,"ssl_ca_certs":null,"ssl_keyfile_password":null,"enable_ssl_refresh":false,"api_key":null,"admin_api_key":null,"served_model_name":"kimi-k3","weight_version":"default","chat_template":null,"hf_chat_template_name":null,"completion_template":null,"file_storage_path":"sglang_storage","enable_cache_report":false,"reasoning_parser":null,"default_chat_template_kwargs":null,"strip_thinking_cache":false,"enable_strict_thinking":false,"tool_call_parser":null,"tool_server":null,"sampling_defaults":"model","asr_max_buffer_seconds":60,"asr_max_concurrent_sessions":32,"preferred_sampling_params":null,"allow_auto_truncate":false,"stream_interval":1,"batch_notify_size":16,"stream_response_default_include_usage":false,"incremental_streaming_output":false,"enable_streaming_session":false,"enable_session_radix_cache":false,"log_level":"info","log_level_http":null,"log_requests":false,"log_requests_level":2,"log_requests_format":"text","log_requests_target":null,"uvicorn_access_log_exclude_prefixes":[],"crash_dump_folder":null,"show_time_cost":false,"enable_metrics":false,"smg_http_sidecar_port":null,"enable_mfu_metrics":false,"enable_metrics_for_all_schedulers":false,"load_snapshot_publish_interval":15,"tokenizer_metrics_custom_labels_header":"x-custom-labels","tokenizer_metrics_allowed_custom_labels":null,"extra_metric_labels":null,"bucket_time_to_first_token":null,"bucket_inter_token_latency":null,"bucket_e2e_request_latency":null,"prompt_tokens_buckets":null,"generation_tokens_buckets":null,"gc_warning_threshold_secs":0.0,"decode_log_interval":40,"enable_request_time_stats_logging":false,"kv_events_config":null,"enable_forward_pass_metrics":false,"forward_pass_metrics_worker_id":"","forward_pass_metrics_ipc_name":null,"enable_trace":false,"trace_modules":"request","otlp_traces_endpoint":"localhost:4317","export_metrics_to_file":false,"export_metrics_to_file_dir":null,"stat_loggers":null,"constrained_json_whitespace_pattern":null,"constrained_json_disable_any_whitespace":false,"attention_backend":"triton","decode_attention_backend":null,"prefill_attention_backend":null,"sampling_backend":"flashinfer","grammar_backend":"xgrammar","radix_cache_backend":null,"mm_attention_backend":null,"fp8_gemm_runner_backend":"auto","fp4_gemm_runner_backend":"auto","bf16_gemm_backend":"auto","dsa_prefill_backend":null,"dsa_decode_backend":null,"dsa_paged_mqa_logits_backend":"auto","dsa_topk_backend":"sgl-kernel","disable_flashinfer_autotune":false,"flashinfer_autotune_skip_ops":null,"mamba_backend":"triton","cuda_graph_config":{"decode":{"backend":"disabled","max_bs":16,"bs":[1,2,4,8,12,16],"tc_compiler":"eager","full_prefill_max_req":null},"prefill":{"backend":"disabled","max_bs":2048,"bs":[4,8,12,16,20,24,28,32,48,64,80,96,112,128,144,160,176,192,208,224,240,256,288,320,352,384,416,448,480,512,576,640,704,768,832,896,960,1024,1280,1536,1792,2048],"tc_compiler":"eager","full_prefill_max_req":null}},"cuda_graph_backend_decode":null,"cuda_graph_backend_prefill":null,"cuda_graph_max_bs_decode":16,"cuda_graph_max_bs_prefill":null,"cuda_graph_bs_decode":null,"cuda_graph_bs_prefill":null,"cuda_graph_tc_compiler":null,"disable_prefill_cuda_graph":false,"disable_decode_cuda_graph":false,"disable_cuda_graph":false,"disable_cuda_graph_padding":false,"enable_profile_cuda_graph":false,"enable_cudagraph_gc":false,"debug_cuda_graph":false,"enable_layerwise_nvtx_marker":false,"enable_nccl_nvls":false,"enable_symm_mem":false,"triton_attention_reduce_in_fp32":false,"triton_attention_num_kv_splits":8,"triton_attention_split_tile_size":null,"flashinfer_mla_disable_ragged":false,"enable_fused_qk_norm_rope":false,"enable_precise_embedding_interpolation":false,"enable_fused_moe_sum_all_reduce":false,"enable_deepseek_v4_fp4_indexer":false,"disable_custom_all_reduce":false,"enable_mscclpp":false,"enable_torch_symm_mem":false,"enable_scattered_sconv":false,"pre_warm_nccl":false,"enable_quant_communications":false,"enable_flashinfer_allreduce_fusion":false,"enforce_disable_flashinfer_allreduce_fusion":false,"flashinfer_allreduce_fusion_backend":null,"enable_aiter_allreduce_fusion":false,"enable_torch_compile":false,"enable_torch_compile_debug_mode":false,"torch_compile_max_bs":32,"torchao_config":"","speculative_algorithm":null,"speculative_draft_model_path":null,"speculative_draft_model_revision":null,"speculative_draft_load_format":null,"speculative_num_steps":null,"speculative_eagle_topk":null,"speculative_num_draft_tokens":null,"speculative_dflash_block_size":null,"speculative_dspark_block_size":null,"speculative_dspark_sps_table_path":null,"speculative_dspark_confidence_sts_path":null,"speculative_dspark_align_verify_tokens_to_graph_tier":false,"speculative_accept_threshold_single":1.0,"speculative_accept_threshold_acc":1.0,"speculative_use_rejection_sampling":false,"speculative_token_map":null,"speculative_attention_mode":"prefill","speculative_draft_attention_backend":null,"speculative_draft_window_size":null,"speculative_moe_runner_backend":"flashinfer_mxfp4","speculative_moe_a2a_backend":null,"speculative_draft_model_quantization":null,"speculative_skip_dp_mlp_sync":false,"enable_multi_layer_eagle":false,"speculative_adaptive":false,"speculative_adaptive_config":null,"decoupled_spec_bind_endpoint":null,"decoupled_spec_connect_endpoints":null,"decoupled_spec_rank":null,"decoupled_spec_role":"null","spec_trace_dir":null,"speculative_ngram_min_bfs_breadth":1,"speculative_ngram_max_bfs_breadth":10,"speculative_ngram_match_type":"BFS","speculative_ngram_max_trie_depth":18,"speculative_ngram_capacity":10000000,"speculative_ngram_external_corpus_path":null,"speculative_ngram_external_sam_budget":0,"speculative_ngram_external_corpus_max_tokens":10000000,"ep_size":4,"moe_a2a_backend":"none","moe_runner_backend":"flashinfer_mxfp4","flashinfer_mxfp4_moe_precision":"default","deepep_mode":"auto","fuseep_mode":2,"deepep_dispatcher_output_dtype":"auto","ep_num_redundant_experts":0,"ep_dispatch_algorithm":null,"init_expert_location":"trivial","enable_eplb":false,"eplb_algorithm":"auto","eplb_rebalance_num_iterations":1000,"eplb_rebalance_layers_per_chunk":null,"eplb_min_rebalancing_utilization_threshold":1.0,"expert_distribution_recorder_mode":null,"expert_distribution_recorder_buffer_size":1000,"enable_expert_distribution_metrics":false,"deepep_config":null,"moe_dense_tp_size":null,"elastic_ep_backend":null,"enable_elastic_expert_backup":false,"mooncake_ib_device":null,"enable_waterfill":false,"ep_join_mode":null,"ep_join_rank_offset":0,"elastic_ep_initial_size":null,"max_ep_size":null,"elastic_ep_scale_timeout":600,"elastic_ep_rejoin":false,"disable_flashinfer_cutlass_moe_fp4_allgather":false,"disable_shared_experts_fusion":false,"enforce_shared_experts_fusion":false,"max_mamba_cache_size":54,"mamba_ssm_dtype":null,"mamba_max_states_per_path":-1,"enable_mamba_cache_stochastic_rounding":false,"mamba_cache_philox_rounds":0,"mamba_full_memory_ratio":0.36,"mamba_radix_cache_strategy":"extra_buffer_lazy","uses_mamba_radix_cache":false,"mamba_track_interval":256,"enable_int8_mamba_checkpoint":false,"int8_mamba_ckpt_size":null,"linear_attn_backend":"triton","linear_attn_decode_backend":null,"linear_attn_prefill_backend":null,"linear_attn_verify_backend":null,"enable_linear_replayssm":false,"linear_replayssm_cache_len":16,"enable_linear_replayssm_spec":false,"enable_hierarchical_cache":false,"hicache_ratio":2.0,"hicache_size":0,"hicache_write_policy":"write_through","hicache_io_backend":"kernel","hicache_mem_layout":"page_first","hicache_storage_backend":null,"hicache_storage_prefetch_policy":"timeout","hicache_storage_backend_extra_config":null,"enable_hisparse":false,"hisparse_config":null,"enable_broadcast_mm_inputs_process":false,"enable_prefix_mm_cache":false,"mm_enable_dp_encoder":false,"mm_process_config":{},"mm_processor_worker_num":0,"mm_io_worker_num":0,"limit_mm_data_per_request":null,"enable_mm_global_cache":false,"disable_fast_image_processor":false,"mm_feature_transport":"cpu","keep_mm_feature_on_device":false,"enable_lora":null,"enable_lora_overlap_loading":null,"max_lora_rank":null,"lora_target_modules":null,"lora_paths":null,"max_loaded_loras":null,"max_loras_per_batch":8,"lora_eviction_policy":"lru","lora_backend":"csgmv","max_lora_chunk_size":16,"experts_shared_outer_loras":null,"lora_use_virtual_experts":false,"lora_strict_loading":false,"lora_drain_wait_threshold":0.0,"enable_two_batch_overlap":false,"enable_single_batch_overlap":false,"tbo_token_distribution_threshold":0.48,"cpu_offload_gb":0,"offload_group_size":-1,"offload_num_in_group":1,"offload_prefetch_step":1,"offload_mode":"cpu","enable_lmcache":false,"lmcache_config_file":null,"enable_flexkv":false,"flexkv_config_file":null,"kt_weight_path":null,"kt_method":"AMXINT4","kt_cpuinfer":null,"kt_threadpool_count":2,"kt_num_gpu_experts":null,"kt_max_deferred_experts_per_token":null,"dllm_algorithm":null,"dllm_algorithm_config":null,"dllm_fdfo":true,"disaggregation_mode":"prefill","disaggregation_transfer_backend":"mooncake","disaggregation_bootstrap_port":28800,"disaggregation_ib_device":"mlx5_0,mlx5_1,mlx5_2,mlx5_3","disaggregation_decode_enable_radix_cache":false,"disaggregation_decode_enable_offload_kvcache":false,"num_reserved_decode_tokens":512,"disaggregation_decode_extra_slots":null,"disaggregation_decode_polling_interval":1,"optimistic_prefill_attempts":0,"encoder_only":false,"language_only":false,"encoder_transfer_backend":"zmq_to_tokenizer","encoder_urls":[],"encoder_bootstrap_port":8997,"encoder_register_urls":[],"enable_adaptive_dispatch_to_encoder":false,"enable_pdmux":false,"pdmux_config_path":null,"sm_group_num":8,"custom_weight_loader":[],"weight_loader_disable_mmap":false,"weight_loader_prefetch_checkpoints":false,"weight_loader_prefetch_num_threads":4,"weight_loader_drop_cache_after_load":false,"remote_instance_weight_loader_seed_instance_ip":null,"remote_instance_weight_loader_seed_instance_service_port":null,"remote_instance_weight_loader_send_weights_group_ports":null,"remote_instance_weight_loader_backend":"nccl","remote_instance_weight_loader_start_seed_via_transfer_engine":false,"engine_info_bootstrap_port":6789,"modelexpress_config":null,"download_dir":null,"model_checksum":null,"delete_ckpt_after_loading":false,"decrypted_config_file":null,"decrypted_draft_config_file":null,"checkpoint_engine_wait_weights_before_ready":false,"enable_prefill_delayer":false,"prefill_delayer_max_delay_passes":30,"prefill_delayer_token_usage_low_watermark":null,"prefill_delayer_forward_passes_buckets":null,"prefill_delayer_wait_seconds_buckets":null,"prefill_delayer_queue_min_ratio":null,"prefill_delayer_max_delay_ms":null,"min_free_slots_delay":null,"enable_deterministic_inference":false,"rl_on_policy_target":null,"kv_canary":"none","kv_canary_real_data":"none","kv_canary_sweep_interval":0,"enable_dynamic_batch_tokenizer":false,"dynamic_batch_tokenizer_batch_size":32,"dynamic_batch_tokenizer_batch_timeout":0.002,"enable_tokenizer_batch_encode":false,"disable_tokenizer_batch_decode":false,"debug_tensor_dump_output_folder":null,"debug_tensor_dump_layers":null,"debug_tensor_dump_input_file":null,"enable_memory_saver":false,"enable_weights_cpu_backup":false,"enable_draft_weights_cpu_backup":false,"enable_custom_logit_processor":false,"enable_return_hidden_states":false,"enable_return_routed_experts":false,"enable_return_indexer_topk":false,"disable_outlines_disk_cache":false,"enable_mis":false,"weight_cache_mode":"off","weight_cache_socket":null,"weight_cache_timeout":1800,"forward_hooks":null,"msprobe_dump_config":null,"_resolved_overrides":[["_sampling_backend_default",{"sampling_backend":"flashinfer"}],["_attention_backend_default",{"attention_backend":"triton"}],["_data_parallelism_defaults",{"enable_dp_attention":false,"enable_dp_lm_head":false}],["_pipeline_parallel_overlap_disable",{"disable_overlap_schedule":true}],["_speculative_moe_runner_default",{"speculative_moe_runner_backend":"flashinfer_mxfp4"}]],"grpc_worker_threads":4,"_quantization_explicitly_unset":false,"_cuda_graph_config_locked":[["decode","max_bs"]],"_declarations_materialized":true,"_runtime_mutations":[["model_runner.chunked_prefix_cache_gate",{"disable_chunked_prefix_cache":true}],["mamba_pool.memory_budget",{"max_mamba_cache_size":54}]],"_in_override":false,"_mx_config_cache":{},"max_speculative_num_draft_tokens":null,"last_gen_throughput":0.0,"memory_usage":{"weight":43.09,"kvcache":8.16,"token_capacity":3801408,"graph":0},"effective_max_running_requests_per_dp":54}],"version":"0.5.16","kv_events":null}],"decode":[{"model_path":"/data/hf_models/Kimi-K3","tokenizer_path":"/data/hf_models/Kimi-K3","tokenizer_mode":"auto","tokenizer_backend":"huggingface","tokenizer_worker_num":1,"detokenizer_worker_num":1,"skip_tokenizer_init":false,"load_format":"auto","model_loader_extra_config":"{}","trust_remote_code":true,"context_length":null,"is_embedding":false,"enable_multimodal":null,"revision":null,"model_impl":"auto","model_config_parser":"auto","json_model_override_args":"{}","dtype":"auto","quantization":null,"quantization_param_path":null,"kv_cache_dtype":"auto","enable_fp32_lm_head":false,"modelopt_quant":null,"modelopt_checkpoint_restore_path":null,"modelopt_checkpoint_save_path":null,"modelopt_export_path":null,"quantize_and_serve":false,"rl_quant_profile":null,"enable_tf32_matmul":false,"mem_fraction_static":0.88,"max_running_requests":null,"max_queued_requests":null,"max_total_tokens":null,"chunked_prefill_size":8192,"enable_dynamic_chunking":false,"max_prefill_tokens":16384,"prefill_max_requests":null,"schedule_policy":"fcfs","enable_priority_scheduling":false,"disable_priority_preemption":false,"default_priority_value":null,"abort_on_priority_when_disabled":false,"schedule_low_priority_values_first":false,"priority_scheduling_preemption_threshold":10,"retraction_policy":"length","schedule_conservativeness":1.0,"page_size":64,"swa_full_tokens_ratio":0.8,"disable_hybrid_swa_memory":false,"radix_eviction_policy":"lru","prefill_only_disable_kv_cache":false,"disable_radix_cache":true,"enable_page_major_kv_layout":false,"enable_unified_memory":false,"disable_chunked_prefix_cache":false,"disable_overlap_schedule":false,"num_continuous_decode_steps":1,"scheduler_recv_interval":1,"enable_mixed_chunk":false,"nccl_port":null,"dist_timeout":3600,"dist_init_addr":"174.1.60.5:20000","nnodes":4,"node_rank":0,"tp_size":32,"dcp_size":1,"pp_size":1,"pp_max_micro_batch_size":null,"pp_async_batch_depth":0,"dp_size":1,"load_balance_method":"round_robin","attn_cp_size":1,"moe_dp_size":1,"dcp_comm_backend":"ag_rs","dcp_replicate_q_proj":null,"dwdp_size":1,"enable_prefill_cp":false,"cp_strategy":null,"enable_dsa_cache_layer_split":false,"enable_dsa_prefill_context_parallel":false,"dsa_prefill_cp_mode":"round-robin-split","enable_prefill_context_parallel":false,"prefill_cp_mode":"in-seq-split","enable_cp_decode_attn_tp":false,"enable_dp_attention":false,"enable_dp_attention_local_control_broadcast":false,"enable_dp_lm_head":false,"enable_attn_tp_input_scattered":false,"disable_attn_tp_gather":false,"enable_p2p_check":false,"device":"cuda","base_gpu_id":0,"gpu_id_step":1,"random_seed":369235288,"watchdog_timeout":300,"soft_watchdog_timeout":null,"sleep_on_idle":false,"use_ray":false,"custom_sigquit_handler":null,"numa_node":null,"gc_threshold":null,"host":"0.0.0.0","port":30000,"fastapi_root_path":"","smg_grpc_mode":false,"grpc_mode":false,"grpc_port":null,"sidecar":null,"sidecar_args":null,"skip_server_warmup":false,"warmups":null,"enable_http2":false,"ssl_keyfile":null,"ssl_certfile":null,"ssl_ca_certs":null,"ssl_keyfile_password":null,"enable_ssl_refresh":false,"api_key":null,"admin_api_key":null,"served_model_name":"kimi-k3","weight_version":"default","chat_template":null,"hf_chat_template_name":null,"completion_template":null,"file_storage_path":"sglang_storage","enable_cache_report":false,"reasoning_parser":null,"default_chat_template_kwargs":null,"strip_thinking_cache":false,"enable_strict_thinking":false,"tool_call_parser":null,"tool_server":null,"sampling_defaults":"model","asr_max_buffer_seconds":60,"asr_max_concurrent_sessions":32,"preferred_sampling_params":null,"allow_auto_truncate":false,"stream_interval":1,"batch_notify_size":16,"stream_response_default_include_usage":false,"incremental_streaming_output":false,"enable_streaming_session":false,"enable_session_radix_cache":false,"log_level":"info","log_level_http":null,"log_requests":false,"log_requests_level":2,"log_requests_format":"text","log_requests_target":null,"uvicorn_access_log_exclude_prefixes":[],"crash_dump_folder":null,"show_time_cost":false,"enable_metrics":false,"smg_http_sidecar_port":null,"enable_mfu_metrics":false,"enable_metrics_for_all_schedulers":false,"load_snapshot_publish_interval":15,"tokenizer_metrics_custom_labels_header":"x-custom-labels","tokenizer_metrics_allowed_custom_labels":null,"extra_metric_labels":null,"bucket_time_to_first_token":null,"bucket_inter_token_latency":null,"bucket_e2e_request_latency":null,"prompt_tokens_buckets":null,"generation_tokens_buckets":null,"gc_warning_threshold_secs":0.0,"decode_log_interval":40,"enable_request_time_stats_logging":false,"kv_events_config":null,"enable_forward_pass_metrics":false,"forward_pass_metrics_worker_id":"","forward_pass_metrics_ipc_name":null,"enable_trace":false,"trace_modules":"request","otlp_traces_endpoint":"localhost:4317","export_metrics_to_file":false,"export_metrics_to_file_dir":null,"stat_loggers":null,"constrained_json_whitespace_pattern":null,"constrained_json_disable_any_whitespace":false,"attention_backend":"triton","decode_attention_backend":null,"prefill_attention_backend":null,"sampling_backend":"flashinfer","grammar_backend":"xgrammar","radix_cache_backend":null,"mm_attention_backend":null,"fp8_gemm_runner_backend":"auto","fp4_gemm_runner_backend":"auto","bf16_gemm_backend":"auto","dsa_prefill_backend":null,"dsa_decode_backend":null,"dsa_paged_mqa_logits_backend":"auto","dsa_topk_backend":"sgl-kernel","disable_flashinfer_autotune":false,"flashinfer_autotune_skip_ops":null,"mamba_backend":"triton","cuda_graph_config":{"decode":{"backend":"full","max_bs":16,"bs":[1,2,4,8,12,16],"tc_compiler":"eager","full_prefill_max_req":null},"prefill":{"backend":"disabled","max_bs":2048,"bs":[4,8,12,16,20,24,28,32,48,64,80,96,112,128,144,160,176,192,208,224,240,256,288,320,352,384,416,448,480,512,576,640,704,768,832,896,960,1024,1280,1536,1792,2048],"tc_compiler":"eager","full_prefill_max_req":null}},"cuda_graph_backend_decode":null,"cuda_graph_backend_prefill":null,"cuda_graph_max_bs_decode":16,"cuda_graph_max_bs_prefill":null,"cuda_graph_bs_decode":null,"cuda_graph_bs_prefill":null,"cuda_graph_tc_compiler":null,"disable_prefill_cuda_graph":false,"disable_decode_cuda_graph":false,"disable_cuda_graph":false,"disable_cuda_graph_padding":false,"enable_profile_cuda_graph":false,"enable_cudagraph_gc":false,"debug_cuda_graph":false,"enable_layerwise_nvtx_marker":false,"enable_nccl_nvls":false,"enable_symm_mem":false,"triton_attention_reduce_in_fp32":false,"triton_attention_num_kv_splits":8,"triton_attention_split_tile_size":null,"flashinfer_mla_disable_ragged":false,"enable_fused_qk_norm_rope":false,"enable_precise_embedding_interpolation":false,"enable_fused_moe_sum_all_reduce":false,"enable_deepseek_v4_fp4_indexer":false,"disable_custom_all_reduce":false,"enable_mscclpp":false,"enable_torch_symm_mem":false,"enable_scattered_sconv":false,"pre_warm_nccl":false,"enable_quant_communications":false,"enable_flashinfer_allreduce_fusion":false,"enforce_disable_flashinfer_allreduce_fusion":false,"flashinfer_allreduce_fusion_backend":null,"enable_aiter_allreduce_fusion":false,"enable_torch_compile":false,"enable_torch_compile_debug_mode":false,"torch_compile_max_bs":32,"torchao_config":"","speculative_algorithm":null,"speculative_draft_model_path":null,"speculative_draft_model_revision":null,"speculative_draft_load_format":null,"speculative_num_steps":null,"speculative_eagle_topk":null,"speculative_num_draft_tokens":null,"speculative_dflash_block_size":null,"speculative_dspark_block_size":null,"speculative_dspark_sps_table_path":null,"speculative_dspark_confidence_sts_path":null,"speculative_dspark_align_verify_tokens_to_graph_tier":false,"speculative_accept_threshold_single":1.0,"speculative_accept_threshold_acc":1.0,"speculative_use_rejection_sampling":false,"speculative_token_map":null,"speculative_attention_mode":"prefill","speculative_draft_attention_backend":null,"speculative_draft_window_size":null,"speculative_moe_runner_backend":"marlin","speculative_moe_a2a_backend":null,"speculative_draft_model_quantization":null,"speculative_skip_dp_mlp_sync":false,"enable_multi_layer_eagle":false,"speculative_adaptive":false,"speculative_adaptive_config":null,"decoupled_spec_bind_endpoint":null,"decoupled_spec_connect_endpoints":null,"decoupled_spec_rank":null,"decoupled_spec_role":"null","spec_trace_dir":null,"speculative_ngram_min_bfs_breadth":1,"speculative_ngram_max_bfs_breadth":10,"speculative_ngram_match_type":"BFS","speculative_ngram_max_trie_depth":18,"speculative_ngram_capacity":10000000,"speculative_ngram_external_corpus_path":null,"speculative_ngram_external_sam_budget":0,"speculative_ngram_external_corpus_max_tokens":10000000,"ep_size":32,"moe_a2a_backend":"none","moe_runner_backend":"marlin","flashinfer_mxfp4_moe_precision":"default","deepep_mode":"auto","fuseep_mode":2,"deepep_dispatcher_output_dtype":"auto","ep_num_redundant_experts":0,"ep_dispatch_algorithm":null,"init_expert_location":"trivial","enable_eplb":false,"eplb_algorithm":"auto","eplb_rebalance_num_iterations":1000,"eplb_rebalance_layers_per_chunk":null,"eplb_min_rebalancing_utilization_threshold":1.0,"expert_distribution_recorder_mode":null,"expert_distribution_recorder_buffer_size":1000,"enable_expert_distribution_metrics":false,"deepep_config":null,"moe_dense_tp_size":null,"elastic_ep_backend":null,"enable_elastic_expert_backup":false,"mooncake_ib_device":null,"enable_waterfill":false,"ep_join_mode":null,"ep_join_rank_offset":0,"elastic_ep_initial_size":null,"max_ep_size":null,"elastic_ep_scale_timeout":600,"elastic_ep_rejoin":false,"disable_flashinfer_cutlass_moe_fp4_allgather":false,"disable_shared_experts_fusion":false,"enforce_shared_experts_fusion":false,"max_mamba_cache_size":null,"mamba_ssm_dtype":null,"mamba_max_states_per_path":-1,"enable_mamba_cache_stochastic_rounding":false,"mamba_cache_philox_rounds":0,"mamba_full_memory_ratio":0.21,"mamba_radix_cache_strategy":"extra_buffer_lazy","uses_mamba_radix_cache":false,"mamba_track_interval":256,"enable_int8_mamba_checkpoint":false,"int8_mamba_ckpt_size":null,"linear_attn_backend":"triton","linear_attn_decode_backend":null,"linear_attn_prefill_backend":null,"linear_attn_verify_backend":null,"enable_linear_replayssm":false,"linear_replayssm_cache_len":16,"enable_linear_replayssm_spec":false,"enable_hierarchical_cache":false,"hicache_ratio":2.0,"hicache_size":0,"hicache_write_policy":"write_through","hicache_io_backend":"kernel","hicache_mem_layout":"page_first","hicache_storage_backend":null,"hicache_storage_prefetch_policy":"timeout","hicache_storage_backend_extra_config":null,"enable_hisparse":false,"hisparse_config":null,"enable_broadcast_mm_inputs_process":false,"enable_prefix_mm_cache":false,"mm_enable_dp_encoder":false,"mm_process_config":{},"mm_processor_worker_num":0,"mm_io_worker_num":0,"limit_mm_data_per_request":null,"enable_mm_global_cache":false,"disable_fast_image_processor":false,"mm_feature_transport":"cpu","keep_mm_feature_on_device":false,"enable_lora":null,"enable_lora_overlap_loading":null,"max_lora_rank":null,"lora_target_modules":null,"lora_paths":null,"max_loaded_loras":null,"max_loras_per_batch":8,"lora_eviction_policy":"lru","lora_backend":"csgmv","max_lora_chunk_size":16,"experts_shared_outer_loras":null,"lora_use_virtual_experts":false,"lora_strict_loading":false,"lora_drain_wait_threshold":0.0,"enable_two_batch_overlap":false,"enable_single_batch_overlap":false,"tbo_token_distribution_threshold":0.48,"cpu_offload_gb":0,"offload_group_size":-1,"offload_num_in_group":1,"offload_prefetch_step":1,"offload_mode":"cpu","enable_lmcache":false,"lmcache_config_file":null,"enable_flexkv":false,"flexkv_config_file":null,"kt_weight_path":null,"kt_method":"AMXINT4","kt_cpuinfer":null,"kt_threadpool_count":2,"kt_num_gpu_experts":null,"kt_max_deferred_experts_per_token":null,"dllm_algorithm":null,"dllm_algorithm_config":null,"dllm_fdfo":true,"disaggregation_mode":"decode","disaggregation_transfer_backend":"mooncake","disaggregation_bootstrap_port":28800,"disaggregation_ib_device":"mlx5_0,mlx5_1,mlx5_2,mlx5_3","disaggregation_decode_enable_radix_cache":false,"disaggregation_decode_enable_offload_kvcache":false,"num_reserved_decode_tokens":512,"disaggregation_decode_extra_slots":0,"disaggregation_decode_polling_interval":1,"optimistic_prefill_attempts":0,"encoder_only":false,"language_only":false,"encoder_transfer_backend":"zmq_to_tokenizer","encoder_urls":[],"encoder_bootstrap_port":8997,"encoder_register_urls":[],"enable_adaptive_dispatch_to_encoder":false,"enable_pdmux":false,"pdmux_config_path":null,"sm_group_num":8,"custom_weight_loader":[],"weight_loader_disable_mmap":false,"weight_loader_prefetch_checkpoints":false,"weight_loader_prefetch_num_threads":4,"weight_loader_drop_cache_after_load":false,"remote_instance_weight_loader_seed_instance_ip":null,"remote_instance_weight_loader_seed_instance_service_port":null,"remote_instance_weight_loader_send_weights_group_ports":null,"remote_instance_weight_loader_backend":"nccl","remote_instance_weight_loader_start_seed_via_transfer_engine":false,"engine_info_bootstrap_port":6789,"modelexpress_config":null,"download_dir":null,"model_checksum":null,"delete_ckpt_after_loading":false,"decrypted_config_file":null,"decrypted_draft_config_file":null,"checkpoint_engine_wait_weights_before_ready":false,"enable_prefill_delayer":false,"prefill_delayer_max_delay_passes":30,"prefill_delayer_token_usage_low_watermark":null,"prefill_delayer_forward_passes_buckets":null,"prefill_delayer_wait_seconds_buckets":null,"prefill_delayer_queue_min_ratio":null,"prefill_delayer_max_delay_ms":null,"min_free_slots_delay":null,"enable_deterministic_inference":false,"rl_on_policy_target":null,"kv_canary":"none","kv_canary_real_data":"none","kv_canary_sweep_interval":0,"enable_dynamic_batch_tokenizer":false,"dynamic_batch_tokenizer_batch_size":32,"dynamic_batch_tokenizer_batch_timeout":0.002,"enable_tokenizer_batch_encode":false,"disable_tokenizer_batch_decode":false,"debug_tensor_dump_output_folder":null,"debug_tensor_dump_layers":null,"debug_tensor_dump_input_file":null,"enable_memory_saver":false,"enable_weights_cpu_backup":false,"enable_draft_weights_cpu_backup":false,"enable_custom_logit_processor":false,"enable_return_hidden_states":false,"enable_return_routed_experts":false,"enable_return_indexer_topk":false,"disable_outlines_disk_cache":false,"enable_mis":false,"weight_cache_mode":"off","weight_cache_socket":null,"weight_cache_timeout":1800,"forward_hooks":null,"msprobe_dump_config":null,"status":"ready","max_total_num_tokens":440256,"max_req_input_len":440250,"internal_states":[{"model_path":"/data/hf_models/Kimi-K3","tokenizer_path":"/data/hf_models/Kimi-K3","tokenizer_mode":"auto","tokenizer_backend":"huggingface","tokenizer_worker_num":1,"detokenizer_worker_num":1,"skip_tokenizer_init":false,"load_format":"auto","model_loader_extra_config":"{}","trust_remote_code":true,"context_length":null,"is_embedding":false,"enable_multimodal":null,"revision":null,"model_impl":"auto","model_config_parser":"auto","json_model_override_args":"{}","dtype":"auto","quantization":null,"quantization_param_path":null,"kv_cache_dtype":"auto","enable_fp32_lm_head":false,"modelopt_quant":null,"modelopt_checkpoint_restore_path":null,"modelopt_checkpoint_save_path":null,"modelopt_export_path":null,"quantize_and_serve":false,"rl_quant_profile":null,"enable_tf32_matmul":false,"mem_fraction_static":0.88,"max_running_requests":null,"max_queued_requests":null,"max_total_tokens":null,"chunked_prefill_size":8192,"enable_dynamic_chunking":false,"max_prefill_tokens":16384,"prefill_max_requests":null,"schedule_policy":"fcfs","enable_priority_scheduling":false,"disable_priority_preemption":false,"default_priority_value":null,"abort_on_priority_when_disabled":false,"schedule_low_priority_values_first":false,"priority_scheduling_preemption_threshold":10,"retraction_policy":"length","schedule_conservativeness":1.0,"page_size":64,"swa_full_tokens_ratio":0.8,"disable_hybrid_swa_memory":false,"radix_eviction_policy":"lru","prefill_only_disable_kv_cache":false,"disable_radix_cache":true,"enable_page_major_kv_layout":false,"enable_unified_memory":false,"disable_chunked_prefix_cache":true,"disable_overlap_schedule":false,"num_continuous_decode_steps":1,"scheduler_recv_interval":1,"enable_mixed_chunk":false,"nccl_port":null,"dist_timeout":3600,"dist_init_addr":"174.1.60.5:20000","nnodes":4,"node_rank":0,"tp_size":32,"dcp_size":1,"pp_size":1,"pp_max_micro_batch_size":181,"pp_async_batch_depth":0,"dp_size":1,"load_balance_method":"round_robin","attn_cp_size":1,"moe_dp_size":1,"dcp_comm_backend":"ag_rs","dcp_replicate_q_proj":null,"dwdp_size":1,"enable_prefill_cp":false,"cp_strategy":null,"enable_dsa_cache_layer_split":false,"enable_dsa_prefill_context_parallel":false,"dsa_prefill_cp_mode":"round-robin-split","enable_prefill_context_parallel":false,"prefill_cp_mode":"in-seq-split","enable_cp_decode_attn_tp":false,"enable_dp_attention":false,"enable_dp_attention_local_control_broadcast":false,"enable_dp_lm_head":false,"enable_attn_tp_input_scattered":false,"disable_attn_tp_gather":false,"enable_p2p_check":false,"device":"cuda","base_gpu_id":0,"gpu_id_step":1,"random_seed":369235288,"watchdog_timeout":300,"soft_watchdog_timeout":null,"sleep_on_idle":false,"use_ray":false,"numa_node":null,"gc_threshold":null,"host":"0.0.0.0","port":30000,"fastapi_root_path":"","smg_grpc_mode":false,"grpc_mode":false,"grpc_port":null,"sidecar":null,"sidecar_args":null,"skip_server_warmup":false,"warmups":null,"enable_http2":false,"ssl_keyfile":null,"ssl_certfile":null,"ssl_ca_certs":null,"ssl_keyfile_password":null,"enable_ssl_refresh":false,"api_key":null,"admin_api_key":null,"served_model_name":"kimi-k3","weight_version":"default","chat_template":null,"hf_chat_template_name":null,"completion_template":null,"file_storage_path":"sglang_storage","enable_cache_report":false,"reasoning_parser":null,"default_chat_template_kwargs":null,"strip_thinking_cache":false,"enable_strict_thinking":false,"tool_call_parser":null,"tool_server":null,"sampling_defaults":"model","asr_max_buffer_seconds":60,"asr_max_concurrent_sessions":32,"preferred_sampling_params":null,"allow_auto_truncate":false,"stream_interval":1,"batch_notify_size":16,"stream_response_default_include_usage":false,"incremental_streaming_output":false,"enable_streaming_session":false,"enable_session_radix_cache":false,"log_level":"info","log_level_http":null,"log_requests":false,"log_requests_level":2,"log_requests_format":"text","log_requests_target":null,"uvicorn_access_log_exclude_prefixes":[],"crash_dump_folder":null,"show_time_cost":false,"enable_metrics":false,"smg_http_sidecar_port":null,"enable_mfu_metrics":false,"enable_metrics_for_all_schedulers":false,"load_snapshot_publish_interval":15,"tokenizer_metrics_custom_labels_header":"x-custom-labels","tokenizer_metrics_allowed_custom_labels":null,"extra_metric_labels":null,"bucket_time_to_first_token":null,"bucket_inter_token_latency":null,"bucket_e2e_request_latency":null,"prompt_tokens_buckets":null,"generation_tokens_buckets":null,"gc_warning_threshold_secs":0.0,"decode_log_interval":40,"enable_request_time_stats_logging":false,"kv_events_config":null,"enable_forward_pass_metrics":false,"forward_pass_metrics_worker_id":"","forward_pass_metrics_ipc_name":null,"enable_trace":false,"trace_modules":"request","otlp_traces_endpoint":"localhost:4317","export_metrics_to_file":false,"export_metrics_to_file_dir":null,"stat_loggers":null,"constrained_json_whitespace_pattern":null,"constrained_json_disable_any_whitespace":false,"attention_backend":"triton","decode_attention_backend":null,"prefill_attention_backend":null,"sampling_backend":"flashinfer","grammar_backend":"xgrammar","radix_cache_backend":null,"mm_attention_backend":null,"fp8_gemm_runner_backend":"auto","fp4_gemm_runner_backend":"auto","bf16_gemm_backend":"auto","dsa_prefill_backend":null,"dsa_decode_backend":null,"dsa_paged_mqa_logits_backend":"auto","dsa_topk_backend":"sgl-kernel","disable_flashinfer_autotune":false,"flashinfer_autotune_skip_ops":null,"mamba_backend":"triton","cuda_graph_config":{"decode":{"backend":"full","max_bs":16,"bs":[1,2,4,8,12,16],"tc_compiler":"eager","full_prefill_max_req":null},"prefill":{"backend":"disabled","max_bs":2048,"bs":[4,8,12,16,20,24,28,32,48,64,80,96,112,128,144,160,176,192,208,224,240,256,288,320,352,384,416,448,480,512,576,640,704,768,832,896,960,1024,1280,1536,1792,2048],"tc_compiler":"eager","full_prefill_max_req":null}},"cuda_graph_backend_decode":null,"cuda_graph_backend_prefill":null,"cuda_graph_max_bs_decode":16,"cuda_graph_max_bs_prefill":null,"cuda_graph_bs_decode":null,"cuda_graph_bs_prefill":null,"cuda_graph_tc_compiler":null,"disable_prefill_cuda_graph":false,"disable_decode_cuda_graph":false,"disable_cuda_graph":false,"disable_cuda_graph_padding":false,"enable_profile_cuda_graph":false,"enable_cudagraph_gc":false,"debug_cuda_graph":false,"enable_layerwise_nvtx_marker":false,"enable_nccl_nvls":false,"enable_symm_mem":false,"triton_attention_reduce_in_fp32":false,"triton_attention_num_kv_splits":8,"triton_attention_split_tile_size":null,"flashinfer_mla_disable_ragged":false,"enable_fused_qk_norm_rope":false,"enable_precise_embedding_interpolation":false,"enable_fused_moe_sum_all_reduce":false,"enable_deepseek_v4_fp4_indexer":false,"disable_custom_all_reduce":false,"enable_mscclpp":false,"enable_torch_symm_mem":false,"enable_scattered_sconv":false,"pre_warm_nccl":false,"enable_quant_communications":false,"enable_flashinfer_allreduce_fusion":false,"enforce_disable_flashinfer_allreduce_fusion":false,"flashinfer_allreduce_fusion_backend":null,"enable_aiter_allreduce_fusion":false,"enable_torch_compile":false,"enable_torch_compile_debug_mode":false,"torch_compile_max_bs":32,"torchao_config":"","speculative_algorithm":null,"speculative_draft_model_path":null,"speculative_draft_model_revision":null,"speculative_draft_load_format":null,"speculative_num_steps":null,"speculative_eagle_topk":null,"speculative_num_draft_tokens":null,"speculative_dflash_block_size":null,"speculative_dspark_block_size":null,"speculative_dspark_sps_table_path":null,"speculative_dspark_confidence_sts_path":null,"speculative_dspark_align_verify_tokens_to_graph_tier":false,"speculative_accept_threshold_single":1.0,"speculative_accept_threshold_acc":1.0,"speculative_use_rejection_sampling":false,"speculative_token_map":null,"speculative_attention_mode":"prefill","speculative_draft_attention_backend":null,"speculative_draft_window_size":null,"speculative_moe_runner_backend":"marlin","speculative_moe_a2a_backend":null,"speculative_draft_model_quantization":null,"speculative_skip_dp_mlp_sync":false,"enable_multi_layer_eagle":false,"speculative_adaptive":false,"speculative_adaptive_config":null,"decoupled_spec_bind_endpoint":null,"decoupled_spec_connect_endpoints":null,"decoupled_spec_rank":null,"decoupled_spec_role":"null","spec_trace_dir":null,"speculative_ngram_min_bfs_breadth":1,"speculative_ngram_max_bfs_breadth":10,"speculative_ngram_match_type":"BFS","speculative_ngram_max_trie_depth":18,"speculative_ngram_capacity":10000000,"speculative_ngram_external_corpus_path":null,"speculative_ngram_external_sam_budget":0,"speculative_ngram_external_corpus_max_tokens":10000000,"ep_size":32,"moe_a2a_backend":"none","moe_runner_backend":"marlin","flashinfer_mxfp4_moe_precision":"default","deepep_mode":"auto","fuseep_mode":2,"deepep_dispatcher_output_dtype":"auto","ep_num_redundant_experts":0,"ep_dispatch_algorithm":null,"init_expert_location":"trivial","enable_eplb":false,"eplb_algorithm":"auto","eplb_rebalance_num_iterations":1000,"eplb_rebalance_layers_per_chunk":null,"eplb_min_rebalancing_utilization_threshold":1.0,"expert_distribution_recorder_mode":null,"expert_distribution_recorder_buffer_size":1000,"enable_expert_distribution_metrics":false,"deepep_config":null,"moe_dense_tp_size":null,"elastic_ep_backend":null,"enable_elastic_expert_backup":false,"mooncake_ib_device":null,"enable_waterfill":false,"ep_join_mode":null,"ep_join_rank_offset":0,"elastic_ep_initial_size":null,"max_ep_size":null,"elastic_ep_scale_timeout":600,"elastic_ep_rejoin":false,"disable_flashinfer_cutlass_moe_fp4_allgather":false,"disable_shared_experts_fusion":false,"enforce_shared_experts_fusion":false,"max_mamba_cache_size":181,"mamba_ssm_dtype":null,"mamba_max_states_per_path":-1,"enable_mamba_cache_stochastic_rounding":false,"mamba_cache_philox_rounds":0,"mamba_full_memory_ratio":0.21,"mamba_radix_cache_strategy":"extra_buffer_lazy","uses_mamba_radix_cache":false,"mamba_track_interval":256,"enable_int8_mamba_checkpoint":false,"int8_mamba_ckpt_size":null,"linear_attn_backend":"triton","linear_attn_decode_backend":null,"linear_attn_prefill_backend":null,"linear_attn_verify_backend":null,"enable_linear_replayssm":false,"linear_replayssm_cache_len":16,"enable_linear_replayssm_spec":false,"enable_hierarchical_cache":false,"hicache_ratio":2.0,"hicache_size":0,"hicache_write_policy":"write_through","hicache_io_backend":"kernel","hicache_mem_layout":"page_first","hicache_storage_backend":null,"hicache_storage_prefetch_policy":"timeout","hicache_storage_backend_extra_config":null,"enable_hisparse":false,"hisparse_config":null,"enable_broadcast_mm_inputs_process":false,"enable_prefix_mm_cache":false,"mm_enable_dp_encoder":false,"mm_process_config":{},"mm_processor_worker_num":0,"mm_io_worker_num":0,"limit_mm_data_per_request":null,"enable_mm_global_cache":false,"disable_fast_image_processor":false,"mm_feature_transport":"cpu","keep_mm_feature_on_device":false,"enable_lora":null,"enable_lora_overlap_loading":null,"max_lora_rank":null,"lora_target_modules":null,"lora_paths":null,"max_loaded_loras":null,"max_loras_per_batch":8,"lora_eviction_policy":"lru","lora_backend":"csgmv","max_lora_chunk_size":16,"experts_shared_outer_loras":null,"lora_use_virtual_experts":false,"lora_strict_loading":false,"lora_drain_wait_threshold":0.0,"enable_two_batch_overlap":false,"enable_single_batch_overlap":false,"tbo_token_distribution_threshold":0.48,"cpu_offload_gb":0,"offload_group_size":-1,"offload_num_in_group":1,"offload_prefetch_step":1,"offload_mode":"cpu","enable_lmcache":false,"lmcache_config_file":null,"enable_flexkv":false,"flexkv_config_file":null,"kt_weight_path":null,"kt_method":"AMXINT4","kt_cpuinfer":null,"kt_threadpool_count":2,"kt_num_gpu_experts":null,"kt_max_deferred_experts_per_token":null,"dllm_algorithm":null,"dllm_algorithm_config":null,"dllm_fdfo":true,"disaggregation_mode":"decode","disaggregation_transfer_backend":"mooncake","disaggregation_bootstrap_port":28800,"disaggregation_ib_device":"mlx5_0,mlx5_1,mlx5_2,mlx5_3","disaggregation_decode_enable_radix_cache":false,"disaggregation_decode_enable_offload_kvcache":false,"num_reserved_decode_tokens":512,"disaggregation_decode_extra_slots":0,"disaggregation_decode_polling_interval":1,"optimistic_prefill_attempts":0,"encoder_only":false,"language_only":false,"encoder_transfer_backend":"zmq_to_tokenizer","encoder_urls":[],"encoder_bootstrap_port":8997,"encoder_register_urls":[],"enable_adaptive_dispatch_to_encoder":false,"enable_pdmux":false,"pdmux_config_path":null,"sm_group_num":8,"custom_weight_loader":[],"weight_loader_disable_mmap":false,"weight_loader_prefetch_checkpoints":false,"weight_loader_prefetch_num_threads":4,"weight_loader_drop_cache_after_load":false,"remote_instance_weight_loader_seed_instance_ip":null,"remote_instance_weight_loader_seed_instance_service_port":null,"remote_instance_weight_loader_send_weights_group_ports":null,"remote_instance_weight_loader_backend":"nccl","remote_instance_weight_loader_start_seed_via_transfer_engine":false,"engine_info_bootstrap_port":6789,"modelexpress_config":null,"download_dir":null,"model_checksum":null,"delete_ckpt_after_loading":false,"decrypted_config_file":null,"decrypted_draft_config_file":null,"checkpoint_engine_wait_weights_before_ready":false,"enable_prefill_delayer":false,"prefill_delayer_max_delay_passes":30,"prefill_delayer_token_usage_low_watermark":null,"prefill_delayer_forward_passes_buckets":null,"prefill_delayer_wait_seconds_buckets":null,"prefill_delayer_queue_min_ratio":null,"prefill_delayer_max_delay_ms":null,"min_free_slots_delay":null,"enable_deterministic_inference":false,"rl_on_policy_target":null,"kv_canary":"none","kv_canary_real_data":"none","kv_canary_sweep_interval":0,"enable_dynamic_batch_tokenizer":false,"dynamic_batch_tokenizer_batch_size":32,"dynamic_batch_tokenizer_batch_timeout":0.002,"enable_tokenizer_batch_encode":false,"disable_tokenizer_batch_decode":false,"debug_tensor_dump_output_folder":null,"debug_tensor_dump_layers":null,"debug_tensor_dump_input_file":null,"enable_memory_saver":false,"enable_weights_cpu_backup":false,"enable_draft_weights_cpu_backup":false,"enable_custom_logit_processor":false,"enable_return_hidden_states":false,"enable_return_routed_experts":false,"enable_return_indexer_topk":false,"disable_outlines_disk_cache":false,"enable_mis":false,"weight_cache_mode":"off","weight_cache_socket":null,"weight_cache_timeout":1800,"forward_hooks":null,"msprobe_dump_config":null,"_resolved_overrides":[["_sampling_backend_default",{"sampling_backend":"flashinfer"}],["_attention_backend_default",{"attention_backend":"triton"}],["_data_parallelism_defaults",{"enable_dp_attention":false,"enable_dp_lm_head":false}],["_speculative_moe_runner_default",{"speculative_moe_runner_backend":"marlin"}]],"grpc_worker_threads":4,"_quantization_explicitly_unset":false,"_cuda_graph_config_locked":[["decode","max_bs"]],"_declarations_materialized":true,"_runtime_mutations":[["model_runner.chunked_prefix_cache_gate",{"disable_chunked_prefix_cache":true}],["mamba_pool.memory_budget",{"max_mamba_cache_size":181}]],"_in_override":false,"_mx_config_cache":{},"max_speculative_num_draft_tokens":null,"last_gen_throughput":26.412623015816244,"memory_usage":{"weight":58.57,"kvcache":11.34,"token_capacity":440256,"graph":0.6},"effective_max_running_requests_per_dp":181}],"version":"0.5.16","kv_events":null}]}
\ No newline at end of file
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-20260827-1325/metadata/status_after_validation.txt b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-20260827-1325/metadata/status_after_validation.txt
new file mode 100644
index 0000000..e150278
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-20260827-1325/metadata/status_after_validation.txt
@@ -0,0 +1,85 @@
+===== node 1 =====
+kimi3_pro6000_pd_pp8_standard_prefill_node0|local/sglang:kimi-k3-sm120-flashinfer-mxfp4-pd1|Up About an hour
+kimi3_pro6000_pd_pp8_standard_mc_master|local/sglang:kimi-k3-sm120-flashinfer-mxfp4-pd1|Up About an hour
+0, 59711 MiB, 0 %
+1, 59711 MiB, 0 %
+2, 59711 MiB, 0 %
+3, 59711 MiB, 0 %
+4, 67385 MiB, 0 %
+5, 67389 MiB, 0 %
+6, 67383 MiB, 0 %
+7, 67395 MiB, 0 %
+===== node 2 =====
+kimi3_pro6000_pd_pp8_standard_prefill_node1|local/sglang:kimi-k3-sm120-flashinfer-mxfp4-pd1|Up About an hour
+0, 70433 MiB, 0 %
+1, 70439 MiB, 0 %
+2, 70439 MiB, 0 %
+3, 70439 MiB, 0 %
+4, 74811 MiB, 0 %
+5, 74811 MiB, 0 %
+6, 74811 MiB, 0 %
+7, 74811 MiB, 0 %
+===== node 3 =====
+kimi3_pro6000_pd_pp8_standard_prefill_node2|local/sglang:kimi-k3-sm120-flashinfer-mxfp4-pd1|Up About an hour
+0, 75679 MiB, 0 %
+1, 75675 MiB, 0 %
+2, 75675 MiB, 0 %
+3, 75675 MiB, 0 %
+4, 76687 MiB, 0 %
+5, 76687 MiB, 0 %
+6, 76683 MiB, 0 %
+7, 76687 MiB, 0 %
+===== node 4 =====
+kimi3_pro6000_pd_pp8_standard_prefill_node3|local/sglang:kimi-k3-sm120-flashinfer-mxfp4-pd1|Up About an hour
+0, 77185 MiB, 0 %
+1, 77185 MiB, 0 %
+2, 77185 MiB, 0 %
+3, 77189 MiB, 0 %
+4, 79665 MiB, 0 %
+5, 79665 MiB, 0 %
+6, 79665 MiB, 0 %
+7, 79665 MiB, 0 %
+===== node 5 =====
+kimi3_pro6000_pd_pp8_standard_router|local/sglang:kimi-k3-sm120-flashinfer-mxfp4-pd1|Up 53 minutes
+kimi3_pro6000_pd_pp8_standard_decode_node0|local/sglang:kimi-k3-sm120-flashinfer-mxfp4-pd1|Up About an hour
+0, 76277 MiB, 0 %
+1, 76261 MiB, 0 %
+2, 76245 MiB, 0 %
+3, 76293 MiB, 0 %
+4, 76317 MiB, 0 %
+5, 76261 MiB, 0 %
+6, 76245 MiB, 0 %
+7, 76333 MiB, 0 %
+===== node 6 =====
+kimi3_pro6000_pd_pp8_standard_decode_node1|local/sglang:kimi-k3-sm120-flashinfer-mxfp4-pd1|Up About an hour
+0, 76317 MiB, 0 %
+1, 76261 MiB, 0 %
+2, 76245 MiB, 0 %
+3, 76333 MiB, 0 %
+4, 76317 MiB, 0 %
+5, 76261 MiB, 0 %
+6, 76245 MiB, 0 %
+7, 76333 MiB, 0 %
+===== node 7 =====
+kimi3_pro6000_pd_pp8_standard_decode_node2|local/sglang:kimi-k3-sm120-flashinfer-mxfp4-pd1|Up About an hour
+0, 76317 MiB, 0 %
+1, 76261 MiB, 0 %
+2, 76245 MiB, 0 %
+3, 76333 MiB, 0 %
+4, 76317 MiB, 0 %
+5, 76261 MiB, 0 %
+6, 76245 MiB, 0 %
+7, 76333 MiB, 0 %
+===== node 8 =====
+kimi3_pro6000_pd_pp8_standard_decode_node3|local/sglang:kimi-k3-sm120-flashinfer-mxfp4-pd1|Up About an hour
+0, 76317 MiB, 0 %
+1, 76261 MiB, 0 %
+2, 76245 MiB, 0 %
+3, 76333 MiB, 0 %
+4, 76277 MiB, 0 %
+5, 76261 MiB, 0 %
+6, 76245 MiB, 0 %
+7, 76293 MiB, 0 %
+p=healthy
+d=healthy
+router=healthy
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-20260827-1325/orchestrator.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-20260827-1325/orchestrator.log
new file mode 100644
index 0000000..f867711
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-20260827-1325/orchestrator.log
@@ -0,0 +1 @@
+[2026-08-27 13:32:54] case=decode_16k_512_c1 isl=16384 osl=512 c=1 prompts=1
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-20260827-1325/summary.csv b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-20260827-1325/summary.csv
new file mode 100644
index 0000000..b853668
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-20260827-1325/summary.csv
@@ -0,0 +1,2 @@
+case,completed,input_throughput,request_throughput,median_ttft_ms,p95_ttft_ms,median_tpot_ms,p95_tpot_ms,median_e2e_latency_ms,p95_e2e_latency_ms
+decode_16k_512_c1,1,595.2741690862215,0.03633265192176645,7701.911773066968,7701.911773066968,38.655149595039205,38.655149595039205,27454.693216132,27454.693216132
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-c8-20260827-1340/bench/decode_16k_512_c8.cmd.txt b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-c8-20260827-1340/bench/decode_16k_512_c8.cmd.txt
new file mode 100644
index 0000000..79278f0
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-c8-20260827-1340/bench/decode_16k_512_c8.cmd.txt
@@ -0,0 +1 @@
+docker run --rm --network host -v /data/hf_models/Kimi-K3:/data/hf_models/Kimi-K3:ro -v /data/hzy/sskj/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-c8-20260827-1340:/data/hzy/sskj/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-c8-20260827-1340 --entrypoint python3 local/sglang:kimi-k3-sm120-flashinfer-mxfp4-pd1 -m sglang.benchmark.serving --backend sglang --host 174.1.60.5 --port 31000 --tokenizer /data/hf_models/Kimi-K3 --dataset-name random-ids --random-input-len 16384 --random-output-len 512 --random-range-ratio 1.0 --num-prompts 40 --max-concurrency 8 --request-rate 10000 --warmup-requests 0 --output-file /data/hzy/sskj/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-c8-20260827-1340/bench/decode_16k_512_c8.jsonl --output-details --disable-tqdm 
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-c8-20260827-1340/bench/decode_16k_512_c8.jsonl b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-c8-20260827-1340/bench/decode_16k_512_c8.jsonl
new file mode 100644
index 0000000..4389cd6
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-c8-20260827-1340/bench/decode_16k_512_c8.jsonl
@@ -0,0 +1 @@
+{"tag": null, "backend": "sglang", "dataset_name": "random-ids", "request_rate": 10000.0, "max_concurrency": 8, "sharegpt_output_len": null, "random_input_len": 16384, "random_output_len": 512, "random_range_ratio": 1.0, "server_info": {"internal_states": [{"model_path": "/data/hf_models/Kimi-K3", "tokenizer_path": "/data/hf_models/Kimi-K3", "tokenizer_mode": "auto", "tokenizer_backend": "huggingface", "tokenizer_worker_num": 1, "detokenizer_worker_num": 1, "skip_tokenizer_init": false, "load_format": "auto", "model_loader_extra_config": "{}", "trust_remote_code": true, "context_length": null, "is_embedding": false, "enable_multimodal": null, "revision": null, "model_impl": "auto", "model_config_parser": "auto", "json_model_override_args": "{}", "dtype": "auto", "quantization": null, "quantization_param_path": null, "kv_cache_dtype": "auto", "enable_fp32_lm_head": false, "modelopt_quant": null, "modelopt_checkpoint_restore_path": null, "modelopt_checkpoint_save_path": null, "modelopt_export_path": null, "quantize_and_serve": false, "rl_quant_profile": null, "enable_tf32_matmul": false, "mem_fraction_static": 0.88, "max_running_requests": null, "max_queued_requests": null, "max_total_tokens": null, "chunked_prefill_size": 8192, "enable_dynamic_chunking": false, "max_prefill_tokens": 16384, "prefill_max_requests": null, "schedule_policy": "fcfs", "enable_priority_scheduling": false, "disable_priority_preemption": false, "default_priority_value": null, "abort_on_priority_when_disabled": false, "schedule_low_priority_values_first": false, "priority_scheduling_preemption_threshold": 10, "retraction_policy": "length", "schedule_conservativeness": 1.0, "page_size": 64, "swa_full_tokens_ratio": 0.8, "disable_hybrid_swa_memory": false, "radix_eviction_policy": "lru", "prefill_only_disable_kv_cache": false, "disable_radix_cache": true, "enable_page_major_kv_layout": false, "enable_unified_memory": false, "disable_chunked_prefix_cache": true, "disable_overlap_schedule": false, "num_continuous_decode_steps": 1, "scheduler_recv_interval": 1, "enable_mixed_chunk": false, "nccl_port": null, "dist_timeout": 3600, "dist_init_addr": "174.1.60.5:20000", "nnodes": 4, "node_rank": 0, "tp_size": 32, "dcp_size": 1, "pp_size": 1, "pp_max_micro_batch_size": 181, "pp_async_batch_depth": 0, "dp_size": 1, "load_balance_method": "round_robin", "attn_cp_size": 1, "moe_dp_size": 1, "dcp_comm_backend": "ag_rs", "dcp_replicate_q_proj": null, "dwdp_size": 1, "enable_prefill_cp": false, "cp_strategy": null, "enable_dsa_cache_layer_split": false, "enable_dsa_prefill_context_parallel": false, "dsa_prefill_cp_mode": "round-robin-split", "enable_prefill_context_parallel": false, "prefill_cp_mode": "in-seq-split", "enable_cp_decode_attn_tp": false, "enable_dp_attention": false, "enable_dp_attention_local_control_broadcast": false, "enable_dp_lm_head": false, "enable_attn_tp_input_scattered": false, "disable_attn_tp_gather": false, "enable_p2p_check": false, "device": "cuda", "base_gpu_id": 0, "gpu_id_step": 1, "random_seed": 369235288, "watchdog_timeout": 300, "soft_watchdog_timeout": null, "sleep_on_idle": false, "use_ray": false, "numa_node": null, "gc_threshold": null, "host": "0.0.0.0", "port": 30000, "fastapi_root_path": "", "smg_grpc_mode": false, "grpc_mode": false, "grpc_port": null, "sidecar": null, "sidecar_args": null, "skip_server_warmup": false, "warmups": null, "enable_http2": false, "ssl_keyfile": null, "ssl_certfile": null, "ssl_ca_certs": null, "ssl_keyfile_password": null, "enable_ssl_refresh": false, "api_key": null, "admin_api_key": null, "served_model_name": "kimi-k3", "weight_version": "default", "chat_template": null, "hf_chat_template_name": null, "completion_template": null, "file_storage_path": "sglang_storage", "enable_cache_report": false, "reasoning_parser": null, "default_chat_template_kwargs": null, "strip_thinking_cache": false, "enable_strict_thinking": false, "tool_call_parser": null, "tool_server": null, "sampling_defaults": "model", "asr_max_buffer_seconds": 60, "asr_max_concurrent_sessions": 32, "preferred_sampling_params": null, "allow_auto_truncate": false, "stream_interval": 1, "batch_notify_size": 16, "stream_response_default_include_usage": false, "incremental_streaming_output": false, "enable_streaming_session": false, "enable_session_radix_cache": false, "log_level": "info", "log_level_http": null, "log_requests": false, "log_requests_level": 2, "log_requests_format": "text", "log_requests_target": null, "uvicorn_access_log_exclude_prefixes": [], "crash_dump_folder": null, "show_time_cost": false, "enable_metrics": false, "smg_http_sidecar_port": null, "enable_mfu_metrics": false, "enable_metrics_for_all_schedulers": false, "load_snapshot_publish_interval": 15, "tokenizer_metrics_custom_labels_header": "x-custom-labels", "tokenizer_metrics_allowed_custom_labels": null, "extra_metric_labels": null, "bucket_time_to_first_token": null, "bucket_inter_token_latency": null, "bucket_e2e_request_latency": null, "prompt_tokens_buckets": null, "generation_tokens_buckets": null, "gc_warning_threshold_secs": 0.0, "decode_log_interval": 40, "enable_request_time_stats_logging": false, "kv_events_config": null, "enable_forward_pass_metrics": false, "forward_pass_metrics_worker_id": "", "forward_pass_metrics_ipc_name": null, "enable_trace": false, "trace_modules": "request", "otlp_traces_endpoint": "localhost:4317", "export_metrics_to_file": false, "export_metrics_to_file_dir": null, "stat_loggers": null, "constrained_json_whitespace_pattern": null, "constrained_json_disable_any_whitespace": false, "attention_backend": "triton", "decode_attention_backend": null, "prefill_attention_backend": null, "sampling_backend": "flashinfer", "grammar_backend": "xgrammar", "radix_cache_backend": null, "mm_attention_backend": null, "fp8_gemm_runner_backend": "auto", "fp4_gemm_runner_backend": "auto", "bf16_gemm_backend": "auto", "dsa_prefill_backend": null, "dsa_decode_backend": null, "dsa_paged_mqa_logits_backend": "auto", "dsa_topk_backend": "sgl-kernel", "disable_flashinfer_autotune": false, "flashinfer_autotune_skip_ops": null, "mamba_backend": "triton", "cuda_graph_config": {"decode": {"backend": "full", "max_bs": 16, "bs": [1, 2, 4, 8, 12, 16], "tc_compiler": "eager", "full_prefill_max_req": null}, "prefill": {"backend": "disabled", "max_bs": 2048, "bs": [4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], "tc_compiler": "eager", "full_prefill_max_req": null}}, "cuda_graph_backend_decode": null, "cuda_graph_backend_prefill": null, "cuda_graph_max_bs_decode": 16, "cuda_graph_max_bs_prefill": null, "cuda_graph_bs_decode": null, "cuda_graph_bs_prefill": null, "cuda_graph_tc_compiler": null, "disable_prefill_cuda_graph": false, "disable_decode_cuda_graph": false, "disable_cuda_graph": false, "disable_cuda_graph_padding": false, "enable_profile_cuda_graph": false, "enable_cudagraph_gc": false, "debug_cuda_graph": false, "enable_layerwise_nvtx_marker": false, "enable_nccl_nvls": false, "enable_symm_mem": false, "triton_attention_reduce_in_fp32": false, "triton_attention_num_kv_splits": 8, "triton_attention_split_tile_size": null, "flashinfer_mla_disable_ragged": false, "enable_fused_qk_norm_rope": false, "enable_precise_embedding_interpolation": false, "enable_fused_moe_sum_all_reduce": false, "enable_deepseek_v4_fp4_indexer": false, "disable_custom_all_reduce": false, "enable_mscclpp": false, "enable_torch_symm_mem": false, "enable_scattered_sconv": false, "pre_warm_nccl": false, "enable_quant_communications": false, "enable_flashinfer_allreduce_fusion": false, "enforce_disable_flashinfer_allreduce_fusion": false, "flashinfer_allreduce_fusion_backend": null, "enable_aiter_allreduce_fusion": false, "enable_torch_compile": false, "enable_torch_compile_debug_mode": false, "torch_compile_max_bs": 32, "torchao_config": "", "speculative_algorithm": null, "speculative_draft_model_path": null, "speculative_draft_model_revision": null, "speculative_draft_load_format": null, "speculative_num_steps": null, "speculative_eagle_topk": null, "speculative_num_draft_tokens": null, "speculative_dflash_block_size": null, "speculative_dspark_block_size": null, "speculative_dspark_sps_table_path": null, "speculative_dspark_confidence_sts_path": null, "speculative_dspark_align_verify_tokens_to_graph_tier": false, "speculative_accept_threshold_single": 1.0, "speculative_accept_threshold_acc": 1.0, "speculative_use_rejection_sampling": false, "speculative_token_map": null, "speculative_attention_mode": "prefill", "speculative_draft_attention_backend": null, "speculative_draft_window_size": null, "speculative_moe_runner_backend": "marlin", "speculative_moe_a2a_backend": null, "speculative_draft_model_quantization": null, "speculative_skip_dp_mlp_sync": false, "enable_multi_layer_eagle": false, "speculative_adaptive": false, "speculative_adaptive_config": null, "decoupled_spec_bind_endpoint": null, "decoupled_spec_connect_endpoints": null, "decoupled_spec_rank": null, "decoupled_spec_role": "null", "spec_trace_dir": null, "speculative_ngram_min_bfs_breadth": 1, "speculative_ngram_max_bfs_breadth": 10, "speculative_ngram_match_type": "BFS", "speculative_ngram_max_trie_depth": 18, "speculative_ngram_capacity": 10000000, "speculative_ngram_external_corpus_path": null, "speculative_ngram_external_sam_budget": 0, "speculative_ngram_external_corpus_max_tokens": 10000000, "ep_size": 32, "moe_a2a_backend": "none", "moe_runner_backend": "marlin", "flashinfer_mxfp4_moe_precision": "default", "deepep_mode": "auto", "fuseep_mode": 2, "deepep_dispatcher_output_dtype": "auto", "ep_num_redundant_experts": 0, "ep_dispatch_algorithm": null, "init_expert_location": "trivial", "enable_eplb": false, "eplb_algorithm": "auto", "eplb_rebalance_num_iterations": 1000, "eplb_rebalance_layers_per_chunk": null, "eplb_min_rebalancing_utilization_threshold": 1.0, "expert_distribution_recorder_mode": null, "expert_distribution_recorder_buffer_size": 1000, "enable_expert_distribution_metrics": false, "deepep_config": null, "moe_dense_tp_size": null, "elastic_ep_backend": null, "enable_elastic_expert_backup": false, "mooncake_ib_device": null, "enable_waterfill": false, "ep_join_mode": null, "ep_join_rank_offset": 0, "elastic_ep_initial_size": null, "max_ep_size": null, "elastic_ep_scale_timeout": 600, "elastic_ep_rejoin": false, "disable_flashinfer_cutlass_moe_fp4_allgather": false, "disable_shared_experts_fusion": false, "enforce_shared_experts_fusion": false, "max_mamba_cache_size": 181, "mamba_ssm_dtype": null, "mamba_max_states_per_path": -1, "enable_mamba_cache_stochastic_rounding": false, "mamba_cache_philox_rounds": 0, "mamba_full_memory_ratio": 0.21, "mamba_radix_cache_strategy": "extra_buffer_lazy", "uses_mamba_radix_cache": false, "mamba_track_interval": 256, "enable_int8_mamba_checkpoint": false, "int8_mamba_ckpt_size": null, "linear_attn_backend": "triton", "linear_attn_decode_backend": null, "linear_attn_prefill_backend": null, "linear_attn_verify_backend": null, "enable_linear_replayssm": false, "linear_replayssm_cache_len": 16, "enable_linear_replayssm_spec": false, "enable_hierarchical_cache": false, "hicache_ratio": 2.0, "hicache_size": 0, "hicache_write_policy": "write_through", "hicache_io_backend": "kernel", "hicache_mem_layout": "page_first", "hicache_storage_backend": null, "hicache_storage_prefetch_policy": "timeout", "hicache_storage_backend_extra_config": null, "enable_hisparse": false, "hisparse_config": null, "enable_broadcast_mm_inputs_process": false, "enable_prefix_mm_cache": false, "mm_enable_dp_encoder": false, "mm_process_config": {}, "mm_processor_worker_num": 0, "mm_io_worker_num": 0, "limit_mm_data_per_request": null, "enable_mm_global_cache": false, "disable_fast_image_processor": false, "mm_feature_transport": "cpu", "keep_mm_feature_on_device": false, "enable_lora": null, "enable_lora_overlap_loading": null, "max_lora_rank": null, "lora_target_modules": null, "lora_paths": null, "max_loaded_loras": null, "max_loras_per_batch": 8, "lora_eviction_policy": "lru", "lora_backend": "csgmv", "max_lora_chunk_size": 16, "experts_shared_outer_loras": null, "lora_use_virtual_experts": false, "lora_strict_loading": false, "lora_drain_wait_threshold": 0.0, "enable_two_batch_overlap": false, "enable_single_batch_overlap": false, "tbo_token_distribution_threshold": 0.48, "cpu_offload_gb": 0, "offload_group_size": -1, "offload_num_in_group": 1, "offload_prefetch_step": 1, "offload_mode": "cpu", "enable_lmcache": false, "lmcache_config_file": null, "enable_flexkv": false, "flexkv_config_file": null, "kt_weight_path": null, "kt_method": "AMXINT4", "kt_cpuinfer": null, "kt_threadpool_count": 2, "kt_num_gpu_experts": null, "kt_max_deferred_experts_per_token": null, "dllm_algorithm": null, "dllm_algorithm_config": null, "dllm_fdfo": true, "disaggregation_mode": "decode", "disaggregation_transfer_backend": "mooncake", "disaggregation_bootstrap_port": 28800, "disaggregation_ib_device": "mlx5_0,mlx5_1,mlx5_2,mlx5_3", "disaggregation_decode_enable_radix_cache": false, "disaggregation_decode_enable_offload_kvcache": false, "num_reserved_decode_tokens": 512, "disaggregation_decode_extra_slots": 0, "disaggregation_decode_polling_interval": 1, "optimistic_prefill_attempts": 0, "encoder_only": false, "language_only": false, "encoder_transfer_backend": "zmq_to_tokenizer", "encoder_urls": [], "encoder_bootstrap_port": 8997, "encoder_register_urls": [], "enable_adaptive_dispatch_to_encoder": false, "enable_pdmux": false, "pdmux_config_path": null, "sm_group_num": 8, "custom_weight_loader": [], "weight_loader_disable_mmap": false, "weight_loader_prefetch_checkpoints": false, "weight_loader_prefetch_num_threads": 4, "weight_loader_drop_cache_after_load": false, "remote_instance_weight_loader_seed_instance_ip": null, "remote_instance_weight_loader_seed_instance_service_port": null, "remote_instance_weight_loader_send_weights_group_ports": null, "remote_instance_weight_loader_backend": "nccl", "remote_instance_weight_loader_start_seed_via_transfer_engine": false, "engine_info_bootstrap_port": 6789, "modelexpress_config": null, "download_dir": null, "model_checksum": null, "delete_ckpt_after_loading": false, "decrypted_config_file": null, "decrypted_draft_config_file": null, "checkpoint_engine_wait_weights_before_ready": false, "enable_prefill_delayer": false, "prefill_delayer_max_delay_passes": 30, "prefill_delayer_token_usage_low_watermark": null, "prefill_delayer_forward_passes_buckets": null, "prefill_delayer_wait_seconds_buckets": null, "prefill_delayer_queue_min_ratio": null, "prefill_delayer_max_delay_ms": null, "min_free_slots_delay": null, "enable_deterministic_inference": false, "rl_on_policy_target": null, "kv_canary": "none", "kv_canary_real_data": "none", "kv_canary_sweep_interval": 0, "enable_dynamic_batch_tokenizer": false, "dynamic_batch_tokenizer_batch_size": 32, "dynamic_batch_tokenizer_batch_timeout": 0.002, "enable_tokenizer_batch_encode": false, "disable_tokenizer_batch_decode": false, "debug_tensor_dump_output_folder": null, "debug_tensor_dump_layers": null, "debug_tensor_dump_input_file": null, "enable_memory_saver": false, "enable_weights_cpu_backup": false, "enable_draft_weights_cpu_backup": false, "enable_custom_logit_processor": false, "enable_return_hidden_states": false, "enable_return_routed_experts": false, "enable_return_indexer_topk": false, "disable_outlines_disk_cache": false, "enable_mis": false, "weight_cache_mode": "off", "weight_cache_socket": null, "weight_cache_timeout": 1800, "forward_hooks": null, "msprobe_dump_config": null, "_resolved_overrides": [["_sampling_backend_default", {"sampling_backend": "flashinfer"}], ["_attention_backend_default", {"attention_backend": "triton"}], ["_data_parallelism_defaults", {"enable_dp_attention": false, "enable_dp_lm_head": false}], ["_speculative_moe_runner_default", {"speculative_moe_runner_backend": "marlin"}]], "grpc_worker_threads": 4, "_quantization_explicitly_unset": false, "_cuda_graph_config_locked": [["decode", "max_bs"]], "_declarations_materialized": true, "_runtime_mutations": [["model_runner.chunked_prefix_cache_gate", {"disable_chunked_prefix_cache": true}], ["mamba_pool.memory_budget", {"max_mamba_cache_size": 181}]], "_in_override": false, "_mx_config_cache": {}, "max_speculative_num_draft_tokens": null, "last_gen_throughput": 48.2592441209455, "memory_usage": {"weight": 58.57, "kvcache": 11.34, "token_capacity": 440256, "graph": 0.6}, "effective_max_running_requests_per_dp": 181}], "prefill": [{"model_path": "/data/hf_models/Kimi-K3", "tokenizer_path": "/data/hf_models/Kimi-K3", "tokenizer_mode": "auto", "tokenizer_backend": "huggingface", "tokenizer_worker_num": 1, "detokenizer_worker_num": 1, "skip_tokenizer_init": false, "load_format": "auto", "model_loader_extra_config": "{}", "trust_remote_code": true, "context_length": null, "is_embedding": false, "enable_multimodal": null, "revision": null, "model_impl": "auto", "model_config_parser": "auto", "json_model_override_args": "{}", "dtype": "auto", "quantization": null, "quantization_param_path": null, "kv_cache_dtype": "auto", "enable_fp32_lm_head": false, "modelopt_quant": null, "modelopt_checkpoint_restore_path": null, "modelopt_checkpoint_save_path": null, "modelopt_export_path": null, "quantize_and_serve": false, "rl_quant_profile": null, "enable_tf32_matmul": false, "mem_fraction_static": 0.88, "max_running_requests": null, "max_queued_requests": null, "max_total_tokens": null, "chunked_prefill_size": 8192, "enable_dynamic_chunking": false, "max_prefill_tokens": 16384, "prefill_max_requests": null, "schedule_policy": "fcfs", "enable_priority_scheduling": false, "disable_priority_preemption": false, "default_priority_value": null, "abort_on_priority_when_disabled": false, "schedule_low_priority_values_first": false, "priority_scheduling_preemption_threshold": 10, "retraction_policy": "length", "schedule_conservativeness": 1.0, "page_size": 64, "swa_full_tokens_ratio": 0.8, "disable_hybrid_swa_memory": false, "radix_eviction_policy": "lru", "prefill_only_disable_kv_cache": false, "disable_radix_cache": true, "enable_page_major_kv_layout": false, "enable_unified_memory": false, "disable_chunked_prefix_cache": false, "disable_overlap_schedule": true, "num_continuous_decode_steps": 1, "scheduler_recv_interval": 1, "enable_mixed_chunk": false, "nccl_port": null, "dist_timeout": 3600, "dist_init_addr": "174.1.60.1:20000", "nnodes": 4, "node_rank": 0, "tp_size": 4, "dcp_size": 1, "pp_size": 8, "pp_max_micro_batch_size": null, "pp_async_batch_depth": 0, "dp_size": 1, "load_balance_method": "follow_bootstrap_room", "attn_cp_size": 1, "moe_dp_size": 1, "dcp_comm_backend": "ag_rs", "dcp_replicate_q_proj": null, "dwdp_size": 1, "enable_prefill_cp": false, "cp_strategy": null, "enable_dsa_cache_layer_split": false, "enable_dsa_prefill_context_parallel": false, "dsa_prefill_cp_mode": "round-robin-split", "enable_prefill_context_parallel": false, "prefill_cp_mode": "in-seq-split", "enable_cp_decode_attn_tp": false, "enable_dp_attention": false, "enable_dp_attention_local_control_broadcast": false, "enable_dp_lm_head": false, "enable_attn_tp_input_scattered": false, "disable_attn_tp_gather": false, "enable_p2p_check": false, "device": "cuda", "base_gpu_id": 0, "gpu_id_step": 1, "random_seed": 984607766, "watchdog_timeout": 300, "soft_watchdog_timeout": null, "sleep_on_idle": false, "use_ray": false, "custom_sigquit_handler": null, "numa_node": null, "gc_threshold": null, "host": "0.0.0.0", "port": 30000, "fastapi_root_path": "", "smg_grpc_mode": false, "grpc_mode": false, "grpc_port": null, "sidecar": null, "sidecar_args": null, "skip_server_warmup": false, "warmups": null, "enable_http2": false, "ssl_keyfile": null, "ssl_certfile": null, "ssl_ca_certs": null, "ssl_keyfile_password": null, "enable_ssl_refresh": false, "api_key": null, "admin_api_key": null, "served_model_name": "kimi-k3", "weight_version": "default", "chat_template": null, "hf_chat_template_name": null, "completion_template": null, "file_storage_path": "sglang_storage", "enable_cache_report": false, "reasoning_parser": null, "default_chat_template_kwargs": null, "strip_thinking_cache": false, "enable_strict_thinking": false, "tool_call_parser": null, "tool_server": null, "sampling_defaults": "model", "asr_max_buffer_seconds": 60, "asr_max_concurrent_sessions": 32, "preferred_sampling_params": null, "allow_auto_truncate": false, "stream_interval": 1, "batch_notify_size": 16, "stream_response_default_include_usage": false, "incremental_streaming_output": false, "enable_streaming_session": false, "enable_session_radix_cache": false, "log_level": "info", "log_level_http": null, "log_requests": false, "log_requests_level": 2, "log_requests_format": "text", "log_requests_target": null, "uvicorn_access_log_exclude_prefixes": [], "crash_dump_folder": null, "show_time_cost": false, "enable_metrics": false, "smg_http_sidecar_port": null, "enable_mfu_metrics": false, "enable_metrics_for_all_schedulers": false, "load_snapshot_publish_interval": 15, "tokenizer_metrics_custom_labels_header": "x-custom-labels", "tokenizer_metrics_allowed_custom_labels": null, "extra_metric_labels": null, "bucket_time_to_first_token": null, "bucket_inter_token_latency": null, "bucket_e2e_request_latency": null, "prompt_tokens_buckets": null, "generation_tokens_buckets": null, "gc_warning_threshold_secs": 0.0, "decode_log_interval": 40, "enable_request_time_stats_logging": false, "kv_events_config": null, "enable_forward_pass_metrics": false, "forward_pass_metrics_worker_id": "", "forward_pass_metrics_ipc_name": null, "enable_trace": false, "trace_modules": "request", "otlp_traces_endpoint": "localhost:4317", "export_metrics_to_file": false, "export_metrics_to_file_dir": null, "stat_loggers": null, "constrained_json_whitespace_pattern": null, "constrained_json_disable_any_whitespace": false, "attention_backend": "triton", "decode_attention_backend": null, "prefill_attention_backend": null, "sampling_backend": "flashinfer", "grammar_backend": "xgrammar", "radix_cache_backend": null, "mm_attention_backend": null, "fp8_gemm_runner_backend": "auto", "fp4_gemm_runner_backend": "auto", "bf16_gemm_backend": "auto", "dsa_prefill_backend": null, "dsa_decode_backend": null, "dsa_paged_mqa_logits_backend": "auto", "dsa_topk_backend": "sgl-kernel", "disable_flashinfer_autotune": false, "flashinfer_autotune_skip_ops": null, "mamba_backend": "triton", "cuda_graph_config": {"decode": {"backend": "disabled", "max_bs": 16, "bs": [1, 2, 4, 8, 12, 16], "tc_compiler": "eager", "full_prefill_max_req": null}, "prefill": {"backend": "disabled", "max_bs": 2048, "bs": [4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], "tc_compiler": "eager", "full_prefill_max_req": null}}, "cuda_graph_backend_decode": null, "cuda_graph_backend_prefill": null, "cuda_graph_max_bs_decode": 16, "cuda_graph_max_bs_prefill": null, "cuda_graph_bs_decode": null, "cuda_graph_bs_prefill": null, "cuda_graph_tc_compiler": null, "disable_prefill_cuda_graph": false, "disable_decode_cuda_graph": false, "disable_cuda_graph": false, "disable_cuda_graph_padding": false, "enable_profile_cuda_graph": false, "enable_cudagraph_gc": false, "debug_cuda_graph": false, "enable_layerwise_nvtx_marker": false, "enable_nccl_nvls": false, "enable_symm_mem": false, "triton_attention_reduce_in_fp32": false, "triton_attention_num_kv_splits": 8, "triton_attention_split_tile_size": null, "flashinfer_mla_disable_ragged": false, "enable_fused_qk_norm_rope": false, "enable_precise_embedding_interpolation": false, "enable_fused_moe_sum_all_reduce": false, "enable_deepseek_v4_fp4_indexer": false, "disable_custom_all_reduce": false, "enable_mscclpp": false, "enable_torch_symm_mem": false, "enable_scattered_sconv": false, "pre_warm_nccl": false, "enable_quant_communications": false, "enable_flashinfer_allreduce_fusion": false, "enforce_disable_flashinfer_allreduce_fusion": false, "flashinfer_allreduce_fusion_backend": null, "enable_aiter_allreduce_fusion": false, "enable_torch_compile": false, "enable_torch_compile_debug_mode": false, "torch_compile_max_bs": 32, "torchao_config": "", "speculative_algorithm": null, "speculative_draft_model_path": null, "speculative_draft_model_revision": null, "speculative_draft_load_format": null, "speculative_num_steps": null, "speculative_eagle_topk": null, "speculative_num_draft_tokens": null, "speculative_dflash_block_size": null, "speculative_dspark_block_size": null, "speculative_dspark_sps_table_path": null, "speculative_dspark_confidence_sts_path": null, "speculative_dspark_align_verify_tokens_to_graph_tier": false, "speculative_accept_threshold_single": 1.0, "speculative_accept_threshold_acc": 1.0, "speculative_use_rejection_sampling": false, "speculative_token_map": null, "speculative_attention_mode": "prefill", "speculative_draft_attention_backend": null, "speculative_draft_window_size": null, "speculative_moe_runner_backend": "flashinfer_mxfp4", "speculative_moe_a2a_backend": null, "speculative_draft_model_quantization": null, "speculative_skip_dp_mlp_sync": false, "enable_multi_layer_eagle": false, "speculative_adaptive": false, "speculative_adaptive_config": null, "decoupled_spec_bind_endpoint": null, "decoupled_spec_connect_endpoints": null, "decoupled_spec_rank": null, "decoupled_spec_role": "null", "spec_trace_dir": null, "speculative_ngram_min_bfs_breadth": 1, "speculative_ngram_max_bfs_breadth": 10, "speculative_ngram_match_type": "BFS", "speculative_ngram_max_trie_depth": 18, "speculative_ngram_capacity": 10000000, "speculative_ngram_external_corpus_path": null, "speculative_ngram_external_sam_budget": 0, "speculative_ngram_external_corpus_max_tokens": 10000000, "ep_size": 4, "moe_a2a_backend": "none", "moe_runner_backend": "flashinfer_mxfp4", "flashinfer_mxfp4_moe_precision": "default", "deepep_mode": "auto", "fuseep_mode": 2, "deepep_dispatcher_output_dtype": "auto", "ep_num_redundant_experts": 0, "ep_dispatch_algorithm": null, "init_expert_location": "trivial", "enable_eplb": false, "eplb_algorithm": "auto", "eplb_rebalance_num_iterations": 1000, "eplb_rebalance_layers_per_chunk": null, "eplb_min_rebalancing_utilization_threshold": 1.0, "expert_distribution_recorder_mode": null, "expert_distribution_recorder_buffer_size": 1000, "enable_expert_distribution_metrics": false, "deepep_config": null, "moe_dense_tp_size": null, "elastic_ep_backend": null, "enable_elastic_expert_backup": false, "mooncake_ib_device": null, "enable_waterfill": false, "ep_join_mode": null, "ep_join_rank_offset": 0, "elastic_ep_initial_size": null, "max_ep_size": null, "elastic_ep_scale_timeout": 600, "elastic_ep_rejoin": false, "disable_flashinfer_cutlass_moe_fp4_allgather": false, "disable_shared_experts_fusion": false, "enforce_shared_experts_fusion": false, "max_mamba_cache_size": null, "mamba_ssm_dtype": null, "mamba_max_states_per_path": -1, "enable_mamba_cache_stochastic_rounding": false, "mamba_cache_philox_rounds": 0, "mamba_full_memory_ratio": 0.36, "mamba_radix_cache_strategy": "extra_buffer_lazy", "uses_mamba_radix_cache": false, "mamba_track_interval": 256, "enable_int8_mamba_checkpoint": false, "int8_mamba_ckpt_size": null, "linear_attn_backend": "triton", "linear_attn_decode_backend": null, "linear_attn_prefill_backend": null, "linear_attn_verify_backend": null, "enable_linear_replayssm": false, "linear_replayssm_cache_len": 16, "enable_linear_replayssm_spec": false, "enable_hierarchical_cache": false, "hicache_ratio": 2.0, "hicache_size": 0, "hicache_write_policy": "write_through", "hicache_io_backend": "kernel", "hicache_mem_layout": "page_first", "hicache_storage_backend": null, "hicache_storage_prefetch_policy": "timeout", "hicache_storage_backend_extra_config": null, "enable_hisparse": false, "hisparse_config": null, "enable_broadcast_mm_inputs_process": false, "enable_prefix_mm_cache": false, "mm_enable_dp_encoder": false, "mm_process_config": {}, "mm_processor_worker_num": 0, "mm_io_worker_num": 0, "limit_mm_data_per_request": null, "enable_mm_global_cache": false, "disable_fast_image_processor": false, "mm_feature_transport": "cpu", "keep_mm_feature_on_device": false, "enable_lora": null, "enable_lora_overlap_loading": null, "max_lora_rank": null, "lora_target_modules": null, "lora_paths": null, "max_loaded_loras": null, "max_loras_per_batch": 8, "lora_eviction_policy": "lru", "lora_backend": "csgmv", "max_lora_chunk_size": 16, "experts_shared_outer_loras": null, "lora_use_virtual_experts": false, "lora_strict_loading": false, "lora_drain_wait_threshold": 0.0, "enable_two_batch_overlap": false, "enable_single_batch_overlap": false, "tbo_token_distribution_threshold": 0.48, "cpu_offload_gb": 0, "offload_group_size": -1, "offload_num_in_group": 1, "offload_prefetch_step": 1, "offload_mode": "cpu", "enable_lmcache": false, "lmcache_config_file": null, "enable_flexkv": false, "flexkv_config_file": null, "kt_weight_path": null, "kt_method": "AMXINT4", "kt_cpuinfer": null, "kt_threadpool_count": 2, "kt_num_gpu_experts": null, "kt_max_deferred_experts_per_token": null, "dllm_algorithm": null, "dllm_algorithm_config": null, "dllm_fdfo": true, "disaggregation_mode": "prefill", "disaggregation_transfer_backend": "mooncake", "disaggregation_bootstrap_port": 28800, "disaggregation_ib_device": "mlx5_0,mlx5_1,mlx5_2,mlx5_3", "disaggregation_decode_enable_radix_cache": false, "disaggregation_decode_enable_offload_kvcache": false, "num_reserved_decode_tokens": 512, "disaggregation_decode_extra_slots": null, "disaggregation_decode_polling_interval": 1, "optimistic_prefill_attempts": 0, "encoder_only": false, "language_only": false, "encoder_transfer_backend": "zmq_to_tokenizer", "encoder_urls": [], "encoder_bootstrap_port": 8997, "encoder_register_urls": [], "enable_adaptive_dispatch_to_encoder": false, "enable_pdmux": false, "pdmux_config_path": null, "sm_group_num": 8, "custom_weight_loader": [], "weight_loader_disable_mmap": false, "weight_loader_prefetch_checkpoints": false, "weight_loader_prefetch_num_threads": 4, "weight_loader_drop_cache_after_load": false, "remote_instance_weight_loader_seed_instance_ip": null, "remote_instance_weight_loader_seed_instance_service_port": null, "remote_instance_weight_loader_send_weights_group_ports": null, "remote_instance_weight_loader_backend": "nccl", "remote_instance_weight_loader_start_seed_via_transfer_engine": false, "engine_info_bootstrap_port": 6789, "modelexpress_config": null, "download_dir": null, "model_checksum": null, "delete_ckpt_after_loading": false, "decrypted_config_file": null, "decrypted_draft_config_file": null, "checkpoint_engine_wait_weights_before_ready": false, "enable_prefill_delayer": false, "prefill_delayer_max_delay_passes": 30, "prefill_delayer_token_usage_low_watermark": null, "prefill_delayer_forward_passes_buckets": null, "prefill_delayer_wait_seconds_buckets": null, "prefill_delayer_queue_min_ratio": null, "prefill_delayer_max_delay_ms": null, "min_free_slots_delay": null, "enable_deterministic_inference": false, "rl_on_policy_target": null, "kv_canary": "none", "kv_canary_real_data": "none", "kv_canary_sweep_interval": 0, "enable_dynamic_batch_tokenizer": false, "dynamic_batch_tokenizer_batch_size": 32, "dynamic_batch_tokenizer_batch_timeout": 0.002, "enable_tokenizer_batch_encode": false, "disable_tokenizer_batch_decode": false, "debug_tensor_dump_output_folder": null, "debug_tensor_dump_layers": null, "debug_tensor_dump_input_file": null, "enable_memory_saver": false, "enable_weights_cpu_backup": false, "enable_draft_weights_cpu_backup": false, "enable_custom_logit_processor": false, "enable_return_hidden_states": false, "enable_return_routed_experts": false, "enable_return_indexer_topk": false, "disable_outlines_disk_cache": false, "enable_mis": false, "weight_cache_mode": "off", "weight_cache_socket": null, "weight_cache_timeout": 1800, "forward_hooks": null, "msprobe_dump_config": null, "status": "ready", "max_total_num_tokens": 3801408, "max_req_input_len": 1048570, "internal_states": [{"model_path": "/data/hf_models/Kimi-K3", "tokenizer_path": "/data/hf_models/Kimi-K3", "tokenizer_mode": "auto", "tokenizer_backend": "huggingface", "tokenizer_worker_num": 1, "detokenizer_worker_num": 1, "skip_tokenizer_init": false, "load_format": "auto", "model_loader_extra_config": "{}", "trust_remote_code": true, "context_length": null, "is_embedding": false, "enable_multimodal": null, "revision": null, "model_impl": "auto", "model_config_parser": "auto", "json_model_override_args": "{}", "dtype": "auto", "quantization": null, "quantization_param_path": null, "kv_cache_dtype": "auto", "enable_fp32_lm_head": false, "modelopt_quant": null, "modelopt_checkpoint_restore_path": null, "modelopt_checkpoint_save_path": null, "modelopt_export_path": null, "quantize_and_serve": false, "rl_quant_profile": null, "enable_tf32_matmul": false, "mem_fraction_static": 0.88, "max_running_requests": null, "max_queued_requests": null, "max_total_tokens": null, "chunked_prefill_size": 8192, "enable_dynamic_chunking": false, "max_prefill_tokens": 16384, "prefill_max_requests": null, "schedule_policy": "fcfs", "enable_priority_scheduling": false, "disable_priority_preemption": false, "default_priority_value": null, "abort_on_priority_when_disabled": false, "schedule_low_priority_values_first": false, "priority_scheduling_preemption_threshold": 10, "retraction_policy": "length", "schedule_conservativeness": 1.0, "page_size": 64, "swa_full_tokens_ratio": 0.8, "disable_hybrid_swa_memory": false, "radix_eviction_policy": "lru", "prefill_only_disable_kv_cache": false, "disable_radix_cache": true, "enable_page_major_kv_layout": false, "enable_unified_memory": false, "disable_chunked_prefix_cache": true, "disable_overlap_schedule": true, "num_continuous_decode_steps": 1, "scheduler_recv_interval": 1, "enable_mixed_chunk": false, "nccl_port": null, "dist_timeout": 3600, "dist_init_addr": "174.1.60.1:20000", "nnodes": 4, "node_rank": 0, "tp_size": 4, "dcp_size": 1, "pp_size": 8, "pp_max_micro_batch_size": 6, "pp_async_batch_depth": 0, "dp_size": 1, "load_balance_method": "follow_bootstrap_room", "attn_cp_size": 1, "moe_dp_size": 1, "dcp_comm_backend": "ag_rs", "dcp_replicate_q_proj": null, "dwdp_size": 1, "enable_prefill_cp": false, "cp_strategy": null, "enable_dsa_cache_layer_split": false, "enable_dsa_prefill_context_parallel": false, "dsa_prefill_cp_mode": "round-robin-split", "enable_prefill_context_parallel": false, "prefill_cp_mode": "in-seq-split", "enable_cp_decode_attn_tp": false, "enable_dp_attention": false, "enable_dp_attention_local_control_broadcast": false, "enable_dp_lm_head": false, "enable_attn_tp_input_scattered": false, "disable_attn_tp_gather": false, "enable_p2p_check": false, "device": "cuda", "base_gpu_id": 0, "gpu_id_step": 1, "random_seed": 984607766, "watchdog_timeout": 300, "soft_watchdog_timeout": null, "sleep_on_idle": false, "use_ray": false, "numa_node": null, "gc_threshold": null, "host": "0.0.0.0", "port": 30000, "fastapi_root_path": "", "smg_grpc_mode": false, "grpc_mode": false, "grpc_port": null, "sidecar": null, "sidecar_args": null, "skip_server_warmup": false, "warmups": null, "enable_http2": false, "ssl_keyfile": null, "ssl_certfile": null, "ssl_ca_certs": null, "ssl_keyfile_password": null, "enable_ssl_refresh": false, "api_key": null, "admin_api_key": null, "served_model_name": "kimi-k3", "weight_version": "default", "chat_template": null, "hf_chat_template_name": null, "completion_template": null, "file_storage_path": "sglang_storage", "enable_cache_report": false, "reasoning_parser": null, "default_chat_template_kwargs": null, "strip_thinking_cache": false, "enable_strict_thinking": false, "tool_call_parser": null, "tool_server": null, "sampling_defaults": "model", "asr_max_buffer_seconds": 60, "asr_max_concurrent_sessions": 32, "preferred_sampling_params": null, "allow_auto_truncate": false, "stream_interval": 1, "batch_notify_size": 16, "stream_response_default_include_usage": false, "incremental_streaming_output": false, "enable_streaming_session": false, "enable_session_radix_cache": false, "log_level": "info", "log_level_http": null, "log_requests": false, "log_requests_level": 2, "log_requests_format": "text", "log_requests_target": null, "uvicorn_access_log_exclude_prefixes": [], "crash_dump_folder": null, "show_time_cost": false, "enable_metrics": false, "smg_http_sidecar_port": null, "enable_mfu_metrics": false, "enable_metrics_for_all_schedulers": false, "load_snapshot_publish_interval": 15, "tokenizer_metrics_custom_labels_header": "x-custom-labels", "tokenizer_metrics_allowed_custom_labels": null, "extra_metric_labels": null, "bucket_time_to_first_token": null, "bucket_inter_token_latency": null, "bucket_e2e_request_latency": null, "prompt_tokens_buckets": null, "generation_tokens_buckets": null, "gc_warning_threshold_secs": 0.0, "decode_log_interval": 40, "enable_request_time_stats_logging": false, "kv_events_config": null, "enable_forward_pass_metrics": false, "forward_pass_metrics_worker_id": "", "forward_pass_metrics_ipc_name": null, "enable_trace": false, "trace_modules": "request", "otlp_traces_endpoint": "localhost:4317", "export_metrics_to_file": false, "export_metrics_to_file_dir": null, "stat_loggers": null, "constrained_json_whitespace_pattern": null, "constrained_json_disable_any_whitespace": false, "attention_backend": "triton", "decode_attention_backend": null, "prefill_attention_backend": null, "sampling_backend": "flashinfer", "grammar_backend": "xgrammar", "radix_cache_backend": null, "mm_attention_backend": null, "fp8_gemm_runner_backend": "auto", "fp4_gemm_runner_backend": "auto", "bf16_gemm_backend": "auto", "dsa_prefill_backend": null, "dsa_decode_backend": null, "dsa_paged_mqa_logits_backend": "auto", "dsa_topk_backend": "sgl-kernel", "disable_flashinfer_autotune": false, "flashinfer_autotune_skip_ops": null, "mamba_backend": "triton", "cuda_graph_config": {"decode": {"backend": "disabled", "max_bs": 16, "bs": [1, 2, 4, 8, 12, 16], "tc_compiler": "eager", "full_prefill_max_req": null}, "prefill": {"backend": "disabled", "max_bs": 2048, "bs": [4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], "tc_compiler": "eager", "full_prefill_max_req": null}}, "cuda_graph_backend_decode": null, "cuda_graph_backend_prefill": null, "cuda_graph_max_bs_decode": 16, "cuda_graph_max_bs_prefill": null, "cuda_graph_bs_decode": null, "cuda_graph_bs_prefill": null, "cuda_graph_tc_compiler": null, "disable_prefill_cuda_graph": false, "disable_decode_cuda_graph": false, "disable_cuda_graph": false, "disable_cuda_graph_padding": false, "enable_profile_cuda_graph": false, "enable_cudagraph_gc": false, "debug_cuda_graph": false, "enable_layerwise_nvtx_marker": false, "enable_nccl_nvls": false, "enable_symm_mem": false, "triton_attention_reduce_in_fp32": false, "triton_attention_num_kv_splits": 8, "triton_attention_split_tile_size": null, "flashinfer_mla_disable_ragged": false, "enable_fused_qk_norm_rope": false, "enable_precise_embedding_interpolation": false, "enable_fused_moe_sum_all_reduce": false, "enable_deepseek_v4_fp4_indexer": false, "disable_custom_all_reduce": false, "enable_mscclpp": false, "enable_torch_symm_mem": false, "enable_scattered_sconv": false, "pre_warm_nccl": false, "enable_quant_communications": false, "enable_flashinfer_allreduce_fusion": false, "enforce_disable_flashinfer_allreduce_fusion": false, "flashinfer_allreduce_fusion_backend": null, "enable_aiter_allreduce_fusion": false, "enable_torch_compile": false, "enable_torch_compile_debug_mode": false, "torch_compile_max_bs": 32, "torchao_config": "", "speculative_algorithm": null, "speculative_draft_model_path": null, "speculative_draft_model_revision": null, "speculative_draft_load_format": null, "speculative_num_steps": null, "speculative_eagle_topk": null, "speculative_num_draft_tokens": null, "speculative_dflash_block_size": null, "speculative_dspark_block_size": null, "speculative_dspark_sps_table_path": null, "speculative_dspark_confidence_sts_path": null, "speculative_dspark_align_verify_tokens_to_graph_tier": false, "speculative_accept_threshold_single": 1.0, "speculative_accept_threshold_acc": 1.0, "speculative_use_rejection_sampling": false, "speculative_token_map": null, "speculative_attention_mode": "prefill", "speculative_draft_attention_backend": null, "speculative_draft_window_size": null, "speculative_moe_runner_backend": "flashinfer_mxfp4", "speculative_moe_a2a_backend": null, "speculative_draft_model_quantization": null, "speculative_skip_dp_mlp_sync": false, "enable_multi_layer_eagle": false, "speculative_adaptive": false, "speculative_adaptive_config": null, "decoupled_spec_bind_endpoint": null, "decoupled_spec_connect_endpoints": null, "decoupled_spec_rank": null, "decoupled_spec_role": "null", "spec_trace_dir": null, "speculative_ngram_min_bfs_breadth": 1, "speculative_ngram_max_bfs_breadth": 10, "speculative_ngram_match_type": "BFS", "speculative_ngram_max_trie_depth": 18, "speculative_ngram_capacity": 10000000, "speculative_ngram_external_corpus_path": null, "speculative_ngram_external_sam_budget": 0, "speculative_ngram_external_corpus_max_tokens": 10000000, "ep_size": 4, "moe_a2a_backend": "none", "moe_runner_backend": "flashinfer_mxfp4", "flashinfer_mxfp4_moe_precision": "default", "deepep_mode": "auto", "fuseep_mode": 2, "deepep_dispatcher_output_dtype": "auto", "ep_num_redundant_experts": 0, "ep_dispatch_algorithm": null, "init_expert_location": "trivial", "enable_eplb": false, "eplb_algorithm": "auto", "eplb_rebalance_num_iterations": 1000, "eplb_rebalance_layers_per_chunk": null, "eplb_min_rebalancing_utilization_threshold": 1.0, "expert_distribution_recorder_mode": null, "expert_distribution_recorder_buffer_size": 1000, "enable_expert_distribution_metrics": false, "deepep_config": null, "moe_dense_tp_size": null, "elastic_ep_backend": null, "enable_elastic_expert_backup": false, "mooncake_ib_device": null, "enable_waterfill": false, "ep_join_mode": null, "ep_join_rank_offset": 0, "elastic_ep_initial_size": null, "max_ep_size": null, "elastic_ep_scale_timeout": 600, "elastic_ep_rejoin": false, "disable_flashinfer_cutlass_moe_fp4_allgather": false, "disable_shared_experts_fusion": false, "enforce_shared_experts_fusion": false, "max_mamba_cache_size": 54, "mamba_ssm_dtype": null, "mamba_max_states_per_path": -1, "enable_mamba_cache_stochastic_rounding": false, "mamba_cache_philox_rounds": 0, "mamba_full_memory_ratio": 0.36, "mamba_radix_cache_strategy": "extra_buffer_lazy", "uses_mamba_radix_cache": false, "mamba_track_interval": 256, "enable_int8_mamba_checkpoint": false, "int8_mamba_ckpt_size": null, "linear_attn_backend": "triton", "linear_attn_decode_backend": null, "linear_attn_prefill_backend": null, "linear_attn_verify_backend": null, "enable_linear_replayssm": false, "linear_replayssm_cache_len": 16, "enable_linear_replayssm_spec": false, "enable_hierarchical_cache": false, "hicache_ratio": 2.0, "hicache_size": 0, "hicache_write_policy": "write_through", "hicache_io_backend": "kernel", "hicache_mem_layout": "page_first", "hicache_storage_backend": null, "hicache_storage_prefetch_policy": "timeout", "hicache_storage_backend_extra_config": null, "enable_hisparse": false, "hisparse_config": null, "enable_broadcast_mm_inputs_process": false, "enable_prefix_mm_cache": false, "mm_enable_dp_encoder": false, "mm_process_config": {}, "mm_processor_worker_num": 0, "mm_io_worker_num": 0, "limit_mm_data_per_request": null, "enable_mm_global_cache": false, "disable_fast_image_processor": false, "mm_feature_transport": "cpu", "keep_mm_feature_on_device": false, "enable_lora": null, "enable_lora_overlap_loading": null, "max_lora_rank": null, "lora_target_modules": null, "lora_paths": null, "max_loaded_loras": null, "max_loras_per_batch": 8, "lora_eviction_policy": "lru", "lora_backend": "csgmv", "max_lora_chunk_size": 16, "experts_shared_outer_loras": null, "lora_use_virtual_experts": false, "lora_strict_loading": false, "lora_drain_wait_threshold": 0.0, "enable_two_batch_overlap": false, "enable_single_batch_overlap": false, "tbo_token_distribution_threshold": 0.48, "cpu_offload_gb": 0, "offload_group_size": -1, "offload_num_in_group": 1, "offload_prefetch_step": 1, "offload_mode": "cpu", "enable_lmcache": false, "lmcache_config_file": null, "enable_flexkv": false, "flexkv_config_file": null, "kt_weight_path": null, "kt_method": "AMXINT4", "kt_cpuinfer": null, "kt_threadpool_count": 2, "kt_num_gpu_experts": null, "kt_max_deferred_experts_per_token": null, "dllm_algorithm": null, "dllm_algorithm_config": null, "dllm_fdfo": true, "disaggregation_mode": "prefill", "disaggregation_transfer_backend": "mooncake", "disaggregation_bootstrap_port": 28800, "disaggregation_ib_device": "mlx5_0,mlx5_1,mlx5_2,mlx5_3", "disaggregation_decode_enable_radix_cache": false, "disaggregation_decode_enable_offload_kvcache": false, "num_reserved_decode_tokens": 512, "disaggregation_decode_extra_slots": null, "disaggregation_decode_polling_interval": 1, "optimistic_prefill_attempts": 0, "encoder_only": false, "language_only": false, "encoder_transfer_backend": "zmq_to_tokenizer", "encoder_urls": [], "encoder_bootstrap_port": 8997, "encoder_register_urls": [], "enable_adaptive_dispatch_to_encoder": false, "enable_pdmux": false, "pdmux_config_path": null, "sm_group_num": 8, "custom_weight_loader": [], "weight_loader_disable_mmap": false, "weight_loader_prefetch_checkpoints": false, "weight_loader_prefetch_num_threads": 4, "weight_loader_drop_cache_after_load": false, "remote_instance_weight_loader_seed_instance_ip": null, "remote_instance_weight_loader_seed_instance_service_port": null, "remote_instance_weight_loader_send_weights_group_ports": null, "remote_instance_weight_loader_backend": "nccl", "remote_instance_weight_loader_start_seed_via_transfer_engine": false, "engine_info_bootstrap_port": 6789, "modelexpress_config": null, "download_dir": null, "model_checksum": null, "delete_ckpt_after_loading": false, "decrypted_config_file": null, "decrypted_draft_config_file": null, "checkpoint_engine_wait_weights_before_ready": false, "enable_prefill_delayer": false, "prefill_delayer_max_delay_passes": 30, "prefill_delayer_token_usage_low_watermark": null, "prefill_delayer_forward_passes_buckets": null, "prefill_delayer_wait_seconds_buckets": null, "prefill_delayer_queue_min_ratio": null, "prefill_delayer_max_delay_ms": null, "min_free_slots_delay": null, "enable_deterministic_inference": false, "rl_on_policy_target": null, "kv_canary": "none", "kv_canary_real_data": "none", "kv_canary_sweep_interval": 0, "enable_dynamic_batch_tokenizer": false, "dynamic_batch_tokenizer_batch_size": 32, "dynamic_batch_tokenizer_batch_timeout": 0.002, "enable_tokenizer_batch_encode": false, "disable_tokenizer_batch_decode": false, "debug_tensor_dump_output_folder": null, "debug_tensor_dump_layers": null, "debug_tensor_dump_input_file": null, "enable_memory_saver": false, "enable_weights_cpu_backup": false, "enable_draft_weights_cpu_backup": false, "enable_custom_logit_processor": false, "enable_return_hidden_states": false, "enable_return_routed_experts": false, "enable_return_indexer_topk": false, "disable_outlines_disk_cache": false, "enable_mis": false, "weight_cache_mode": "off", "weight_cache_socket": null, "weight_cache_timeout": 1800, "forward_hooks": null, "msprobe_dump_config": null, "_resolved_overrides": [["_sampling_backend_default", {"sampling_backend": "flashinfer"}], ["_attention_backend_default", {"attention_backend": "triton"}], ["_data_parallelism_defaults", {"enable_dp_attention": false, "enable_dp_lm_head": false}], ["_pipeline_parallel_overlap_disable", {"disable_overlap_schedule": true}], ["_speculative_moe_runner_default", {"speculative_moe_runner_backend": "flashinfer_mxfp4"}]], "grpc_worker_threads": 4, "_quantization_explicitly_unset": false, "_cuda_graph_config_locked": [["decode", "max_bs"]], "_declarations_materialized": true, "_runtime_mutations": [["model_runner.chunked_prefix_cache_gate", {"disable_chunked_prefix_cache": true}], ["mamba_pool.memory_budget", {"max_mamba_cache_size": 54}]], "_in_override": false, "_mx_config_cache": {}, "max_speculative_num_draft_tokens": null, "last_gen_throughput": 0.0, "memory_usage": {"weight": 43.09, "kvcache": 8.16, "token_capacity": 3801408, "graph": 0}, "effective_max_running_requests_per_dp": 54}], "version": "0.5.16", "kv_events": null}], "decode": [{"model_path": "/data/hf_models/Kimi-K3", "tokenizer_path": "/data/hf_models/Kimi-K3", "tokenizer_mode": "auto", "tokenizer_backend": "huggingface", "tokenizer_worker_num": 1, "detokenizer_worker_num": 1, "skip_tokenizer_init": false, "load_format": "auto", "model_loader_extra_config": "{}", "trust_remote_code": true, "context_length": null, "is_embedding": false, "enable_multimodal": null, "revision": null, "model_impl": "auto", "model_config_parser": "auto", "json_model_override_args": "{}", "dtype": "auto", "quantization": null, "quantization_param_path": null, "kv_cache_dtype": "auto", "enable_fp32_lm_head": false, "modelopt_quant": null, "modelopt_checkpoint_restore_path": null, "modelopt_checkpoint_save_path": null, "modelopt_export_path": null, "quantize_and_serve": false, "rl_quant_profile": null, "enable_tf32_matmul": false, "mem_fraction_static": 0.88, "max_running_requests": null, "max_queued_requests": null, "max_total_tokens": null, "chunked_prefill_size": 8192, "enable_dynamic_chunking": false, "max_prefill_tokens": 16384, "prefill_max_requests": null, "schedule_policy": "fcfs", "enable_priority_scheduling": false, "disable_priority_preemption": false, "default_priority_value": null, "abort_on_priority_when_disabled": false, "schedule_low_priority_values_first": false, "priority_scheduling_preemption_threshold": 10, "retraction_policy": "length", "schedule_conservativeness": 1.0, "page_size": 64, "swa_full_tokens_ratio": 0.8, "disable_hybrid_swa_memory": false, "radix_eviction_policy": "lru", "prefill_only_disable_kv_cache": false, "disable_radix_cache": true, "enable_page_major_kv_layout": false, "enable_unified_memory": false, "disable_chunked_prefix_cache": false, "disable_overlap_schedule": false, "num_continuous_decode_steps": 1, "scheduler_recv_interval": 1, "enable_mixed_chunk": false, "nccl_port": null, "dist_timeout": 3600, "dist_init_addr": "174.1.60.5:20000", "nnodes": 4, "node_rank": 0, "tp_size": 32, "dcp_size": 1, "pp_size": 1, "pp_max_micro_batch_size": null, "pp_async_batch_depth": 0, "dp_size": 1, "load_balance_method": "round_robin", "attn_cp_size": 1, "moe_dp_size": 1, "dcp_comm_backend": "ag_rs", "dcp_replicate_q_proj": null, "dwdp_size": 1, "enable_prefill_cp": false, "cp_strategy": null, "enable_dsa_cache_layer_split": false, "enable_dsa_prefill_context_parallel": false, "dsa_prefill_cp_mode": "round-robin-split", "enable_prefill_context_parallel": false, "prefill_cp_mode": "in-seq-split", "enable_cp_decode_attn_tp": false, "enable_dp_attention": false, "enable_dp_attention_local_control_broadcast": false, "enable_dp_lm_head": false, "enable_attn_tp_input_scattered": false, "disable_attn_tp_gather": false, "enable_p2p_check": false, "device": "cuda", "base_gpu_id": 0, "gpu_id_step": 1, "random_seed": 369235288, "watchdog_timeout": 300, "soft_watchdog_timeout": null, "sleep_on_idle": false, "use_ray": false, "custom_sigquit_handler": null, "numa_node": null, "gc_threshold": null, "host": "0.0.0.0", "port": 30000, "fastapi_root_path": "", "smg_grpc_mode": false, "grpc_mode": false, "grpc_port": null, "sidecar": null, "sidecar_args": null, "skip_server_warmup": false, "warmups": null, "enable_http2": false, "ssl_keyfile": null, "ssl_certfile": null, "ssl_ca_certs": null, "ssl_keyfile_password": null, "enable_ssl_refresh": false, "api_key": null, "admin_api_key": null, "served_model_name": "kimi-k3", "weight_version": "default", "chat_template": null, "hf_chat_template_name": null, "completion_template": null, "file_storage_path": "sglang_storage", "enable_cache_report": false, "reasoning_parser": null, "default_chat_template_kwargs": null, "strip_thinking_cache": false, "enable_strict_thinking": false, "tool_call_parser": null, "tool_server": null, "sampling_defaults": "model", "asr_max_buffer_seconds": 60, "asr_max_concurrent_sessions": 32, "preferred_sampling_params": null, "allow_auto_truncate": false, "stream_interval": 1, "batch_notify_size": 16, "stream_response_default_include_usage": false, "incremental_streaming_output": false, "enable_streaming_session": false, "enable_session_radix_cache": false, "log_level": "info", "log_level_http": null, "log_requests": false, "log_requests_level": 2, "log_requests_format": "text", "log_requests_target": null, "uvicorn_access_log_exclude_prefixes": [], "crash_dump_folder": null, "show_time_cost": false, "enable_metrics": false, "smg_http_sidecar_port": null, "enable_mfu_metrics": false, "enable_metrics_for_all_schedulers": false, "load_snapshot_publish_interval": 15, "tokenizer_metrics_custom_labels_header": "x-custom-labels", "tokenizer_metrics_allowed_custom_labels": null, "extra_metric_labels": null, "bucket_time_to_first_token": null, "bucket_inter_token_latency": null, "bucket_e2e_request_latency": null, "prompt_tokens_buckets": null, "generation_tokens_buckets": null, "gc_warning_threshold_secs": 0.0, "decode_log_interval": 40, "enable_request_time_stats_logging": false, "kv_events_config": null, "enable_forward_pass_metrics": false, "forward_pass_metrics_worker_id": "", "forward_pass_metrics_ipc_name": null, "enable_trace": false, "trace_modules": "request", "otlp_traces_endpoint": "localhost:4317", "export_metrics_to_file": false, "export_metrics_to_file_dir": null, "stat_loggers": null, "constrained_json_whitespace_pattern": null, "constrained_json_disable_any_whitespace": false, "attention_backend": "triton", "decode_attention_backend": null, "prefill_attention_backend": null, "sampling_backend": "flashinfer", "grammar_backend": "xgrammar", "radix_cache_backend": null, "mm_attention_backend": null, "fp8_gemm_runner_backend": "auto", "fp4_gemm_runner_backend": "auto", "bf16_gemm_backend": "auto", "dsa_prefill_backend": null, "dsa_decode_backend": null, "dsa_paged_mqa_logits_backend": "auto", "dsa_topk_backend": "sgl-kernel", "disable_flashinfer_autotune": false, "flashinfer_autotune_skip_ops": null, "mamba_backend": "triton", "cuda_graph_config": {"decode": {"backend": "full", "max_bs": 16, "bs": [1, 2, 4, 8, 12, 16], "tc_compiler": "eager", "full_prefill_max_req": null}, "prefill": {"backend": "disabled", "max_bs": 2048, "bs": [4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], "tc_compiler": "eager", "full_prefill_max_req": null}}, "cuda_graph_backend_decode": null, "cuda_graph_backend_prefill": null, "cuda_graph_max_bs_decode": 16, "cuda_graph_max_bs_prefill": null, "cuda_graph_bs_decode": null, "cuda_graph_bs_prefill": null, "cuda_graph_tc_compiler": null, "disable_prefill_cuda_graph": false, "disable_decode_cuda_graph": false, "disable_cuda_graph": false, "disable_cuda_graph_padding": false, "enable_profile_cuda_graph": false, "enable_cudagraph_gc": false, "debug_cuda_graph": false, "enable_layerwise_nvtx_marker": false, "enable_nccl_nvls": false, "enable_symm_mem": false, "triton_attention_reduce_in_fp32": false, "triton_attention_num_kv_splits": 8, "triton_attention_split_tile_size": null, "flashinfer_mla_disable_ragged": false, "enable_fused_qk_norm_rope": false, "enable_precise_embedding_interpolation": false, "enable_fused_moe_sum_all_reduce": false, "enable_deepseek_v4_fp4_indexer": false, "disable_custom_all_reduce": false, "enable_mscclpp": false, "enable_torch_symm_mem": false, "enable_scattered_sconv": false, "pre_warm_nccl": false, "enable_quant_communications": false, "enable_flashinfer_allreduce_fusion": false, "enforce_disable_flashinfer_allreduce_fusion": false, "flashinfer_allreduce_fusion_backend": null, "enable_aiter_allreduce_fusion": false, "enable_torch_compile": false, "enable_torch_compile_debug_mode": false, "torch_compile_max_bs": 32, "torchao_config": "", "speculative_algorithm": null, "speculative_draft_model_path": null, "speculative_draft_model_revision": null, "speculative_draft_load_format": null, "speculative_num_steps": null, "speculative_eagle_topk": null, "speculative_num_draft_tokens": null, "speculative_dflash_block_size": null, "speculative_dspark_block_size": null, "speculative_dspark_sps_table_path": null, "speculative_dspark_confidence_sts_path": null, "speculative_dspark_align_verify_tokens_to_graph_tier": false, "speculative_accept_threshold_single": 1.0, "speculative_accept_threshold_acc": 1.0, "speculative_use_rejection_sampling": false, "speculative_token_map": null, "speculative_attention_mode": "prefill", "speculative_draft_attention_backend": null, "speculative_draft_window_size": null, "speculative_moe_runner_backend": "marlin", "speculative_moe_a2a_backend": null, "speculative_draft_model_quantization": null, "speculative_skip_dp_mlp_sync": false, "enable_multi_layer_eagle": false, "speculative_adaptive": false, "speculative_adaptive_config": null, "decoupled_spec_bind_endpoint": null, "decoupled_spec_connect_endpoints": null, "decoupled_spec_rank": null, "decoupled_spec_role": "null", "spec_trace_dir": null, "speculative_ngram_min_bfs_breadth": 1, "speculative_ngram_max_bfs_breadth": 10, "speculative_ngram_match_type": "BFS", "speculative_ngram_max_trie_depth": 18, "speculative_ngram_capacity": 10000000, "speculative_ngram_external_corpus_path": null, "speculative_ngram_external_sam_budget": 0, "speculative_ngram_external_corpus_max_tokens": 10000000, "ep_size": 32, "moe_a2a_backend": "none", "moe_runner_backend": "marlin", "flashinfer_mxfp4_moe_precision": "default", "deepep_mode": "auto", "fuseep_mode": 2, "deepep_dispatcher_output_dtype": "auto", "ep_num_redundant_experts": 0, "ep_dispatch_algorithm": null, "init_expert_location": "trivial", "enable_eplb": false, "eplb_algorithm": "auto", "eplb_rebalance_num_iterations": 1000, "eplb_rebalance_layers_per_chunk": null, "eplb_min_rebalancing_utilization_threshold": 1.0, "expert_distribution_recorder_mode": null, "expert_distribution_recorder_buffer_size": 1000, "enable_expert_distribution_metrics": false, "deepep_config": null, "moe_dense_tp_size": null, "elastic_ep_backend": null, "enable_elastic_expert_backup": false, "mooncake_ib_device": null, "enable_waterfill": false, "ep_join_mode": null, "ep_join_rank_offset": 0, "elastic_ep_initial_size": null, "max_ep_size": null, "elastic_ep_scale_timeout": 600, "elastic_ep_rejoin": false, "disable_flashinfer_cutlass_moe_fp4_allgather": false, "disable_shared_experts_fusion": false, "enforce_shared_experts_fusion": false, "max_mamba_cache_size": null, "mamba_ssm_dtype": null, "mamba_max_states_per_path": -1, "enable_mamba_cache_stochastic_rounding": false, "mamba_cache_philox_rounds": 0, "mamba_full_memory_ratio": 0.21, "mamba_radix_cache_strategy": "extra_buffer_lazy", "uses_mamba_radix_cache": false, "mamba_track_interval": 256, "enable_int8_mamba_checkpoint": false, "int8_mamba_ckpt_size": null, "linear_attn_backend": "triton", "linear_attn_decode_backend": null, "linear_attn_prefill_backend": null, "linear_attn_verify_backend": null, "enable_linear_replayssm": false, "linear_replayssm_cache_len": 16, "enable_linear_replayssm_spec": false, "enable_hierarchical_cache": false, "hicache_ratio": 2.0, "hicache_size": 0, "hicache_write_policy": "write_through", "hicache_io_backend": "kernel", "hicache_mem_layout": "page_first", "hicache_storage_backend": null, "hicache_storage_prefetch_policy": "timeout", "hicache_storage_backend_extra_config": null, "enable_hisparse": false, "hisparse_config": null, "enable_broadcast_mm_inputs_process": false, "enable_prefix_mm_cache": false, "mm_enable_dp_encoder": false, "mm_process_config": {}, "mm_processor_worker_num": 0, "mm_io_worker_num": 0, "limit_mm_data_per_request": null, "enable_mm_global_cache": false, "disable_fast_image_processor": false, "mm_feature_transport": "cpu", "keep_mm_feature_on_device": false, "enable_lora": null, "enable_lora_overlap_loading": null, "max_lora_rank": null, "lora_target_modules": null, "lora_paths": null, "max_loaded_loras": null, "max_loras_per_batch": 8, "lora_eviction_policy": "lru", "lora_backend": "csgmv", "max_lora_chunk_size": 16, "experts_shared_outer_loras": null, "lora_use_virtual_experts": false, "lora_strict_loading": false, "lora_drain_wait_threshold": 0.0, "enable_two_batch_overlap": false, "enable_single_batch_overlap": false, "tbo_token_distribution_threshold": 0.48, "cpu_offload_gb": 0, "offload_group_size": -1, "offload_num_in_group": 1, "offload_prefetch_step": 1, "offload_mode": "cpu", "enable_lmcache": false, "lmcache_config_file": null, "enable_flexkv": false, "flexkv_config_file": null, "kt_weight_path": null, "kt_method": "AMXINT4", "kt_cpuinfer": null, "kt_threadpool_count": 2, "kt_num_gpu_experts": null, "kt_max_deferred_experts_per_token": null, "dllm_algorithm": null, "dllm_algorithm_config": null, "dllm_fdfo": true, "disaggregation_mode": "decode", "disaggregation_transfer_backend": "mooncake", "disaggregation_bootstrap_port": 28800, "disaggregation_ib_device": "mlx5_0,mlx5_1,mlx5_2,mlx5_3", "disaggregation_decode_enable_radix_cache": false, "disaggregation_decode_enable_offload_kvcache": false, "num_reserved_decode_tokens": 512, "disaggregation_decode_extra_slots": 0, "disaggregation_decode_polling_interval": 1, "optimistic_prefill_attempts": 0, "encoder_only": false, "language_only": false, "encoder_transfer_backend": "zmq_to_tokenizer", "encoder_urls": [], "encoder_bootstrap_port": 8997, "encoder_register_urls": [], "enable_adaptive_dispatch_to_encoder": false, "enable_pdmux": false, "pdmux_config_path": null, "sm_group_num": 8, "custom_weight_loader": [], "weight_loader_disable_mmap": false, "weight_loader_prefetch_checkpoints": false, "weight_loader_prefetch_num_threads": 4, "weight_loader_drop_cache_after_load": false, "remote_instance_weight_loader_seed_instance_ip": null, "remote_instance_weight_loader_seed_instance_service_port": null, "remote_instance_weight_loader_send_weights_group_ports": null, "remote_instance_weight_loader_backend": "nccl", "remote_instance_weight_loader_start_seed_via_transfer_engine": false, "engine_info_bootstrap_port": 6789, "modelexpress_config": null, "download_dir": null, "model_checksum": null, "delete_ckpt_after_loading": false, "decrypted_config_file": null, "decrypted_draft_config_file": null, "checkpoint_engine_wait_weights_before_ready": false, "enable_prefill_delayer": false, "prefill_delayer_max_delay_passes": 30, "prefill_delayer_token_usage_low_watermark": null, "prefill_delayer_forward_passes_buckets": null, "prefill_delayer_wait_seconds_buckets": null, "prefill_delayer_queue_min_ratio": null, "prefill_delayer_max_delay_ms": null, "min_free_slots_delay": null, "enable_deterministic_inference": false, "rl_on_policy_target": null, "kv_canary": "none", "kv_canary_real_data": "none", "kv_canary_sweep_interval": 0, "enable_dynamic_batch_tokenizer": false, "dynamic_batch_tokenizer_batch_size": 32, "dynamic_batch_tokenizer_batch_timeout": 0.002, "enable_tokenizer_batch_encode": false, "disable_tokenizer_batch_decode": false, "debug_tensor_dump_output_folder": null, "debug_tensor_dump_layers": null, "debug_tensor_dump_input_file": null, "enable_memory_saver": false, "enable_weights_cpu_backup": false, "enable_draft_weights_cpu_backup": false, "enable_custom_logit_processor": false, "enable_return_hidden_states": false, "enable_return_routed_experts": false, "enable_return_indexer_topk": false, "disable_outlines_disk_cache": false, "enable_mis": false, "weight_cache_mode": "off", "weight_cache_socket": null, "weight_cache_timeout": 1800, "forward_hooks": null, "msprobe_dump_config": null, "status": "ready", "max_total_num_tokens": 440256, "max_req_input_len": 440250, "internal_states": [{"model_path": "/data/hf_models/Kimi-K3", "tokenizer_path": "/data/hf_models/Kimi-K3", "tokenizer_mode": "auto", "tokenizer_backend": "huggingface", "tokenizer_worker_num": 1, "detokenizer_worker_num": 1, "skip_tokenizer_init": false, "load_format": "auto", "model_loader_extra_config": "{}", "trust_remote_code": true, "context_length": null, "is_embedding": false, "enable_multimodal": null, "revision": null, "model_impl": "auto", "model_config_parser": "auto", "json_model_override_args": "{}", "dtype": "auto", "quantization": null, "quantization_param_path": null, "kv_cache_dtype": "auto", "enable_fp32_lm_head": false, "modelopt_quant": null, "modelopt_checkpoint_restore_path": null, "modelopt_checkpoint_save_path": null, "modelopt_export_path": null, "quantize_and_serve": false, "rl_quant_profile": null, "enable_tf32_matmul": false, "mem_fraction_static": 0.88, "max_running_requests": null, "max_queued_requests": null, "max_total_tokens": null, "chunked_prefill_size": 8192, "enable_dynamic_chunking": false, "max_prefill_tokens": 16384, "prefill_max_requests": null, "schedule_policy": "fcfs", "enable_priority_scheduling": false, "disable_priority_preemption": false, "default_priority_value": null, "abort_on_priority_when_disabled": false, "schedule_low_priority_values_first": false, "priority_scheduling_preemption_threshold": 10, "retraction_policy": "length", "schedule_conservativeness": 1.0, "page_size": 64, "swa_full_tokens_ratio": 0.8, "disable_hybrid_swa_memory": false, "radix_eviction_policy": "lru", "prefill_only_disable_kv_cache": false, "disable_radix_cache": true, "enable_page_major_kv_layout": false, "enable_unified_memory": false, "disable_chunked_prefix_cache": true, "disable_overlap_schedule": false, "num_continuous_decode_steps": 1, "scheduler_recv_interval": 1, "enable_mixed_chunk": false, "nccl_port": null, "dist_timeout": 3600, "dist_init_addr": "174.1.60.5:20000", "nnodes": 4, "node_rank": 0, "tp_size": 32, "dcp_size": 1, "pp_size": 1, "pp_max_micro_batch_size": 181, "pp_async_batch_depth": 0, "dp_size": 1, "load_balance_method": "round_robin", "attn_cp_size": 1, "moe_dp_size": 1, "dcp_comm_backend": "ag_rs", "dcp_replicate_q_proj": null, "dwdp_size": 1, "enable_prefill_cp": false, "cp_strategy": null, "enable_dsa_cache_layer_split": false, "enable_dsa_prefill_context_parallel": false, "dsa_prefill_cp_mode": "round-robin-split", "enable_prefill_context_parallel": false, "prefill_cp_mode": "in-seq-split", "enable_cp_decode_attn_tp": false, "enable_dp_attention": false, "enable_dp_attention_local_control_broadcast": false, "enable_dp_lm_head": false, "enable_attn_tp_input_scattered": false, "disable_attn_tp_gather": false, "enable_p2p_check": false, "device": "cuda", "base_gpu_id": 0, "gpu_id_step": 1, "random_seed": 369235288, "watchdog_timeout": 300, "soft_watchdog_timeout": null, "sleep_on_idle": false, "use_ray": false, "numa_node": null, "gc_threshold": null, "host": "0.0.0.0", "port": 30000, "fastapi_root_path": "", "smg_grpc_mode": false, "grpc_mode": false, "grpc_port": null, "sidecar": null, "sidecar_args": null, "skip_server_warmup": false, "warmups": null, "enable_http2": false, "ssl_keyfile": null, "ssl_certfile": null, "ssl_ca_certs": null, "ssl_keyfile_password": null, "enable_ssl_refresh": false, "api_key": null, "admin_api_key": null, "served_model_name": "kimi-k3", "weight_version": "default", "chat_template": null, "hf_chat_template_name": null, "completion_template": null, "file_storage_path": "sglang_storage", "enable_cache_report": false, "reasoning_parser": null, "default_chat_template_kwargs": null, "strip_thinking_cache": false, "enable_strict_thinking": false, "tool_call_parser": null, "tool_server": null, "sampling_defaults": "model", "asr_max_buffer_seconds": 60, "asr_max_concurrent_sessions": 32, "preferred_sampling_params": null, "allow_auto_truncate": false, "stream_interval": 1, "batch_notify_size": 16, "stream_response_default_include_usage": false, "incremental_streaming_output": false, "enable_streaming_session": false, "enable_session_radix_cache": false, "log_level": "info", "log_level_http": null, "log_requests": false, "log_requests_level": 2, "log_requests_format": "text", "log_requests_target": null, "uvicorn_access_log_exclude_prefixes": [], "crash_dump_folder": null, "show_time_cost": false, "enable_metrics": false, "smg_http_sidecar_port": null, "enable_mfu_metrics": false, "enable_metrics_for_all_schedulers": false, "load_snapshot_publish_interval": 15, "tokenizer_metrics_custom_labels_header": "x-custom-labels", "tokenizer_metrics_allowed_custom_labels": null, "extra_metric_labels": null, "bucket_time_to_first_token": null, "bucket_inter_token_latency": null, "bucket_e2e_request_latency": null, "prompt_tokens_buckets": null, "generation_tokens_buckets": null, "gc_warning_threshold_secs": 0.0, "decode_log_interval": 40, "enable_request_time_stats_logging": false, "kv_events_config": null, "enable_forward_pass_metrics": false, "forward_pass_metrics_worker_id": "", "forward_pass_metrics_ipc_name": null, "enable_trace": false, "trace_modules": "request", "otlp_traces_endpoint": "localhost:4317", "export_metrics_to_file": false, "export_metrics_to_file_dir": null, "stat_loggers": null, "constrained_json_whitespace_pattern": null, "constrained_json_disable_any_whitespace": false, "attention_backend": "triton", "decode_attention_backend": null, "prefill_attention_backend": null, "sampling_backend": "flashinfer", "grammar_backend": "xgrammar", "radix_cache_backend": null, "mm_attention_backend": null, "fp8_gemm_runner_backend": "auto", "fp4_gemm_runner_backend": "auto", "bf16_gemm_backend": "auto", "dsa_prefill_backend": null, "dsa_decode_backend": null, "dsa_paged_mqa_logits_backend": "auto", "dsa_topk_backend": "sgl-kernel", "disable_flashinfer_autotune": false, "flashinfer_autotune_skip_ops": null, "mamba_backend": "triton", "cuda_graph_config": {"decode": {"backend": "full", "max_bs": 16, "bs": [1, 2, 4, 8, 12, 16], "tc_compiler": "eager", "full_prefill_max_req": null}, "prefill": {"backend": "disabled", "max_bs": 2048, "bs": [4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], "tc_compiler": "eager", "full_prefill_max_req": null}}, "cuda_graph_backend_decode": null, "cuda_graph_backend_prefill": null, "cuda_graph_max_bs_decode": 16, "cuda_graph_max_bs_prefill": null, "cuda_graph_bs_decode": null, "cuda_graph_bs_prefill": null, "cuda_graph_tc_compiler": null, "disable_prefill_cuda_graph": false, "disable_decode_cuda_graph": false, "disable_cuda_graph": false, "disable_cuda_graph_padding": false, "enable_profile_cuda_graph": false, "enable_cudagraph_gc": false, "debug_cuda_graph": false, "enable_layerwise_nvtx_marker": false, "enable_nccl_nvls": false, "enable_symm_mem": false, "triton_attention_reduce_in_fp32": false, "triton_attention_num_kv_splits": 8, "triton_attention_split_tile_size": null, "flashinfer_mla_disable_ragged": false, "enable_fused_qk_norm_rope": false, "enable_precise_embedding_interpolation": false, "enable_fused_moe_sum_all_reduce": false, "enable_deepseek_v4_fp4_indexer": false, "disable_custom_all_reduce": false, "enable_mscclpp": false, "enable_torch_symm_mem": false, "enable_scattered_sconv": false, "pre_warm_nccl": false, "enable_quant_communications": false, "enable_flashinfer_allreduce_fusion": false, "enforce_disable_flashinfer_allreduce_fusion": false, "flashinfer_allreduce_fusion_backend": null, "enable_aiter_allreduce_fusion": false, "enable_torch_compile": false, "enable_torch_compile_debug_mode": false, "torch_compile_max_bs": 32, "torchao_config": "", "speculative_algorithm": null, "speculative_draft_model_path": null, "speculative_draft_model_revision": null, "speculative_draft_load_format": null, "speculative_num_steps": null, "speculative_eagle_topk": null, "speculative_num_draft_tokens": null, "speculative_dflash_block_size": null, "speculative_dspark_block_size": null, "speculative_dspark_sps_table_path": null, "speculative_dspark_confidence_sts_path": null, "speculative_dspark_align_verify_tokens_to_graph_tier": false, "speculative_accept_threshold_single": 1.0, "speculative_accept_threshold_acc": 1.0, "speculative_use_rejection_sampling": false, "speculative_token_map": null, "speculative_attention_mode": "prefill", "speculative_draft_attention_backend": null, "speculative_draft_window_size": null, "speculative_moe_runner_backend": "marlin", "speculative_moe_a2a_backend": null, "speculative_draft_model_quantization": null, "speculative_skip_dp_mlp_sync": false, "enable_multi_layer_eagle": false, "speculative_adaptive": false, "speculative_adaptive_config": null, "decoupled_spec_bind_endpoint": null, "decoupled_spec_connect_endpoints": null, "decoupled_spec_rank": null, "decoupled_spec_role": "null", "spec_trace_dir": null, "speculative_ngram_min_bfs_breadth": 1, "speculative_ngram_max_bfs_breadth": 10, "speculative_ngram_match_type": "BFS", "speculative_ngram_max_trie_depth": 18, "speculative_ngram_capacity": 10000000, "speculative_ngram_external_corpus_path": null, "speculative_ngram_external_sam_budget": 0, "speculative_ngram_external_corpus_max_tokens": 10000000, "ep_size": 32, "moe_a2a_backend": "none", "moe_runner_backend": "marlin", "flashinfer_mxfp4_moe_precision": "default", "deepep_mode": "auto", "fuseep_mode": 2, "deepep_dispatcher_output_dtype": "auto", "ep_num_redundant_experts": 0, "ep_dispatch_algorithm": null, "init_expert_location": "trivial", "enable_eplb": false, "eplb_algorithm": "auto", "eplb_rebalance_num_iterations": 1000, "eplb_rebalance_layers_per_chunk": null, "eplb_min_rebalancing_utilization_threshold": 1.0, "expert_distribution_recorder_mode": null, "expert_distribution_recorder_buffer_size": 1000, "enable_expert_distribution_metrics": false, "deepep_config": null, "moe_dense_tp_size": null, "elastic_ep_backend": null, "enable_elastic_expert_backup": false, "mooncake_ib_device": null, "enable_waterfill": false, "ep_join_mode": null, "ep_join_rank_offset": 0, "elastic_ep_initial_size": null, "max_ep_size": null, "elastic_ep_scale_timeout": 600, "elastic_ep_rejoin": false, "disable_flashinfer_cutlass_moe_fp4_allgather": false, "disable_shared_experts_fusion": false, "enforce_shared_experts_fusion": false, "max_mamba_cache_size": 181, "mamba_ssm_dtype": null, "mamba_max_states_per_path": -1, "enable_mamba_cache_stochastic_rounding": false, "mamba_cache_philox_rounds": 0, "mamba_full_memory_ratio": 0.21, "mamba_radix_cache_strategy": "extra_buffer_lazy", "uses_mamba_radix_cache": false, "mamba_track_interval": 256, "enable_int8_mamba_checkpoint": false, "int8_mamba_ckpt_size": null, "linear_attn_backend": "triton", "linear_attn_decode_backend": null, "linear_attn_prefill_backend": null, "linear_attn_verify_backend": null, "enable_linear_replayssm": false, "linear_replayssm_cache_len": 16, "enable_linear_replayssm_spec": false, "enable_hierarchical_cache": false, "hicache_ratio": 2.0, "hicache_size": 0, "hicache_write_policy": "write_through", "hicache_io_backend": "kernel", "hicache_mem_layout": "page_first", "hicache_storage_backend": null, "hicache_storage_prefetch_policy": "timeout", "hicache_storage_backend_extra_config": null, "enable_hisparse": false, "hisparse_config": null, "enable_broadcast_mm_inputs_process": false, "enable_prefix_mm_cache": false, "mm_enable_dp_encoder": false, "mm_process_config": {}, "mm_processor_worker_num": 0, "mm_io_worker_num": 0, "limit_mm_data_per_request": null, "enable_mm_global_cache": false, "disable_fast_image_processor": false, "mm_feature_transport": "cpu", "keep_mm_feature_on_device": false, "enable_lora": null, "enable_lora_overlap_loading": null, "max_lora_rank": null, "lora_target_modules": null, "lora_paths": null, "max_loaded_loras": null, "max_loras_per_batch": 8, "lora_eviction_policy": "lru", "lora_backend": "csgmv", "max_lora_chunk_size": 16, "experts_shared_outer_loras": null, "lora_use_virtual_experts": false, "lora_strict_loading": false, "lora_drain_wait_threshold": 0.0, "enable_two_batch_overlap": false, "enable_single_batch_overlap": false, "tbo_token_distribution_threshold": 0.48, "cpu_offload_gb": 0, "offload_group_size": -1, "offload_num_in_group": 1, "offload_prefetch_step": 1, "offload_mode": "cpu", "enable_lmcache": false, "lmcache_config_file": null, "enable_flexkv": false, "flexkv_config_file": null, "kt_weight_path": null, "kt_method": "AMXINT4", "kt_cpuinfer": null, "kt_threadpool_count": 2, "kt_num_gpu_experts": null, "kt_max_deferred_experts_per_token": null, "dllm_algorithm": null, "dllm_algorithm_config": null, "dllm_fdfo": true, "disaggregation_mode": "decode", "disaggregation_transfer_backend": "mooncake", "disaggregation_bootstrap_port": 28800, "disaggregation_ib_device": "mlx5_0,mlx5_1,mlx5_2,mlx5_3", "disaggregation_decode_enable_radix_cache": false, "disaggregation_decode_enable_offload_kvcache": false, "num_reserved_decode_tokens": 512, "disaggregation_decode_extra_slots": 0, "disaggregation_decode_polling_interval": 1, "optimistic_prefill_attempts": 0, "encoder_only": false, "language_only": false, "encoder_transfer_backend": "zmq_to_tokenizer", "encoder_urls": [], "encoder_bootstrap_port": 8997, "encoder_register_urls": [], "enable_adaptive_dispatch_to_encoder": false, "enable_pdmux": false, "pdmux_config_path": null, "sm_group_num": 8, "custom_weight_loader": [], "weight_loader_disable_mmap": false, "weight_loader_prefetch_checkpoints": false, "weight_loader_prefetch_num_threads": 4, "weight_loader_drop_cache_after_load": false, "remote_instance_weight_loader_seed_instance_ip": null, "remote_instance_weight_loader_seed_instance_service_port": null, "remote_instance_weight_loader_send_weights_group_ports": null, "remote_instance_weight_loader_backend": "nccl", "remote_instance_weight_loader_start_seed_via_transfer_engine": false, "engine_info_bootstrap_port": 6789, "modelexpress_config": null, "download_dir": null, "model_checksum": null, "delete_ckpt_after_loading": false, "decrypted_config_file": null, "decrypted_draft_config_file": null, "checkpoint_engine_wait_weights_before_ready": false, "enable_prefill_delayer": false, "prefill_delayer_max_delay_passes": 30, "prefill_delayer_token_usage_low_watermark": null, "prefill_delayer_forward_passes_buckets": null, "prefill_delayer_wait_seconds_buckets": null, "prefill_delayer_queue_min_ratio": null, "prefill_delayer_max_delay_ms": null, "min_free_slots_delay": null, "enable_deterministic_inference": false, "rl_on_policy_target": null, "kv_canary": "none", "kv_canary_real_data": "none", "kv_canary_sweep_interval": 0, "enable_dynamic_batch_tokenizer": false, "dynamic_batch_tokenizer_batch_size": 32, "dynamic_batch_tokenizer_batch_timeout": 0.002, "enable_tokenizer_batch_encode": false, "disable_tokenizer_batch_decode": false, "debug_tensor_dump_output_folder": null, "debug_tensor_dump_layers": null, "debug_tensor_dump_input_file": null, "enable_memory_saver": false, "enable_weights_cpu_backup": false, "enable_draft_weights_cpu_backup": false, "enable_custom_logit_processor": false, "enable_return_hidden_states": false, "enable_return_routed_experts": false, "enable_return_indexer_topk": false, "disable_outlines_disk_cache": false, "enable_mis": false, "weight_cache_mode": "off", "weight_cache_socket": null, "weight_cache_timeout": 1800, "forward_hooks": null, "msprobe_dump_config": null, "_resolved_overrides": [["_sampling_backend_default", {"sampling_backend": "flashinfer"}], ["_attention_backend_default", {"attention_backend": "triton"}], ["_data_parallelism_defaults", {"enable_dp_attention": false, "enable_dp_lm_head": false}], ["_speculative_moe_runner_default", {"speculative_moe_runner_backend": "marlin"}]], "grpc_worker_threads": 4, "_quantization_explicitly_unset": false, "_cuda_graph_config_locked": [["decode", "max_bs"]], "_declarations_materialized": true, "_runtime_mutations": [["model_runner.chunked_prefix_cache_gate", {"disable_chunked_prefix_cache": true}], ["mamba_pool.memory_budget", {"max_mamba_cache_size": 181}]], "_in_override": false, "_mx_config_cache": {}, "max_speculative_num_draft_tokens": null, "last_gen_throughput": 48.2592441209455, "memory_usage": {"weight": 58.57, "kvcache": 11.34, "token_capacity": 440256, "graph": 0.6}, "effective_max_running_requests_per_dp": 181}], "version": "0.5.16", "kv_events": null}]}, "duration": 257.582706164103, "completed": 40, "total_input_tokens": 655360, "total_input_text_tokens": 655360, "total_input_vision_tokens": 0, "total_output_tokens": 20480, "total_output_tokens_retokenized": 20467, "request_throughput": 0.15528992841047512, "input_throughput": 2544.2701870772244, "output_throughput": 79.50844334616326, "total_throughput": 2623.7786304233873, "mean_e2e_latency_ms": 47570.72458656621, "median_e2e_latency_ms": 51368.52010944858, "std_e2e_latency_ms": 8285.295136840656, "p90_e2e_latency_ms": 54690.13653679285, "p95_e2e_latency_ms": 56342.666492424905, "p99_e2e_latency_ms": 56439.381064786576, "mean_ttft_ms": 17104.554165247828, "median_ttft_ms": 18754.675634438172, "std_ttft_ms": 5004.340325496838, "p90_ttft_ms": 21811.55602070503, "p95_ttft_ms": 23181.66090096347, "p99_ttft_ms": 23331.115312571637, "mean_tpot_ms": 59.62068575600467, "median_tpot_ms": 63.195876082325256, "std_tpot_ms": 7.394110008150321, "p90_tpot_ms": 66.46822736299654, "p95_tpot_ms": 66.5476718326072, "p99_tpot_ms": 66.7804115926464, "mean_itl_ms": 59.717039065885054, "median_itl_ms": 64.61505009792745, "std_itl_ms": 10.00643470443197, "p90_itl_ms": 67.3073009122163, "p95_itl_ms": 69.48161544278264, "p99_itl_ms": 70.99116549361497, "concurrency": 7.387254415482295, "accept_length": null, "max_output_tokens_per_s": 128.0, "max_concurrent_requests": 13, "input_lens": [16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384], "output_lens": [512, 512, 512, 512, 512, 512, 512, 512, 512, 512, 512, 512, 512, 512, 512, 512, 512, 512, 512, 512, 512, 512, 512, 512, 512, 512, 512, 512, 512, 512, 512, 512, 512, 512, 512, 512, 512, 512, 512, 512], "ttfts": [14.64687219192274, 16.564496570033953, 18.298474935116246, 19.87542626983486, 20.614703704835847, 20.610898945014924, 20.67768107308075, 20.59513547900133, 9.775098412064835, 17.973148954100907, 18.818385819904506, 18.68196496204473, 21.133975910954177, 21.195403756108135, 21.15704563516192, 21.154398849001154, 7.709285725839436, 7.7133990249130875, 13.800879123155028, 21.10020899097435, 21.80463355500251, 23.268064484000206, 23.177113343961537, 23.371426498051733, 18.690965448971838, 11.579551477218047, 7.740074336994439, 7.747202174039558, 10.888869322836399, 20.014686041977257, 20.670319276861846, 21.8738582120277, 21.365312944166362, 19.96640943782404, 18.370060433866456, 12.858996473019943, 7.714939045952633, 11.18145534908399, 15.81500570406206, 13.986338716931641], "itls": [[0.040845107985660434, 0.040183834033086896, 0.0398549388628453, 0.039899663999676704, 0.03982460009865463, 0.039983403868973255, 0.040057809092104435, 0.03985685296356678, 0.03982349601574242, 0.03978567011654377, 0.039899451890960336, 0.03984042792581022, 0.039930541068315506, 0.03989121108315885, 0.03985703689977527, 0.0397656240966171, 0.03985107084736228, 0.04004426207393408, 0.04187800013460219, 0.039281368954107165, 0.03977725701406598, 0.040212071035057306, 0.03997956891544163, 0.03986830194480717, 0.03980333497747779, 0.03983982000499964, 0.039774584118276834, 0.03990201698616147, 0.039864397840574384, 0.03973031113855541, 0.039741582004353404, 0.03976752096787095, 0.039800581987947226, 0.0398338430095464, 0.03975323308259249, 0.03990055690519512, 0.039711954072117805, 0.04010096192359924, 0.04006033204495907, 0.03964369185268879, 0.03980391612276435, 0.04107579309493303, 0.20355705078691244, 0.039795226184651256, 0.04627524199895561, 0.0440539310220629, 0.0438022178132087, 0.04410447203554213, 0.04376112716272473, 0.04374827281571925, 0.0437465391587466, 0.04395621782168746, 0.04384080716408789, 0.04409342282451689, 0.043764914153143764, 0.04396215290762484, 0.044511192943900824, 0.04466991499066353, 0.043791214004158974, 0.04422111017629504, 0.043631708016619086, 0.043868261855095625, 0.043719717068597674, 0.04387079388834536, 0.04364808415994048, 0.043825503904372454, 0.044050229946151376, 0.04391428013332188, 0.043903050012886524, 0.04374063992872834, 0.043759508058428764, 0.043676062021404505, 0.043750904966145754, 0.0437681470066309, 0.04382818494923413, 0.04396291705779731, 0.04418168589472771, 0.044835043139755726, 0.19522147183306515, 0.04681411501951516, 0.05171125615015626, 0.05091699887998402, 0.05069550103507936, 0.0513237330596894, 0.05078040389344096, 0.05092244897969067, 0.050971137126907706, 0.050769499968737364, 0.05185486190021038, 0.05032689613290131, 0.05099174682982266, 0.05660002422519028, 0.045710010919719934, 0.05102255288511515, 0.05075378017500043, 0.050703421933576465, 0.050929060904309154, 0.05071582295931876, 0.050982589134946465, 0.05080440500751138, 0.05068004596978426, 0.05087511497549713, 0.05096764396876097, 0.05018901498988271, 0.0508456421084702, 0.050545471953228116, 0.05115674901753664, 0.05051489290781319, 0.05194948706775904, 0.054510371992364526, 0.04864287003874779, 0.05069826706312597, 0.05055634886957705, 0.05083969607949257, 0.05094286589883268, 0.050534808076918125, 0.05129895987920463, 0.050686988048255444, 0.05209041899070144, 0.05134362308308482, 0.2077766580041498, 0.07552284887060523, 0.0704422271810472, 0.07155090197920799, 0.07326677301898599, 0.07208359404467046, 0.07030183984898031, 0.07012850814498961, 0.0713950409553945, 0.07083909399807453, 0.07022962905466557, 0.07042941986583173, 0.07033255393616855, 0.07110600103624165, 0.07079600915312767, 0.07043985486961901, 0.06980050913989544, 0.07082231598906219, 0.07040712283924222, 0.07092684600502253, 0.0701467371545732, 0.07103004097007215, 0.07028592494316399, 0.07098962087184191, 0.07174199121072888, 0.07126273796893656, 0.07137430598959327, 0.07045084703713655, 0.07075169286690652, 0.07094220700673759, 0.07074561691842973, 0.0701374071650207, 0.07017179601825774, 0.07083517801947892, 0.07098543690517545, 0.07063362095504999, 0.07015946111641824, 0.07102706097066402, 0.07152164285071194, 0.07089847605675459, 0.07018725504167378, 0.0702636269852519, 0.07089983206242323, 0.07045831996947527, 0.07023222791031003, 0.07015638402663171, 0.07047030003741384, 0.07144869095645845, 0.07026047515682876, 0.07025162898935378, 0.07101702084764838, 0.0709902320522815, 0.07058189995586872, 0.07064133509993553, 0.0708994590677321, 0.07039081794209778, 0.06998711288906634, 0.0697753201238811, 0.0693676508963108, 0.07000313210301101, 0.06967868888750672, 0.06946328398771584, 0.0698302979581058, 0.06986359017901123, 0.06968386098742485, 0.06964746094308794, 0.06987892091274261, 0.0697979861870408, 0.07055598590523005, 0.07032728288322687, 0.06909689516760409, 0.06957929604686797, 0.06945536285638809, 0.0702136280015111, 0.06984014413319528, 0.07017991482280195, 0.0692841790150851, 0.07021644897758961, 0.06977175106294453, 0.07000144897028804, 0.06993598002009094, 0.06993261608295143, 0.06948832888156176, 0.06990648410283029, 0.0696564270183444, 0.06960327993147075, 0.0697993200737983, 0.07006004103459418, 0.06996808690018952, 0.0701950560323894, 0.06951577495783567, 0.0699248961172998, 0.0700146418530494, 0.06929207104258239, 0.06991985999047756, 0.06967497593723238, 0.07014615298248827, 0.07026117015630007, 0.06990722683258355, 0.07062241504900157, 0.06977670104242861, 0.07015504804439843, 0.06910112802870572, 0.0698830799665302, 0.06993522308766842, 0.06991828000172973, 0.06945076398551464, 0.07050055684521794, 0.06977363605983555, 0.06991028506308794, 0.07014458603225648, 0.06954263290390372, 0.06962596904486418, 0.06983697786927223, 0.06962970714084804, 0.06992104183882475, 0.07007552706636488, 0.06986824492923915, 0.06994608417153358, 0.06968488497659564, 0.06984057696536183, 0.06950763310305774, 0.06896763597615063, 0.0688950028270483, 0.06886459398083389, 0.0694511781912297, 0.06850644992664456, 0.06855996209196746, 0.0686699419748038, 0.06913216481916606, 0.0689796400256455, 0.06916136504150927, 0.0688267070800066, 0.06906311796046793, 0.06907671294175088, 0.06925253896042705, 0.06856040307320654, 0.06829942995682359, 0.06876491312868893, 0.06871200283057988, 0.06877467315644026, 0.06909602787345648, 0.06867817509919405, 0.06833098502829671, 0.06781225395388901, 0.06787658901885152, 0.06787174008786678, 0.06761214090511203, 0.0680003350134939, 0.0678829550743103, 0.06833106791600585, 0.06831603986211121, 0.06756028719246387, 0.06791866291314363, 0.06784685095772147, 0.06754013313911855, 0.06799664697609842, 0.06798115093261003, 0.0676636379212141, 0.0683019810821861, 0.06828547595068812, 0.06690599815919995, 0.06890698382630944, 0.0679285479709506, 0.06747230608016253, 0.06797470501624048, 0.06795123289339244, 0.06768554216250777, 0.06767657399177551, 0.06794113805517554, 0.06771180988289416, 0.06770705501548946, 0.06763970199972391, 0.06755313510075212, 0.06800495181232691, 0.06765814800746739, 0.06817242410033941, 0.06764412089250982, 0.06786438007839024, 0.06762342411093414, 0.06746895797550678, 0.06806815601885319, 0.06765227997675538, 0.0675770128145814, 0.0681559550575912, 0.06820389302447438, 0.06738938996568322, 0.0681314340326935, 0.06796257803216577, 0.06766965799033642, 0.06786330905742943, 0.06750558293424547, 0.06726349401287735, 0.0676859999075532, 0.06791967013850808, 0.068053905852139, 0.06907133618369699, 0.06755280494689941, 0.06814193190075457, 0.06788346916437149, 0.06738643487915397, 0.0677523729391396, 0.06752847810275853, 0.06931959907524288, 0.0657346029765904, 0.06749413185752928, 0.06776508409529924, 0.06724818190559745, 0.06753940205089748, 0.06734127411618829, 0.0677510928362608, 0.06728229811415076, 0.06732688588090241, 0.06740097305737436, 0.06746624410152435, 0.06742823286913335, 0.06731030205264688, 0.06724609108641744, 0.067723847925663, 0.0674367758911103, 0.06725138216279447, 0.06729814992286265, 0.0669436699245125, 0.06738778599537909, 0.06736657512374222, 0.06722005200572312, 0.06718737306073308, 0.06711676786653697, 0.06710959901101887, 0.06772956601344049, 0.06698079500347376, 0.06703148689121008, 0.06736251409165561, 0.06748009007424116, 0.06759581691585481, 0.06706335698254406, 0.066944481106475, 0.06731841992586851, 0.06731980107724667, 0.06771056703291833, 0.06737753096967936, 0.06707975990138948, 0.06738774501718581, 0.06703163590282202, 0.0673079201951623, 0.06761722080409527, 0.06730217020958662, 0.06709848600439727, 0.0670619998127222, 0.06740892608650029, 0.06746347807347775, 0.06692831288091838, 0.0656227171421051, 0.06592035992071033, 0.06541421497240663, 0.06520909909158945, 0.0649426868185401, 0.06547655910253525, 0.06518907193094492, 0.0652884989976883, 0.06526584108360112, 0.06491211988031864, 0.06502334005199373, 0.06486715003848076, 0.06517262803390622, 0.06527726096101105, 0.06521520391106606, 0.06498985015787184, 0.06476294598542154, 0.06510954303666949, 0.06481830589473248, 0.06514763599261642, 0.06555963796563447, 0.06566066900268197, 0.06519838096573949, 0.06554101221263409, 0.06532833795063198, 0.06579288584180176, 0.06517889304086566, 0.06527559505775571, 0.06981584592722356, 0.06686440599150956, 0.06577501306310296, 0.06521331192925572, 0.06479140603914857, 0.0649844640865922, 0.06513514695689082, 0.06502168509177864, 0.06479439791291952, 0.06494060391560197, 0.06468660500831902, 0.0649070511572063, 0.06407913193106651, 0.06509977392852306, 0.06460191612131894, 0.06474206899292767, 0.0645396348554641, 0.0645627430640161, 0.06478523509576917, 0.06431941292248666, 0.06440930790267885, 0.06478694221004844, 0.06474156095646322, 0.06476772483438253, 0.06417173519730568, 0.06486594397574663, 0.06505139102227986, 0.06468405784107745, 0.06485116598196328, 0.06484443717636168, 0.06477996380999684, 0.06459414912387729, 0.06408120901323855, 0.06478804093785584, 0.06474991701543331, 0.0646348181180656, 0.0645066318102181, 0.06427004910074174, 0.0644869259558618, 0.06521507306024432, 0.06393751199357212, 0.06484961207024753, 0.06449780100956559, 0.0645505259744823, 0.06463209982030094, 0.06434735306538641, 0.06498431996442378, 0.06418120209127665, 0.06456407206133008, 0.06441808003000915, 0.06449977890588343, 0.06450553494505584, 0.06458571995608509, 0.06530387699604034, 0.06471825810149312, 0.06455005705356598, 0.0643668610136956, 0.06432291981764138, 0.06528059300035238, 0.06498976214788854, 0.06419600895605981, 0.06509795808233321, 0.06443515280261636, 0.06427874113433063, 0.06515609798952937, 0.06458888691850007, 0.06460663513280451, 0.06444449303671718, 0.06418240093626082, 0.06494385097175837, 0.06486366293393075, 0.06417738506570458, 0.06467007799074054, 0.06476301490329206, 0.06466812803409994, 0.06491840607486665, 0.0642635200638324, 0.06545577687211335, 0.06452652416191995, 0.06432923092506826, 0.06439644191414118, 0.06484281015582383, 0.06465211394242942, 0.06477949302643538, 0.06435513682663441, 0.06497800210490823, 0.06466219900175929, 0.06467105587944388, 0.06426432612352073, 0.06450027599930763, 0.06485692295245826, 0.06472806911915541, 0.06442430499009788, 0.06497175386175513, 0.06479609198868275, 0.06437092018313706, 0.06482963496819139, 0.06542033003643155, 0.06417418993078172, 0.06450403202325106, 0.06487039290368557, 0.06485359813086689, 0.06452917284332216, 0.06425871001556516, 0.06490814196877182, 0.06473751715384424, 0.0648844379466027, 0.06455793092027307, 0.06429335498251021, 0.06581384502351284], [0.04627441707998514, 0.04405109095387161, 0.0438023810274899, 0.044095342978835106, 0.04376923106610775, 0.04375136294402182, 0.04373865691013634, 0.04395660199224949, 0.043840943137183785, 0.04408781696110964, 0.043778205988928676, 0.043953478103503585, 0.04456081986427307, 0.04463029699400067, 0.04376787808723748, 0.04432153492234647, 0.04358836798928678, 0.04384567611850798, 0.04371530096977949, 0.04387377598322928, 0.0436463940422982, 0.043818457052111626, 0.04405577294528484, 0.04391326499171555, 0.04389691189862788, 0.04374195309355855, 0.043760736007243395, 0.04367739101871848, 0.043751563876867294, 0.043772025033831596, 0.04382607713341713, 0.04396683885715902, 0.04417729005217552, 0.04483816307038069, 0.1952189749572426, 0.04692570003680885, 0.05170316295698285, 0.05091803101822734, 0.05070018395781517, 0.05144667602144182, 0.050608983961865306, 0.05098004196770489, 0.05095706996507943, 0.05076203006319702, 0.05185538809746504, 0.050327896839007735, 0.05100030708126724, 0.05670804507099092, 0.04555334197357297, 0.051121569937095046, 0.05071908701211214, 0.050702007953077555, 0.05091658397577703, 0.050718397134914994, 0.0540324819739908, 0.04775481694377959, 0.050691210897639394, 0.050862487172707915, 0.050976225873455405, 0.05018517002463341, 0.050817859126254916, 0.050575605826452374, 0.051193681079894304, 0.05047971103340387, 0.05205633700825274, 0.05450999201275408, 0.048860745038837194, 0.050504261860623956, 0.05055355909280479, 0.05087800300680101, 0.05089651886373758, 0.050537727074697614, 0.05132103990763426, 0.05068656010553241, 0.05219145189039409, 0.05122866202145815, 0.20789623400196433, 0.0753198410384357, 0.07048329594545066, 0.07304564421065152, 0.07175319292582572, 0.07209133985452354, 0.07028869306668639, 0.07011365913785994, 0.07140851183794439, 0.07084042811766267, 0.07022826792672276, 0.07041459693573415, 0.07032925100065768, 0.07116212719120085, 0.07074158685281873, 0.07045095902867615, 0.06983594107441604, 0.0707203010097146, 0.0705223549157381, 0.07094135112129152, 0.07014764682389796, 0.07101796614006162, 0.07031140895560384, 0.07097515393979847, 0.07177156303077936, 0.07120620692148805, 0.07120061409659684, 0.07051741192117333, 0.07074171118438244, 0.0709364649374038, 0.07084665703587234, 0.07019188301637769, 0.07018660381436348, 0.07079549902118742, 0.0709242729935795, 0.07073247316293418, 0.07014966593123972, 0.07101126597262919, 0.07150763692334294, 0.07090045604854822, 0.07021107198670506, 0.07027771300636232, 0.07087933807633817, 0.07052559987641871, 0.07022133003920317, 0.07008743612095714, 0.07047655899077654, 0.07151062600314617, 0.07022875500842929, 0.07022890890948474, 0.07109611411578953, 0.07091922988183796, 0.07055884203873575, 0.07070321589708328, 0.07087435317225754, 0.07039782893843949, 0.07003171788528562, 0.06977032101713121, 0.06936781597323716, 0.07001283904537559, 0.06970463413745165, 0.06943631498143077, 0.06984479399397969, 0.06991459894925356, 0.06968873692676425, 0.06960368505679071, 0.06987358396872878, 0.06979206809774041, 0.07036201492883265, 0.07016796711832285, 0.06929690786637366, 0.06973349093459547, 0.06944042700342834, 0.07022783416323364, 0.0698122070170939, 0.07020979491062462, 0.06929452996701002, 0.07019826304167509, 0.06978563498705626, 0.06999368593096733, 0.06994933914393187, 0.06992651685141027, 0.06948945415206254, 0.0698253468144685, 0.06992372707463801, 0.06932759913615882, 0.06989563279785216, 0.0700587471947074, 0.06995817180722952, 0.07021396397612989, 0.06948555912822485, 0.06997352396138012, 0.06998714501969516, 0.06935022305697203, 0.06987936003133655, 0.06968167796730995, 0.07021879288367927, 0.07018246199004352, 0.06981200305745006, 0.07052176212891936, 0.06976013584062457, 0.07016208115965128, 0.06924591679126024, 0.06993759609758854, 0.06992798508144915, 0.06964791496284306, 0.06978147989138961, 0.07054011197760701, 0.06966370111331344, 0.0701524200849235, 0.06989311682991683, 0.06959487497806549, 0.06961518409661949, 0.06984489294700325, 0.06959544098936021, 0.06999566312879324, 0.0700363318901509, 0.06987138511613011, 0.06994415284134448, 0.06967949611134827, 0.06978287198580801, 0.07403050386346877, 0.06431240704841912, 0.06888334406539798, 0.06901288195513189, 0.06940935715101659, 0.06840720982290804, 0.06854359712451696, 0.06891525886021554, 0.06912698713131249, 0.0689865630120039, 0.0690665019210428, 0.06888615991920233, 0.06908120703883469, 0.06910032709129155, 0.06923313206061721, 0.06855545495636761, 0.06831942196004093, 0.06876409100368619, 0.06870507192797959, 0.06854177103377879, 0.06910062697716057, 0.06867321697063744, 0.06834094412624836, 0.06779956701211631, 0.06788593693636358, 0.06787889311090112, 0.06760317785665393, 0.06800899305380881, 0.06788192107342184, 0.0685238279402256, 0.06811976991593838, 0.06756317312829196, 0.06791506195440888, 0.06785066286101937, 0.06753975502215326, 0.06800223398022354, 0.06798053020611405, 0.06790145789273083, 0.06834499910473824, 0.06792210880666971, 0.06714228820055723, 0.06876782584004104, 0.06795654306188226, 0.06765865907073021, 0.06774895801208913, 0.06807983084581792, 0.06756217614747584, 0.06769770383834839, 0.06792660802602768, 0.06770225497893989, 0.06770732905715704, 0.06776741496287286, 0.06743264803662896, 0.06808759411796927, 0.06758459098637104, 0.06816039490513504, 0.06765742902643979, 0.0678578270599246, 0.06762099685147405, 0.06747036497108638, 0.06806795112788677, 0.06764206988736987, 0.0676729769911617, 0.06808186299167573, 0.06819696119055152, 0.06743076792918146, 0.06810489087365568, 0.06796312611550093, 0.06765470909886062, 0.06798910698853433, 0.06739025097340345, 0.06745086400769651, 0.06761086895130575, 0.06779536488465965, 0.06818721513263881, 0.06894807191565633, 0.06755003007128835, 0.06814071605913341, 0.0679692099802196, 0.06742978794500232, 0.0676453378982842, 0.06752170203253627, 0.06943317502737045, 0.06561252707615495, 0.0675184668507427, 0.06786696705967188, 0.06712469202466309, 0.067542142001912, 0.06734762201085687, 0.06774868094362319, 0.06741561507806182, 0.0672123939730227, 0.06738171610049903, 0.06758689391426742, 0.06731235701590776, 0.06731317797675729, 0.06738007394596934, 0.06771006202325225, 0.06734178308397532, 0.06722246389836073, 0.06730688805691898, 0.06695351493544877, 0.06737999594770372, 0.06763796508312225, 0.06711918907240033, 0.06717276503331959, 0.06711801700294018, 0.06710919085890055, 0.06772212614305317, 0.06699711899273098, 0.06701587699353695, 0.067216238938272, 0.06736105191521347, 0.0679196841083467, 0.06685115187428892, 0.06711067701689899, 0.06731414212845266, 0.06716732797212899, 0.06747906608507037, 0.06737709487788379, 0.0672131369356066, 0.06736322096548975, 0.06703078816644847, 0.06745575787499547, 0.06747943814843893, 0.06729774293489754, 0.06721251388080418, 0.06694852304644883, 0.06741292215883732, 0.06775343092158437, 0.06666266988031566, 0.06562225706875324, 0.06594666093587875, 0.06538081518374383, 0.06522673601284623, 0.06525289290584624, 0.06546470592729747, 0.06488398602232337, 0.06530961813405156, 0.06515847286209464, 0.06517090601846576, 0.06510292110033333, 0.06461503985337913, 0.06554108415730298, 0.06492496398277581, 0.06534560001455247, 0.0651148590259254, 0.06493475800380111, 0.06472060177475214, 0.06492840219289064, 0.06544041982851923, 0.06540264608338475, 0.065392489079386, 0.06550604081712663, 0.06520655704662204, 0.06532357004471123, 0.06607681512832642, 0.06508407299406826, 0.06539666187018156, 0.07007056102156639, 0.06653106701560318, 0.0657294481061399, 0.06529063894413412, 0.06469866400584579, 0.06512668589130044, 0.06483135302551091, 0.06502145691774786, 0.06501060212031007, 0.06498183589428663, 0.06445478298701346, 0.06477902201004326, 0.06433701515197754, 0.06512360600754619, 0.06466378597542644, 0.06465486204251647, 0.0643912588711828, 0.06468562106601894, 0.06467499304562807, 0.0646401229314506, 0.06435228209011257, 0.06466080201789737, 0.06472584884613752, 0.06483596214093268, 0.06411606399342418, 0.06498425081372261, 0.06516005913726985, 0.0648206970654428, 0.06466596387326717, 0.06465052207931876, 0.06480469484813511, 0.06455414020456374, 0.06441513285972178, 0.0645627761259675, 0.06504991394467652, 0.06425850698724389, 0.06486538192257285, 0.06390730408020318, 0.07130834995768964, 0.0584932470228523, 0.06404007994569838, 0.06470670900307596, 0.06453905301168561, 0.06464660493656993, 0.06477287504822016, 0.06403190013952553, 0.06502942787483335, 0.06426635012030602, 0.06478029885329306, 0.06405551498755813, 0.06484302808530629, 0.06420417688786983, 0.06492519308812916, 0.06496587698347867, 0.0647167800925672, 0.06451966683380306, 0.0643823491409421, 0.06443902989849448, 0.06557006412185729, 0.06481761485338211, 0.0643564430065453, 0.06485600001178682, 0.06461736001074314, 0.0638221250846982, 0.06530818389728665, 0.06468571396544576, 0.06463361717760563, 0.06464548385702074, 0.06418397603556514, 0.06470820307731628, 0.06476673088036478, 0.06434111413545907, 0.06448610802181065, 0.06483984785154462, 0.06506238412111998, 0.06445057084783912, 0.06450825207866728, 0.06544545502401888, 0.06429677503183484, 0.06432223785668612, 0.0644519291818142, 0.06494716182351112, 0.06468936405144632, 0.06470916816033423, 0.06423318781889975, 0.06513229198753834, 0.06455961801111698, 0.064637087052688, 0.06440162705257535, 0.06476105889305472, 0.06447463110089302, 0.06489639705978334, 0.06420535081997514, 0.06496119708754122, 0.06479503191076219, 0.06437595910392702, 0.06482652900740504, 0.06558321905322373, 0.06420809403061867, 0.06434895098209381, 0.0648461920209229, 0.06484935292974114, 0.06453018286265433, 0.06443275301717222, 0.06472884304821491, 0.06451855413615704, 0.06511498196050525, 0.06442959699779749, 0.06458862195722759, 0.06585375708527863, 0.07093965681269765, 0.059943479020148516, 0.06960933096706867, 0.06713703507557511, 0.06526493211276829, 0.06540354900062084, 0.0652113298419863, 0.06537377601489425, 0.06534966011531651, 0.06522553390823305, 0.0654965101275593, 0.06596631393767893, 0.06505809188820422, 0.06553447805345058, 0.0646083781030029, 0.06589359301142395, 0.06563127995468676, 0.065584217896685, 0.06529834307730198, 0.06535900896415114, 0.06577107799239457, 0.06582198711112142, 0.06554660992696881, 0.06515898508951068, 0.06552971596829593, 0.06563658500090241, 0.06528728688135743, 0.06557483412325382, 0.06542305182665586, 0.06533308420330286, 0.06575440079905093, 0.06574035808444023, 0.06543481699191034, 0.06535120704211295, 0.06530360202305019, 0.06570062204264104, 0.06531543587334454, 0.06538290600292385, 0.0655318561475724, 0.06555547681637108, 0.06539195217192173, 0.06592089682817459, 0.06524470704607666, 0.06629390991292894], [0.05170718510635197, 0.050917229847982526, 0.050667379051446915, 0.051437064073979855, 0.05066669196821749, 0.050930933095514774, 0.05096328281797469, 0.05079234717413783, 0.05185425584204495, 0.050301168113946915, 0.050994456047192216, 0.05670730001293123, 0.04558008490130305, 0.05111302505247295, 0.05072670290246606, 0.050701700150966644, 0.050890030805021524, 0.05074797407723963, 0.053940417943522334, 0.04784424602985382, 0.05065719597041607, 0.0508982059545815, 0.050941373221576214, 0.05019141477532685, 0.050848381128162146, 0.05053806398063898, 0.051196495071053505, 0.050479089841246605, 0.05205967300571501, 0.05451517319306731, 0.048592318780720234, 0.050676124170422554, 0.05056355404667556, 0.05086144199594855, 0.05091274087317288, 0.050542768090963364, 0.051308830035850406, 0.050698834005743265, 0.05215770099312067, 0.05127148795872927, 0.2078763018362224, 0.07483204104937613, 0.07044348097406328, 0.07158422819338739, 0.07330716890282929, 0.07210808200761676, 0.07025579595938325, 0.07006476703099906, 0.07141717406921089, 0.07087866798974574, 0.0701708400156349, 0.07048037694767118, 0.07034901296719909, 0.07105363090522587, 0.07085511111654341, 0.07035478088073432, 0.06989019620232284, 0.07049762294627726, 0.07064599101431668, 0.07077482482418418, 0.07030056114308536, 0.07101135794073343, 0.07029326702468097, 0.07097165193408728, 0.07174687110818923, 0.07117541087791324, 0.07119194907136261, 0.07047691289335489, 0.07076199003495276, 0.07094130502082407, 0.07077622110955417, 0.07032078690826893, 0.07020152197219431, 0.07087019202299416, 0.07100531412288547, 0.0704268419649452, 0.07030702196061611, 0.0710126010235399, 0.07151752803474665, 0.07089514704421163, 0.07021646178327501, 0.07025510002858937, 0.07098892400972545, 0.07031955709680915, 0.07019926887005568, 0.07024994515813887, 0.0704645998775959, 0.07138544600456953, 0.07033768808469176, 0.07026118389330804, 0.07090494618751109, 0.07110567600466311, 0.07051845081150532, 0.07056507212109864, 0.07103397906757891, 0.07043976685963571, 0.06983361905440688, 0.06978573091328144, 0.06938683311454952, 0.07002008403651416, 0.0697260529268533, 0.06940519390627742, 0.06984571111388505, 0.06982696987688541, 0.06968739302828908, 0.0696312349755317, 0.0698968330398202, 0.06978506105951965, 0.0703504669945687, 0.07020598603412509, 0.06928773387335241, 0.06971329520456493, 0.0694470398593694, 0.07021077908575535, 0.0698137788567692, 0.07022433797828853, 0.06930083804763854, 0.07020818116143346, 0.0697682248428464, 0.07000435097143054, 0.06992151914164424, 0.0699540248606354, 0.06947378604672849, 0.06985272909514606, 0.06974860304035246, 0.06948207295499742, 0.0698849898763001, 0.0700474800541997, 0.06998228002339602, 0.07020350289531052, 0.06947614206001163, 0.06994828605093062, 0.07000767509452999, 0.06938278977759182, 0.06983489007689059, 0.06968007702380419, 0.07013759203255177, 0.07031446904875338, 0.06979120802134275, 0.0705135848838836, 0.06980455899611115, 0.07013270491734147, 0.06923571601510048, 0.06990981218405068, 0.06996498885564506, 0.0696184269618243, 0.06976517708972096, 0.07045866292901337, 0.06980745098553598, 0.06985748116858304, 0.07010834687389433, 0.06961090699769557, 0.06960724503733218, 0.06986365607008338, 0.06956099905073643, 0.0700305518694222, 0.0699950021225959, 0.06988987396471202, 0.06995203089900315, 0.06967553705908358, 0.06980383093468845, 0.0693946301471442, 0.0689451249781996, 0.06890140380710363, 0.06900869403034449, 0.06927729398012161, 0.06855456298217177, 0.06857424601912498, 0.06883063819259405, 0.06912384182214737, 0.06896661617793143, 0.06910019181668758, 0.06890175212174654, 0.06905644806101918, 0.0691157488618046, 0.06924127414822578, 0.06852884986437857, 0.06832336401566863, 0.06875600595958531, 0.06870525213889778, 0.06861393596045673, 0.0691061110701412, 0.06866197986528277, 0.06836007302626967, 0.06778365490026772, 0.0679026641882956, 0.06783674890175462, 0.06765157007612288, 0.06800826685503125, 0.06786997197195888, 0.06833726004697382, 0.06832919013686478, 0.06754000694490969, 0.06791547103784978, 0.06785617489367723, 0.06750315893441439, 0.06802204414270818, 0.06798467691987753, 0.06763620907440782, 0.06800762005150318, 0.06846833182498813, 0.06713324505835772, 0.0687707751058042, 0.06795791885815561, 0.06762074213474989, 0.06785160093568265, 0.06801422592252493, 0.06758018909022212, 0.06770016206428409, 0.06795506295748055, 0.06769981887191534, 0.06770709506236017, 0.06768441805616021, 0.06750070792622864, 0.06804955611005425, 0.06762647698633373, 0.06816156604327261, 0.06765939388424158, 0.0678724991157651, 0.06762143201194704, 0.06748944986611605, 0.06804218492470682, 0.06764083215966821, 0.0677794509101659, 0.0679396209307015, 0.06820920505560935, 0.06852923100814223, 0.06700588506646454, 0.06796928495168686, 0.06766699207946658, 0.06794787500984967, 0.06739602494053543, 0.06737602991051972, 0.06769077898934484, 0.06783205713145435, 0.06813685991801322, 0.06897334195673466, 0.06754918210208416, 0.06815257202833891, 0.0679060819093138, 0.0674525210633874, 0.06766049400903285, 0.0675623849965632, 0.06934203393757343, 0.06564610404893756, 0.06753774196840823, 0.06782739097252488, 0.06713530514389277, 0.06756342784501612, 0.06741666607558727, 0.06769683491438627, 0.06737524317577481, 0.06722552981227636, 0.06739458511583507, 0.06771377706900239, 0.06719417800195515, 0.06729979999363422, 0.06733638886362314, 0.06761420401744545, 0.06743838591501117, 0.06723880511708558, 0.06730123097077012, 0.06695963512174785, 0.06742368498817086, 0.06742216483689845, 0.06718910811468959, 0.06721076695248485, 0.0671119058970362, 0.06709592207334936, 0.06775448098778725, 0.06696133897639811, 0.06703586410731077, 0.06721380795352161, 0.06737815705128014, 0.06794080790132284, 0.06679632002487779, 0.06710988795384765, 0.0673170811496675, 0.06722098682075739, 0.06748938607051969, 0.0673362510278821, 0.06720584910362959, 0.06743279681541026, 0.06700402405112982, 0.06739215599372983, 0.06748970411717892, 0.06732388189993799, 0.06723873293958604, 0.06694552418775856, 0.06735826889052987, 0.06741881603375077, 0.06683649495244026, 0.06582407699897885, 0.06576606701128185, 0.06547235092148185, 0.06503809220157564, 0.06508017494343221, 0.0654627438634634, 0.06515644514001906, 0.0652033039368689, 0.06518877390772104, 0.06520745507441461, 0.06493075308389962, 0.06492893188260496, 0.06536290398798883, 0.06483210297301412, 0.0654298982117325, 0.06490263878367841, 0.06494645518250763, 0.0648420809302479, 0.06490440596826375, 0.06533908797428012, 0.06536363600753248, 0.06561402510851622, 0.06549950200133026, 0.06544231483712792, 0.06545126112177968, 0.06553240888752043, 0.06517546810209751, 0.06540785287506878, 0.0699909960385412, 0.06663490110076964, 0.06568986899219453, 0.06528555508702993, 0.06479726289398968, 0.06499156798236072, 0.06529612210579216, 0.06494802981615067, 0.06455147499218583, 0.06497300020419061, 0.06466727098450065, 0.06482054595835507, 0.06451039086095989, 0.06477159913629293, 0.06479445192962885, 0.06466962001286447, 0.06468785600736737, 0.06460367608815432, 0.06443667481653392, 0.0645876310300082, 0.064258978003636, 0.06502646999433637, 0.06465203408151865, 0.06459250301122665, 0.06448335596360266, 0.06470015412196517, 0.06508263479918242, 0.06471472419798374, 0.06470817979425192, 0.06506965402513742, 0.0645747110247612, 0.0644308419432491, 0.06432710005901754, 0.06497387611307204, 0.06459590303711593, 0.06451971479691565, 0.06462167203426361, 0.06435047695413232, 0.06450074422173202, 0.06477653281763196, 0.06430005514994264, 0.0646076078992337, 0.06462234002538025, 0.06471046688966453, 0.06452972209081054, 0.06443993491120636, 0.06472399109043181, 0.06454456597566605, 0.06438519712537527, 0.06455658888444304, 0.06435142899863422, 0.06467238906770945, 0.06476724101230502, 0.06475848704576492, 0.06494039599783719, 0.06455357791855931, 0.06443701195530593, 0.0644403190817684, 0.06528321583755314, 0.06480504013597965, 0.06436242908239365, 0.06483102194033563, 0.06473477696999907, 0.06449104705825448, 0.06454619695432484, 0.0646722549572587, 0.06461545499041677, 0.06462066108360887, 0.06434272602200508, 0.06471556099131703, 0.06472143391147256, 0.0645430451259017, 0.06450652191415429, 0.06476952694356441, 0.06461245007812977, 0.06487525906413794, 0.0644762518350035, 0.06518250307999551, 0.06447550398297608, 0.06457862909883261, 0.06438687583431602, 0.0647143330425024, 0.06456620409153402, 0.06472727097570896, 0.06463405885733664, 0.06463523698039353, 0.0646494550164789, 0.06482837302610278, 0.06420151307247579, 0.06467767199501395, 0.06481383508071303, 0.06491845892742276, 0.0643036849796772, 0.0650005480274558, 0.06480055209249258, 0.0643677047919482, 0.06480426411144435, 0.06471533607691526, 0.06499459780752659, 0.06436238810420036, 0.06490121805109084, 0.06480668485164642, 0.06454313616268337, 0.06464740890078247, 0.06453031301498413, 0.06493346695788205, 0.0646895479876548, 0.06451857811771333, 0.06444824603386223, 0.06571253086440265, 0.07120962510816753, 0.06004571099765599, 0.06963030784390867, 0.06713516800664365, 0.0652510451618582, 0.06546384003013372, 0.06517206784337759, 0.06539583811536431, 0.06532725691795349, 0.06523754494264722, 0.06544828414916992, 0.06584033090621233, 0.06518975412473083, 0.0653115229215473, 0.06480900500901043, 0.06594860507175326, 0.06560598500072956, 0.06561204884201288, 0.0652754120528698, 0.06536102900281549, 0.06579490494914353, 0.0658026800956577, 0.06547783804126084, 0.06527685490436852, 0.06550192506983876, 0.0653400479350239, 0.06548430398106575, 0.06561361812055111, 0.0654557989910245, 0.06534462794661522, 0.06575463386252522, 0.06572665506973863, 0.06543653993867338, 0.0653562811203301, 0.06532928999513388, 0.0656486488878727, 0.0653536671306938, 0.06536605698056519, 0.06540994299575686, 0.06568948808126152, 0.06538554094731808, 0.06588752707466483, 0.06525025190785527, 0.06601351406425238, 0.06855547102168202, 0.06945489696227014, 0.06669010082259774, 0.06488937418907881, 0.06439049984328449, 0.06468337401747704, 0.06463073613122106, 0.06445614900439978, 0.06497406703419983, 0.06410517590120435, 0.06482975790277123, 0.0645100031979382, 0.06453405693173409, 0.06442549801431596, 0.06469935295172036, 0.06450099498033524, 0.06451688706874847, 0.06464373995549977, 0.06428306899033487, 0.06465922389179468, 0.064438849221915, 0.06506919395178556, 0.06405968498438597, 0.06470921798609197, 0.06455783289857209, 0.06432213610969484, 0.06444179592654109, 0.06464336812496185, 0.0647972219157964, 0.06453900900669396, 0.06457979092374444, 0.06458041816949844, 0.06465546786785126, 0.06452258094213903, 0.06431941012851894, 0.06561301904730499], [0.05066829198040068, 0.05055314605124295, 0.05088121094740927, 0.05089593701995909, 0.05053849192336202, 0.05131625710055232, 0.05069364793598652, 0.05214551696553826, 0.05127341509796679, 0.2078680768609047, 0.0752847061958164, 0.07053992594592273, 0.0729684829711914, 0.07182674994692206, 0.0720889160875231, 0.07028468092903495, 0.07011617813259363, 0.07141154794953763, 0.0708431068342179, 0.07022545300424099, 0.07041893317364156, 0.0703255299013108, 0.07109294109977782, 0.07081839186139405, 0.07044535502791405, 0.06983473594300449, 0.0707067612092942, 0.07054104981943965, 0.07093971106223762, 0.07014271710067987, 0.07102010282687843, 0.07030965317972004, 0.07097406196407974, 0.07176205888390541, 0.07111039804294705, 0.07130593410693109, 0.07051870599389076, 0.07074464485049248, 0.07093526003882289, 0.07074965792708099, 0.07027670019306242, 0.07018484780564904, 0.07080960320308805, 0.0710798988584429, 0.07056940998882055, 0.07015382405370474, 0.07100026891566813, 0.0715224090963602, 0.07089534006081522, 0.07020469103008509, 0.07027971395291388, 0.0708827399648726, 0.07052685203962028, 0.07021941291168332, 0.0700918089132756, 0.07046732306480408, 0.0715207711327821, 0.07020222186110914, 0.07024978706613183, 0.07109785103239119, 0.07091715605929494, 0.07055741385556757, 0.07070006709545851, 0.07087620184756815, 0.0703852609731257, 0.0700405091047287, 0.06977166910655797, 0.06936838780529797, 0.070019057020545, 0.06969884596765041, 0.0694327971432358, 0.06985441991128027, 0.06983572198078036, 0.06968458415940404, 0.06967445486225188, 0.0698712591547519, 0.06979591585695744, 0.0702780659776181, 0.07015759707428515, 0.06938976608216763, 0.06973136798478663, 0.06944215996190906, 0.07022522087208927, 0.06971567519940436, 0.07030523382127285, 0.06929700612090528, 0.07019809796474874, 0.06978578492999077, 0.0699930430855602, 0.06995076895691454, 0.06992454594001174, 0.06948614702560008, 0.0698264620732516, 0.06971556390635669, 0.06954063707962632, 0.0698948500212282, 0.0700536088552326, 0.0699607441201806, 0.07020772388204932, 0.06939310510642827, 0.07006753608584404, 0.06998699181713164, 0.06934335618279874, 0.06988444202579558, 0.06968006677925587, 0.07014121301472187, 0.07025893917307258, 0.06982216285541654, 0.07050586608238518, 0.06977947894483805, 0.07016172795556486, 0.06924108508974314, 0.06993035692721605, 0.0699334959499538, 0.06945866113528609, 0.06988737196661532, 0.07051611109636724, 0.06967025785706937, 0.07021736912429333, 0.06992358388379216, 0.06959681911394, 0.06961688282899559, 0.06984547013416886, 0.06949416105635464, 0.07008573482744396, 0.07004351308569312, 0.06987212295643985, 0.06994391116313636, 0.06967773078940809, 0.0697877760976553, 0.06935237697325647, 0.0689938310533762, 0.06888016988523304, 0.06901156203821301, 0.06937610311433673, 0.06844134884886444, 0.0685458299703896, 0.06890745903365314, 0.06913252198137343, 0.06897611916065216, 0.06906708190217614, 0.0688918880186975, 0.06908340705558658, 0.06909976992756128, 0.06923758098855615, 0.06853620312176645, 0.0683339738752693, 0.06876802700571716, 0.06867536902427673, 0.06857186206616461, 0.06909980205819011, 0.06867420487105846, 0.06833919696509838, 0.06780137005262077, 0.06787914107553661, 0.06788287893868983, 0.06760466587729752, 0.06800464121624827, 0.06788154994137585, 0.06832393305376172, 0.06832173187285662, 0.06756280804984272, 0.0679161089938134, 0.06784876598976552, 0.06753897690214217, 0.0679995461832732, 0.0679814638569951, 0.0676677580922842, 0.06853063497692347, 0.06796384393237531, 0.06713864696212113, 0.0687756179831922, 0.0679485690779984, 0.06765853310935199, 0.0677596228197217, 0.06807380705140531, 0.06756458897143602, 0.06767989997752011, 0.06794550502672791, 0.06770478398539126, 0.06770877912640572, 0.0677574840374291, 0.067436687881127, 0.06800297112204134, 0.0676688109524548, 0.06816221703775227, 0.06764672393910587, 0.06786799198016524, 0.06762006995268166, 0.06747249816544354, 0.06806694692932069, 0.06764739798381925, 0.06758250389248133, 0.06816351599991322, 0.0681992620229721, 0.06730677001178265, 0.06823074398562312, 0.06795915798284113, 0.06765672913752496, 0.06798414885997772, 0.06739091104827821, 0.06736516603268683, 0.06769457110203803, 0.06780059286393225, 0.06818119413219392, 0.06894812383688986, 0.06755410903133452, 0.06814388209022582, 0.06788369594141841, 0.06750995595939457, 0.06763946614228189, 0.06752563500776887, 0.06943422579206526, 0.06561185419559479, 0.06749951280653477, 0.06788513506762683, 0.06712707108817995, 0.06753788003697991, 0.06730475998483598, 0.0677930738311261, 0.06741068908013403, 0.06721760192885995, 0.06737898313440382, 0.067580278031528, 0.06731791398487985, 0.06731348088942468, 0.06737361312843859, 0.06770522682927549, 0.06735041900537908, 0.06722346297465265, 0.0673101912252605, 0.06694714678451419, 0.06738229817710817, 0.06759480596520007, 0.06714910501614213, 0.06718443497084081, 0.06711271987296641, 0.06711128004826605, 0.0677257920615375, 0.06697808392345905, 0.06703663198277354, 0.06720870197750628, 0.06736699515022337, 0.06789513095282018, 0.06687462492845953, 0.06710374704562128, 0.06732074497267604, 0.06716812495142221, 0.06734015815891325, 0.0673765679821372, 0.06733072595670819, 0.06738849193789065, 0.06703063216991723, 0.06743855681270361, 0.06748974602669477, 0.06729889707639813, 0.06721441796980798, 0.06694874097593129, 0.06740920408628881, 0.06770428991876543, 0.06670236494392157, 0.06561758904717863, 0.06594813009724021, 0.06538581987842917, 0.06521641905419528, 0.06521812989376485, 0.06530109909363091, 0.0650850769598037, 0.06529929512180388, 0.06545329699292779, 0.06477344804443419, 0.06502710282802582, 0.06480655516497791, 0.0653103708755225, 0.06514606205746531, 0.06522090500220656, 0.06522897793911397, 0.0648630119394511, 0.06477609509602189, 0.06484327907674015, 0.0654775919392705, 0.06529899802990258, 0.06556637887842953, 0.06546979513950646, 0.06525242002680898, 0.06532574188895524, 0.06605959311127663, 0.06509493594057858, 0.06535997404716909, 0.06996739283204079, 0.06654043705202639, 0.06577653414569795, 0.06523828394711018, 0.06482868385501206, 0.06500225909985602, 0.06496621388942003, 0.06501714908517897, 0.06494039902463555, 0.06491103209555149, 0.06458830181509256, 0.064487983006984, 0.06462027598172426, 0.06497060810215771, 0.064691819017753, 0.06472726701758802, 0.06446053786203265, 0.06455374509096146, 0.06479601794853806, 0.06450761505402625, 0.06447688606567681, 0.0646737408824265, 0.06473435112275183, 0.06482143793255091, 0.06412723893299699, 0.06497241416946054, 0.06503188982605934, 0.06481269397772849, 0.06479379604570568, 0.06467073713429272, 0.0647996838670224, 0.06408783211372793, 0.06487098196521401, 0.06453105388209224, 0.06494573317468166, 0.06440656096674502, 0.06473930086940527, 0.0640361460391432, 0.06477606296539307, 0.0650001100730151, 0.06396283488720655, 0.06479148310609162, 0.06450328510254622, 0.06456907885149121, 0.06481466395780444, 0.06411412917077541, 0.06502160895615816, 0.06414383300580084, 0.06476209592074156, 0.06420779298059642, 0.06468453397974372, 0.06433522701263428, 0.06488115503452718, 0.06502502295188606, 0.06470988411456347, 0.06453633005730808, 0.06437480100430548, 0.0643194408621639, 0.06544722313992679, 0.06500878091901541, 0.06437675794586539, 0.06475196499377489, 0.06463334895670414, 0.06465179310180247, 0.06452081399038434, 0.0646635100711137, 0.06460291682742536, 0.06470371317118406, 0.06414920184761286, 0.06469483813270926, 0.06487090000882745, 0.0643390379846096, 0.0645007430575788, 0.06478388584218919, 0.06495876703411341, 0.06461074599064887, 0.06442511710338295, 0.06550515885464847, 0.06431794911623001, 0.06431619194336236, 0.06439476506784558, 0.0649796100333333, 0.0645755969453603, 0.0648456490598619, 0.06424389383755624, 0.06511607603169978, 0.06456985510885715, 0.06463901302777231, 0.06439686985686421, 0.06457613804377615, 0.06465546507388353, 0.06490526883862913, 0.06421219906769693, 0.06496631191112101, 0.06479185004718602, 0.06437741499394178, 0.06482790014706552, 0.06553248385898769, 0.06424375902861357, 0.06434397911652923, 0.06485914182849228, 0.06485293805599213, 0.06452865712344646, 0.06442972784861922, 0.06473440607078373, 0.06455743196420372, 0.06507171411067247, 0.06474187690764666, 0.0642734388820827, 0.06582615408115089, 0.07059214706532657, 0.060306556057184935, 0.06961040291935205, 0.06713858200237155, 0.06511577893979847, 0.06555443000979722, 0.0652067600749433, 0.06537427101284266, 0.06535517401061952, 0.0652316550258547, 0.06549063487909734, 0.06592460512183607, 0.06508666090667248, 0.06551463901996613, 0.06460107187740505, 0.06592528708279133, 0.0656352280639112, 0.06557797687128186, 0.06530506815761328, 0.06509944796562195, 0.06603235495276749, 0.06581951607950032, 0.06552319694310427, 0.06517422897741199, 0.06552976998500526, 0.06533066602423787, 0.06558369193226099, 0.06559226801618934, 0.06542010116390884, 0.06533924699760973, 0.06575070996768773, 0.06573162600398064, 0.06544877891428769, 0.06524352799169719, 0.06540991202928126, 0.06559464195743203, 0.06541844294406474, 0.06538035813719034, 0.06537823402322829, 0.06560432794503868, 0.06548579293303192, 0.06592871202155948, 0.06513690296560526, 0.06639179517515004, 0.06796721694990993, 0.06974438205361366, 0.06662528379820287, 0.06500258599407971, 0.06428309320472181, 0.06468127481639385, 0.06464060908183455, 0.06442455807700753, 0.0649988017976284, 0.06406425917521119, 0.06482546497136354, 0.06450558104552329, 0.06458373786881566, 0.06442552199587226, 0.06470624706707895, 0.06445776694454253, 0.06451962213031948, 0.06474001193419099, 0.06421179906465113, 0.06465776893310249, 0.06411385908722878, 0.06537692481651902, 0.06408111611381173, 0.06469513685442507, 0.06456744601018727, 0.06416329718194902, 0.06459032581187785, 0.06468462105840445, 0.06470853206701577, 0.06452603009529412, 0.06457058782689273, 0.06466797809116542, 0.06479845498688519, 0.06426947098225355, 0.06425931910052896, 0.06573366583324969, 0.06655492610298097, 0.06965063884854317, 0.06607828405685723, 0.06388756399974227, 0.06427581398747861, 0.06391242798417807, 0.06364357518032193, 0.06372309289872646, 0.06368084391579032, 0.06378161604516208, 0.06390762911178172, 0.0639599550049752, 0.06365489982999861, 0.06398679921403527, 0.06426115799695253, 0.06327530299313366, 0.06372122280299664, 0.06370042311027646, 0.06360690388828516, 0.06433088518679142, 0.0647062489297241, 0.06340062501840293, 0.06379608693532646, 0.06398123595863581, 0.06381423096172512, 0.06403510621748865, 0.06367749581113458, 0.06351068406365812, 0.06436442490667105, 0.06379903620108962, 0.06481131189502776], [0.07054624101147056, 0.07274595787748694, 0.07196154003031552, 0.07217035396024585, 0.07021534605883062, 0.07012063404545188, 0.0714020449668169, 0.07092480198480189, 0.07013770402409136, 0.07050357107073069, 0.07026008400134742, 0.07116180681623518, 0.07081347610801458, 0.07037108088843524, 0.06990916421636939, 0.07071259897202253, 0.07053602999076247, 0.07093680300749838, 0.07014046888798475, 0.07102339412085712, 0.07031051791273057, 0.07084329007193446, 0.07174786482937634, 0.07125665503554046, 0.07120236800983548, 0.0706231330987066, 0.07074803905561566, 0.07093840185552835, 0.07074831309728324, 0.07014258299022913, 0.07017436600290239, 0.07093775505200028, 0.07088567200116813, 0.0707677488680929, 0.07015319005586207, 0.07088898005895317, 0.07163612893782556, 0.07078841095790267, 0.07019508793018758, 0.07039108616299927, 0.0708838109858334, 0.07053524791263044, 0.07021790510043502, 0.07008533389307559, 0.07046727603301406, 0.07146382401697338, 0.0701437450479716, 0.07036053901538253, 0.07111025787889957, 0.07090824493207037, 0.07055874005891383, 0.07060764799825847, 0.07096287212334573, 0.0702709190081805, 0.07004856993444264, 0.06977381906472147, 0.06936427298933268, 0.07013460202142596, 0.06948399590328336, 0.06953163095749915, 0.0697647831402719, 0.06993106380105019, 0.0696785501204431, 0.06964704510755837, 0.06989207980223, 0.06972562219016254, 0.07036632997915149, 0.07025319593958557, 0.06917979894205928, 0.06983393710106611, 0.06945010600611567, 0.07014582096599042, 0.06990487803705037, 0.07018811884336174, 0.06929986597970128, 0.0701986940111965, 0.06978263705968857, 0.06999182305298746, 0.06994864600710571, 0.06991964997723699, 0.06948656192980707, 0.06983175897039473, 0.06998178409412503, 0.0692726019769907, 0.06989377993158996, 0.07005179719999433, 0.0698944958858192, 0.07026889291591942, 0.06952046207152307, 0.06994076492264867, 0.06992189306765795, 0.06938358396291733, 0.06983348517678678, 0.06974576879292727, 0.07015819614753127, 0.07024779985658824, 0.06981911114417017, 0.07063326984643936, 0.0697791981510818, 0.0701596848666668, 0.06923744804225862, 0.06982404994778335, 0.0699271741323173, 0.06948503386229277, 0.06983193405903876, 0.07065611402504146, 0.06967944093048573, 0.07002507499419153, 0.07010587607510388, 0.06947486801072955, 0.06953740189783275, 0.06984003819525242, 0.06971082091331482, 0.06984786107204854, 0.07006714190356433, 0.06986626796424389, 0.06994849303737283, 0.06976519012823701, 0.06990490900352597, 0.06935817189514637, 0.06898594088852406, 0.06888186908327043, 0.06900836410932243, 0.06935556093230844, 0.06846231292001903, 0.06855516997165978, 0.06867346703074872, 0.06913274014368653, 0.06897631892934442, 0.06928987195715308, 0.068695736117661, 0.0690672118216753, 0.06909395102411509, 0.06923912209458649, 0.06855710386298597, 0.06830056314356625, 0.06876778998412192, 0.0687952300067991, 0.06868334487080574, 0.06909981020726264, 0.0686775769572705, 0.06833507283590734, 0.06780472700484097, 0.06787726003676653, 0.06788015901111066, 0.06761069712229073, 0.06799825001507998, 0.06788118090480566, 0.06832910189405084, 0.06831999402493238, 0.06756128603592515, 0.06791755114682019, 0.06784783280454576, 0.06754029798321426, 0.06799741322174668, 0.06798289087601006, 0.06766270194202662, 0.06834347499534488, 0.0680077790748328, 0.06727363099344075, 0.06878148205578327, 0.06793054100126028, 0.06756607186980546, 0.06787549611181021, 0.0680686270352453, 0.06756703997962177, 0.06767635489813983, 0.0679450009483844, 0.0677098510786891, 0.06770746712572873, 0.06775357201695442, 0.06743902899324894, 0.06800142978318036, 0.067667008144781, 0.06816690694540739, 0.06764410901814699, 0.067865771939978, 0.06762547907419503, 0.06746862409636378, 0.06806797091849148, 0.06764933886006474, 0.06758453301154077, 0.06815700512379408, 0.06820153607986867, 0.06813653884455562, 0.0673811580054462, 0.0679634299594909, 0.06766887800768018, 0.06797965010628104, 0.06739139393903315, 0.06737053394317627, 0.06768787815235555, 0.06780938594602048, 0.06817025900818408, 0.06895344704389572, 0.06755733001045883, 0.06814218289218843, 0.06788218300789595, 0.0675078360363841, 0.0676349860150367, 0.06752818799577653, 0.06943504908122122, 0.06561541091650724, 0.06749621988274157, 0.06788225914351642, 0.06713287089951336, 0.06753696617670357, 0.06732304184697568, 0.0677735151257366, 0.0674027418717742, 0.06722300313413143, 0.06738183298148215, 0.06746917893178761, 0.06742600095458329, 0.06731305201537907, 0.0673648260999471, 0.06760322581976652, 0.0674577341414988, 0.0672282250598073, 0.06730352598242462, 0.06694861082360148, 0.06738406117074192, 0.06756593589670956, 0.06716633099131286, 0.06719026411883533, 0.06711641582660377, 0.06710819900035858, 0.06772512919269502, 0.06698244786821306, 0.06703368504531682, 0.0672141860704869, 0.0672183558344841, 0.06802990101277828, 0.06689038197509944, 0.06709260400384665, 0.06732233799993992, 0.0671731021720916, 0.06733160302974284, 0.06738009979017079, 0.06718801218084991, 0.06739161186851561, 0.06702883914113045, 0.06729611987248063, 0.06762876105494797, 0.0673064230941236, 0.0673583538737148, 0.06679648905992508, 0.06740741594694555, 0.06777736311778426, 0.06659633293747902, 0.06562913185916841, 0.06591844605281949, 0.06541489600203931, 0.06519856513477862, 0.06535692396573722, 0.06494191195815802, 0.06531960191205144, 0.06527796317823231, 0.06515265791676939, 0.0650345899630338, 0.06496802694164217, 0.06492089410312474, 0.06547438190318644, 0.06495296605862677, 0.06567394593730569, 0.0644035639706999, 0.06532426318153739, 0.06459911586716771, 0.06490379408933222, 0.06557941995561123, 0.06499473005533218, 0.06571893603540957, 0.06566997687332332, 0.06515429401770234, 0.06533338013105094, 0.06620082398876548, 0.06510673183947802, 0.06534365308471024, 0.06972880708053708, 0.06650012382306159, 0.06577687710523605, 0.0651584449224174, 0.06520432000979781, 0.06461745011620224, 0.0651902889367193, 0.06501662987284362, 0.0646517991553992, 0.06496806186623871, 0.06470649898983538, 0.0646887521725148, 0.064391401829198, 0.06508622015826404, 0.06460299203172326, 0.064705470809713, 0.06458974815905094, 0.06456155702471733, 0.0647687369491905, 0.06465852586552501, 0.064478405052796, 0.06471053906716406, 0.06474028597585857, 0.06480559706687927, 0.06414069794118404, 0.06493112002499402, 0.06498798285610974, 0.0647797950077802, 0.06486710021272302, 0.06471996894106269, 0.06479021604172885, 0.0638190358877182, 0.06512606702744961, 0.06451686494983733, 0.06485215411521494, 0.06453089183196425, 0.06461459305137396, 0.06416146014817059, 0.0644976207986474, 0.06525041512213647, 0.0638977219350636, 0.06487297592684627, 0.06450254004448652, 0.06452253716997802, 0.06474467879161239, 0.06423362111672759, 0.06501126009970903, 0.06415474484674633, 0.06465533701702952, 0.0643218350596726, 0.06456719688139856, 0.06444458616897464, 0.06484550493769348, 0.06505306391045451, 0.06471577403135598, 0.06454259995371103, 0.0643668370321393, 0.06432587094604969, 0.06533777411095798, 0.06494377297349274, 0.06444629491306841, 0.06484897015616298, 0.06442566495388746, 0.06523573910817504, 0.06381138577125967, 0.0649932671803981, 0.06460435083135962, 0.06469095405191183, 0.06411719811148942, 0.06474751397036016, 0.06486389902420342, 0.064176085870713, 0.06466716318391263, 0.0647658808156848, 0.06486100214533508, 0.06472648796625435, 0.06427317298948765, 0.06557786604389548, 0.06439727894030511, 0.0643210590351373, 0.06439746194519103, 0.06494801794178784, 0.06455733301118016, 0.06485270103439689, 0.06427794205956161, 0.06509979604743421, 0.06457458692602813, 0.06464168801903725, 0.06438635103404522, 0.06446443195454776, 0.06478120991960168, 0.06490212702192366, 0.06422516098245978, 0.06496705510653555, 0.06479891901835799, 0.06437586294487119, 0.06482459604740143, 0.06548977294005454, 0.06427514902316034, 0.0643421330023557, 0.06486740405671299, 0.0648560558911413, 0.06452631601132452, 0.06442293711006641, 0.06474417098797858, 0.064644246827811, 0.06498004007153213, 0.06457599392160773, 0.06443008501082659, 0.06569178402423859, 0.07069622003473341, 0.06032985309138894, 0.06948054791428149, 0.06713014305569232, 0.06557574891485274, 0.06511534401215613, 0.06520433397963643, 0.06537294713780284, 0.06535789091140032, 0.06535770697519183, 0.06548237707465887, 0.06590136094018817, 0.06498937495052814, 0.06559211015701294, 0.06451323302462697, 0.0659319928381592, 0.0656198610085994, 0.065585324075073, 0.0653090849518776, 0.06554931495338678, 0.06571836210787296, 0.06580298789776862, 0.06552126398310065, 0.06505867908708751, 0.06565202306956053, 0.06533983279950917, 0.0654042661190033, 0.06576926494017243, 0.06528201210312545, 0.06547898892313242, 0.06575116398744285, 0.06572159705683589, 0.06545957596972585, 0.06524714501574636, 0.06528194295242429, 0.06572076003067195, 0.06527673709206283, 0.0655162779148668, 0.06537877605296671, 0.06545717804692686, 0.06550477282144129, 0.0659254021011293, 0.06527225999161601, 0.06638232804834843, 0.06795555399730802, 0.0696507329121232, 0.06663228594698012, 0.06508139800280333, 0.06430303817614913, 0.06457614293321967, 0.06464337487705052, 0.06442730012349784, 0.06500302907079458, 0.06406555301509798, 0.0648242118768394, 0.06450982694514096, 0.06457610009238124, 0.06441598990932107, 0.06481265998445451, 0.06445790105499327, 0.06451757997274399, 0.06472943094559014, 0.06411586422473192, 0.06466496689245105, 0.0648728848900646, 0.06462256517261267, 0.06408480298705399, 0.0647901059128344, 0.06446835002861917, 0.06424143700860441, 0.06450833706185222, 0.06469017104245722, 0.06480603688396513, 0.06452545407228172, 0.0645666760392487, 0.06467175297439098, 0.06477717193774879, 0.06432961206883192, 0.06428252789191902, 0.0657347859814763, 0.0667729580309242, 0.06969820708036423, 0.06602315092459321, 0.06388770695775747, 0.06435392703860998, 0.06390080112032592, 0.06364846578799188, 0.06390595599077642, 0.06356727122329175, 0.0637799478136003, 0.06392357405275106, 0.06386646092869341, 0.06366016110405326, 0.06406598002649844, 0.06417496385984123, 0.06335888709872961, 0.06364163593389094, 0.06377256312407553, 0.06353343091905117, 0.0644024449866265, 0.0646654290612787, 0.06337040802463889, 0.0639755439478904, 0.06380842695944011, 0.0638153669424355, 0.06409608013927937, 0.063258261885494, 0.06386925606057048, 0.06436234712600708, 0.06380942277610302, 0.06484263320453465, 0.07722666091285646, 0.04759409697726369, 0.048567956080660224, 0.04830363392829895, 0.047006078995764256, 0.04723920510150492, 0.04775703093037009, 0.047048581996932626, 0.04735039593651891, 0.047540300991386175, 0.04853901197202504], [0.07047525001689792, 0.07342292205430567, 0.07148644095286727, 0.07210800889879465, 0.07028695708140731, 0.06998926610685885, 0.0714095369912684, 0.07096046186052263, 0.07010800507850945, 0.07047781790606678, 0.07036547898314893, 0.07106932112947106, 0.07084746891632676, 0.07033576094545424, 0.0699450180400163, 0.07069351710379124, 0.0704446230083704, 0.07094012387096882, 0.07015406410209835, 0.07101281802169979, 0.0703110930044204, 0.07099716598168015, 0.07176053407602012, 0.07120703486725688, 0.07119142496958375, 0.07051954395137727, 0.07081735203973949, 0.07093398901633918, 0.07077896501868963, 0.0701893090736121, 0.07018808485008776, 0.07095923204906285, 0.07097299816086888, 0.07051632786169648, 0.07014733599498868, 0.07101600896567106, 0.07150191511027515, 0.07090303907170892, 0.07022093795239925, 0.07027111784555018, 0.07088199607096612, 0.07051865896210074, 0.07022065110504627, 0.07009018398821354, 0.07047752500511706, 0.0715085060801357, 0.0702344779856503, 0.07022575801238418, 0.07109143398702145, 0.07092255284078419, 0.07056053401902318, 0.07070389902219176, 0.07087681698612869, 0.07040561595931649, 0.07002438511699438, 0.06977401184849441, 0.06938483007252216, 0.06999275204725564, 0.06970265205018222, 0.06944180885329843, 0.06984157301485538, 0.06975746108219028, 0.06968820095062256, 0.06976396404206753, 0.06987599190324545, 0.06979086506180465, 0.07036593114025891, 0.070159827824682, 0.06930110417306423, 0.06973248580470681, 0.06944348313845694, 0.0702266888692975, 0.06981616304256022, 0.07020473410375416, 0.0692959139123559, 0.0701988700311631, 0.06978580891154706, 0.06999512412585318, 0.06994559988379478, 0.06992802815511823, 0.06949429190717638, 0.06982494308613241, 0.06968914694152772, 0.0695600719191134, 0.06989518506452441, 0.07006056001409888, 0.06995962699875236, 0.07021445990540087, 0.0694948451127857, 0.06996337487362325, 0.06998818600550294, 0.06935977912507951, 0.06987481284886599, 0.06967972312122583, 0.07003480894491076, 0.0703678319696337, 0.06981195113621652, 0.07051728898659348, 0.06975848996080458, 0.0701788179576397, 0.06922927615232766, 0.06994158891029656, 0.06992611195892096, 0.06965777394361794, 0.06977037712931633, 0.07055978989228606, 0.06965665798634291, 0.0701515560504049, 0.0698857179377228, 0.06959921005181968, 0.06963334511965513, 0.06989210401661694, 0.06952928192913532, 0.07000004197470844, 0.070030408911407, 0.06987255113199353, 0.06994456588290632, 0.06967629306018353, 0.06978290295228362, 0.07423541904427111, 0.06410726299509406, 0.06888427794910967, 0.06901596509851515, 0.06942556099966168, 0.06838938686996698, 0.06867591617628932, 0.06878701294772327, 0.06912116892635822, 0.06899272301234305, 0.06906397803686559, 0.06889077601954341, 0.06907537207007408, 0.0691022127866745, 0.06922815018333495, 0.06856268201954663, 0.06831883080303669, 0.06876364606432617, 0.06870293407700956, 0.06861495389603078, 0.06909585511311889, 0.06868131388910115, 0.06833725818432868, 0.06780080799944699, 0.06788450898602605, 0.0678813538979739, 0.06759980111382902, 0.06801067478954792, 0.06789026712067425, 0.06823780108243227, 0.068402208853513, 0.0675672700162977, 0.06791019812226295, 0.06784340599551797, 0.06747738597914577, 0.06807378283701837, 0.06797987408936024, 0.0680736219510436, 0.06812833715230227, 0.06789837079122663, 0.06714143301360309, 0.06883630901575089, 0.06795423198491335, 0.06759739317931235, 0.06780628091655672, 0.0680143719073385, 0.06755850999616086, 0.06776852812618017, 0.06794631807133555, 0.0677724399138242, 0.067705123918131, 0.06770211504772305, 0.0674995209556073, 0.06803134502843022, 0.06764504802413285, 0.06816222704946995, 0.06755753699690104, 0.06795300985686481, 0.06752767413854599, 0.06757282698526978, 0.06805960601195693, 0.06764128291979432, 0.06762136309407651, 0.06814292189665139, 0.06819106498733163, 0.06756741809658706, 0.0678795799612999, 0.06805145391263068, 0.06767419399693608, 0.06792264804244041, 0.06744207814335823, 0.06740146386437118, 0.06749929115176201, 0.06786049786023796, 0.06812978000380099, 0.06900675897486508, 0.06754713016562164, 0.06814248580485582, 0.06791379302740097, 0.06742321117781103, 0.06781094591133296, 0.06742515997029841, 0.06936610210686922, 0.06568083981983364, 0.06751450709998608, 0.06780428602360189, 0.06719122501090169, 0.0676428850274533, 0.06716701784171164, 0.06782061816193163, 0.06736208498477936, 0.06727526895701885, 0.06737246597185731, 0.06753662996925414, 0.06736778910271823, 0.06742361886426806, 0.06720968405716121, 0.06771799502894282, 0.06751776300370693, 0.06710695195943117, 0.06741577899083495, 0.06683945609256625, 0.06737681082449853, 0.0675985449925065, 0.06712297606281936, 0.067148108035326, 0.06712246197275817, 0.06710710097104311, 0.06772064510732889, 0.06700260494835675, 0.0670089980121702, 0.06721834489144385, 0.06736409105360508, 0.06792423711158335, 0.06684496905654669, 0.06711429893039167, 0.06731133302673697, 0.06716264598071575, 0.06748714484274387, 0.06747422809712589, 0.0671139641199261, 0.06744385277852416, 0.0669476231560111, 0.06746233394369483, 0.0674750569742173, 0.06729638995602727, 0.06722936197184026, 0.06693222420290112, 0.06741319596767426, 0.06778236199170351, 0.06664331397041678, 0.06562340701930225, 0.06594559084624052, 0.0653765550814569, 0.06523241498507559, 0.0652684720698744, 0.06547155790030956, 0.06486320798285306, 0.06531073106452823, 0.06529909907840192, 0.06514106597751379, 0.06510097999125719, 0.06450047390535474, 0.06516859703697264, 0.06530753104016185, 0.06534425308927894, 0.06512550893239677, 0.06494371290318668, 0.06502848817035556, 0.06493869097903371, 0.06512037781067193, 0.06540501909330487, 0.0653454780112952, 0.06562919006682932, 0.06510776095092297, 0.06542371492832899, 0.06598912016488612, 0.06507519399747252, 0.06543516996316612, 0.07015723385848105, 0.06652302900329232, 0.06563512817956507, 0.06532548880204558, 0.06464855815283954, 0.06524479202926159, 0.06500583188608289, 0.06472401903010905, 0.06502557406201959, 0.06499379989691079, 0.06452788808383048, 0.06477210391312838, 0.06425291299819946, 0.06513251597061753, 0.0648915059864521, 0.0644478399772197, 0.06451934319920838, 0.06458604685030878, 0.06472019804641604, 0.06467294599860907, 0.06421606801450253, 0.06465502292849123, 0.06472142715938389, 0.06484473682940006, 0.06411091703921556, 0.06504148500971496, 0.06523044896312058, 0.06474073603749275, 0.06462616892531514, 0.06463928613811731, 0.06481000292114913, 0.06457489798776805, 0.06440383801236749, 0.06472176313400269, 0.06492879102006555, 0.06421279185451567, 0.06488967593759298, 0.0638814140111208, 0.07218105299398303, 0.057628732174634933, 0.06407168903388083, 0.06467089499346912, 0.06466168886981905, 0.06458589597605169, 0.06473135109990835, 0.06400711601600051, 0.06504299002699554, 0.06427338579669595, 0.06482366099953651, 0.06399511615745723, 0.06493088696151972, 0.06421511597000062, 0.06489907298237085, 0.06490237009711564, 0.06471334700472653, 0.0645160658750683, 0.06438215402886271, 0.06445454899221659, 0.06557672400958836, 0.06491201394237578, 0.06425563408993185, 0.06485751993022859, 0.06466432218439877, 0.06396194780245423, 0.06512589403428137, 0.06495937705039978, 0.06448213593102992, 0.06464470317587256, 0.06413776986300945, 0.06475572194904089, 0.0646591279655695, 0.06431375513784587, 0.06448021298274398, 0.06495006592012942, 0.06508350907824934, 0.06432091491296887, 0.06461476208642125, 0.06535123591311276, 0.0642901691608131, 0.06431621289812028, 0.06452200911007822, 0.06512944889254868, 0.06460814503952861, 0.06454359088093042, 0.06432440807111561, 0.06504465593025088, 0.06455799215473235, 0.06463390099816024, 0.0644061139319092, 0.06482668896205723, 0.06442025699652731, 0.06487901601940393, 0.06420160504058003, 0.06503577600233257, 0.06478581996634603, 0.06439576297998428, 0.06481069908477366, 0.06565893604420125, 0.06407694797962904, 0.06433882284909487, 0.06492207711562514, 0.06476782192476094, 0.0645283239427954, 0.06443604920059443, 0.06472944491542876, 0.06443621288053691, 0.06521616620011628, 0.06455309200100601, 0.06444836989976466, 0.06585559411905706, 0.07095594191923738, 0.059933257987722754, 0.06974821304902434, 0.06699436996132135, 0.06526649091392756, 0.0654029541183263, 0.06521027092821896, 0.06538011692464352, 0.06534861819818616, 0.06521519692614675, 0.06550236605107784, 0.06598517880775034, 0.06505078705959022, 0.06554818502627313, 0.06459939712658525, 0.06588188791647553, 0.06563010392710567, 0.06558794109150767, 0.06529622594825923, 0.06536181108094752, 0.06576147000305355, 0.06582289491780102, 0.06557161803357303, 0.06514829606749117, 0.06552738486789167, 0.06575190904550254, 0.0651756920851767, 0.06556435790844262, 0.06542383204214275, 0.06532995286397636, 0.06576171820051968, 0.06573628587648273, 0.06542856711894274, 0.06536336289718747, 0.06543730408884585, 0.06556904595345259, 0.06530692405067384, 0.06538172694854438, 0.06519609689712524, 0.06590038002468646, 0.0653855090495199, 0.06591909797862172, 0.06525585101917386, 0.06631448096595705, 0.0680481179151684, 0.06944618513807654, 0.06661924999207258, 0.06502044992521405, 0.06427695695310831, 0.06468048016540706, 0.06464007287286222, 0.06441681110300124, 0.06499741389416158, 0.06406246102415025, 0.06482557905837893, 0.06450403109192848, 0.06458896794356406, 0.06442634994164109, 0.06470930390059948, 0.06445369217544794, 0.06451438996009529, 0.06476315297186375, 0.06419815006665885, 0.06465732189826667, 0.06424696207977831, 0.06524092983454466, 0.06407944113016129, 0.06469436106272042, 0.06456666789017618, 0.06411394802853465, 0.06464638491161168, 0.0646805630531162, 0.06479180604219437, 0.06451914994977415, 0.06457416410557926, 0.06457962305285037, 0.06483250088058412, 0.06420479412190616, 0.0642389168497175, 0.06572900107130408, 0.06633080798201263, 0.06965666799806058, 0.06601569196209311, 0.06388647202402353, 0.064244220033288, 0.06392649095505476, 0.06364582711830735, 0.06410416378639638, 0.06331282900646329, 0.06378070916980505, 0.06390722887590528, 0.063864961033687, 0.06365235196426511, 0.06408959813416004, 0.06425638287328184, 0.06327944295480847, 0.06362124299630523, 0.06379713513888419, 0.06360726896673441, 0.06433546403422952, 0.06452852906659245, 0.06356802000664175, 0.06371398479677737, 0.06397717213258147, 0.06381209497340024, 0.0641321069560945, 0.06340158311650157, 0.06369455181993544, 0.06435144320130348, 0.06381278391927481, 0.06477125897072256, 0.07818270800635219, 0.04768044501543045, 0.04836503486149013, 0.04833525000140071, 0.04698760318569839, 0.04722650093026459, 0.04802374402061105, 0.046881044982001185, 0.047130371909588575, 0.04765503108501434, 0.048716288059949875], [0.07304923911578953, 0.07170511898584664, 0.07215986307710409, 0.07022398198023438, 0.07021196885034442, 0.07135085319168866, 0.07086099195294082, 0.07021210482344031, 0.07048424519598484, 0.07020912482403219, 0.07122221798636019, 0.07074272003956139, 0.07043630816042423, 0.06979538081213832, 0.07085477211512625, 0.07050723303109407, 0.07086854102090001, 0.0702268979512155, 0.07101327693089843, 0.07023731293156743, 0.07101632910780609, 0.07174565806053579, 0.07121051684953272, 0.0703480402007699, 0.07141365483403206, 0.07076351111754775, 0.07093875389546156, 0.07077928399667144, 0.07018845994025469, 0.07018582499586046, 0.07083672401495278, 0.07090278901159763, 0.07071101013571024, 0.07022309396415949, 0.0709399338811636, 0.07157760905101895, 0.07082535396330059, 0.07022220105864108, 0.07026890804991126, 0.07094523287378252, 0.07049551722593606, 0.07018070295453072, 0.07014289288781583, 0.07048220513388515, 0.07149078301154077, 0.07019744394347072, 0.070278394035995, 0.07103560492396355, 0.07099122507497668, 0.07055915798991919, 0.07064043590798974, 0.07088020909577608, 0.0704067968763411, 0.0700222379527986, 0.06977200717665255, 0.06938374997116625, 0.06999182398431003, 0.06970589095726609, 0.0694440770894289, 0.06983534200116992, 0.06985251000151038, 0.06968788080848753, 0.06967310002073646, 0.06987791415303946, 0.06979178497567773, 0.07036717399023473, 0.0701568538788706, 0.06930200918577611, 0.06973192398436368, 0.0694416188634932, 0.07022886513732374, 0.06981950299814343, 0.07020353781990707, 0.06929747317917645, 0.07019861694425344, 0.06978637096472085, 0.06999609898775816, 0.06994559708982706, 0.06993007683195174, 0.06949253100901842, 0.06982566299848258, 0.06968968198634684, 0.06955684605054557, 0.06989534199237823, 0.07006433396600187, 0.069960858207196, 0.07021242496557534, 0.06950053991749883, 0.06996160396374762, 0.06998431100510061, 0.06936879293061793, 0.06987282517366111, 0.06968016503378749, 0.07013846095651388, 0.07026064791716635, 0.06980804610066116, 0.07057176297530532, 0.06975458492524922, 0.0701839670073241, 0.06917373999021947, 0.06994438893161714, 0.0699243419803679, 0.06966531416401267, 0.06976526090875268, 0.07056850497610867, 0.0696533601731062, 0.07015841198153794, 0.06987553089857101, 0.06960441498085856, 0.06962887407280505, 0.06974173593334854, 0.06968073104508221, 0.07000325992703438, 0.07002362608909607, 0.06987468805164099, 0.06994434795342386, 0.06967873405665159, 0.06978322681970894, 0.07438671006821096, 0.0639553489163518, 0.06888425210490823, 0.06901553296484053, 0.06944140396080911, 0.06842794502153993, 0.06863651610910892, 0.06877353205345571, 0.06912190397270024, 0.06899751699529588, 0.06905843992717564, 0.06889184890314937, 0.06907935510389507, 0.06909771403297782, 0.06922786985523999, 0.06856513815000653, 0.06831640494056046, 0.0687670879997313, 0.06869830098003149, 0.0686236759647727, 0.06909650098532438, 0.06868388201110065, 0.0683376940432936, 0.06779820099473, 0.06788536603562534, 0.06788311409763992, 0.06760041485540569, 0.06800634600222111, 0.06789381196722388, 0.06835592398419976, 0.06829073000699282, 0.06756500597111881, 0.06791393621824682, 0.06783944484777749, 0.06751373992301524, 0.06804044102318585, 0.06797581212595105, 0.06826981692574918, 0.06794227892532945, 0.06788146821781993, 0.06714384886436164, 0.06884475704282522, 0.06795574794523418, 0.06759035610593855, 0.06780931586399674, 0.06800529407337308, 0.06760691408999264, 0.06772773689590394, 0.06794629292562604, 0.06758220819756389, 0.06770510785281658, 0.06771003501489758, 0.0674924529157579, 0.06803663703612983, 0.06763496017083526, 0.06816742685623467, 0.06775010307319462, 0.06775817507877946, 0.06772330193780363, 0.0673774539027363, 0.0680609040427953, 0.06764294090680778, 0.06762158405035734, 0.068135648034513, 0.0681930729188025, 0.06749209715053439, 0.06815771688707173, 0.06784964702092111, 0.06765424506738782, 0.0679283479694277, 0.06745782610960305, 0.06739817187190056, 0.06769372313283384, 0.06786760780960321, 0.06812361511401832, 0.06901528686285019, 0.06754251313395798, 0.06814811192452908, 0.06790556199848652, 0.06742224306799471, 0.06761395488865674, 0.06763515016064048, 0.0693563858512789, 0.06569234002381563, 0.06751160696148872, 0.06781083019450307, 0.06718381284736097, 0.06742665101774037, 0.06735756411217153, 0.06784562184475362, 0.06735207815654576, 0.06728789885528386, 0.06737394514493644, 0.06752817588858306, 0.06737226992845535, 0.06720761209726334, 0.06742128706537187, 0.06772323395125568, 0.06729648099280894, 0.06733416602946818, 0.06718996190465987, 0.06706387316808105, 0.06778774689882994, 0.06718524498865008, 0.06712307594716549, 0.06714241416193545, 0.06712237885221839, 0.0671032159589231, 0.06772389099933207, 0.06700152601115406, 0.0670102050062269, 0.06721759913489223, 0.06736485683359206, 0.06793448608368635, 0.06683936808258295, 0.06712013483047485, 0.06730251805856824, 0.06716209510341287, 0.06753828004002571, 0.06721095298416913, 0.06738387187942863, 0.06739859096705914, 0.06698575196787715, 0.06741514103487134, 0.06747555499896407, 0.0673444380518049, 0.06723133893683553, 0.0669337990693748, 0.06736064702272415, 0.06780177191831172, 0.06662993296049535, 0.06562861707061529, 0.06594113796018064, 0.06537635298445821, 0.06523815309628844, 0.06526479707099497, 0.06548095587641001, 0.06485196715220809, 0.06534458789974451, 0.06527326395735145, 0.06523383012972772, 0.06511730002239347, 0.06440490996465087, 0.06514746998436749, 0.06531379790976644, 0.06534084607847035, 0.0651305909268558, 0.06495373602956533, 0.06465986906550825, 0.06494075083173811, 0.06555817718617618, 0.06533240783028305, 0.06530899996869266, 0.0656795131508261, 0.0650801439769566, 0.06543669896200299, 0.06598870409652591, 0.06514923088252544, 0.0654684561304748, 0.07009738683700562, 0.06650912715122104, 0.06560678500682116, 0.06536358897574246, 0.06459944089874625, 0.06536387908272445, 0.06455295602791011, 0.06505718897096813, 0.06503644888289273, 0.06499830703251064, 0.0646344069391489, 0.0646753900218755, 0.06423157313838601, 0.0651346449740231, 0.06491885893046856, 0.06442303699441254, 0.06457316898740828, 0.06461656116880476, 0.06463751383125782, 0.06485406612046063, 0.06403308105655015, 0.0646439699921757, 0.064724484924227, 0.06485204305499792, 0.06412781681865454, 0.06513366918079555, 0.06524560786783695, 0.06462694308720529, 0.0646219349000603, 0.0646245339885354, 0.06481492007151246, 0.0646037079859525, 0.06437993701547384, 0.06482169893570244, 0.06484477198682725, 0.06419151811860502, 0.06489940895698965, 0.06387549289502203, 0.07169150002300739, 0.05812130100093782, 0.064157699001953, 0.06458362005650997, 0.06478569796308875, 0.06450498709455132, 0.06472891592420638, 0.06398958712816238, 0.0651141069829464, 0.06425732187926769, 0.06476820399984717, 0.06402712012641132, 0.06490124785341322, 0.06420116312801838, 0.06489806482568383, 0.06489200820215046, 0.06471587298437953, 0.06451108888722956, 0.06438381806947291, 0.06445865798741579, 0.06559056788682938, 0.06502808816730976, 0.0642251439858228, 0.06476496788673103, 0.06470626196824014, 0.06324855401180685, 0.06579903000965714, 0.06507233018055558, 0.06448559579439461, 0.06465464201755822, 0.06403574696741998, 0.06486465316265821, 0.06452376605011523, 0.06438076985068619, 0.06440755305811763, 0.06499245809391141, 0.0650723148137331, 0.06428914307616651, 0.06470583495683968, 0.06534898211248219, 0.06420226185582578, 0.06431739614345133, 0.06463894783519208, 0.06515093008056283, 0.06457885704003274, 0.06277714390307665, 0.06602530810050666, 0.06500697904266417, 0.06455816398374736, 0.064631552901119, 0.06441039498895407, 0.06486529600806534, 0.06438268115743995, 0.06486796890385449, 0.06427707988768816, 0.0649698032066226, 0.06478662486188114, 0.06439197110012174, 0.06481670285575092, 0.06567798717878759, 0.06405104487203062, 0.06439128704369068, 0.06489066896028817, 0.0647385991178453, 0.06453043990768492, 0.06443948904052377, 0.06473045004531741, 0.06435316195711493, 0.06535232486203313, 0.06449817400425673, 0.0644476420711726, 0.0658597310539335, 0.0709648469928652, 0.05992733989842236, 0.0694569309707731, 0.0672857949975878, 0.06526470417156816, 0.06540383794344962, 0.06520931795239449, 0.06538218911737204, 0.06534760585054755, 0.06521355896256864, 0.0655030740890652, 0.06600052095018327, 0.06504720402881503, 0.0655638671014458, 0.06457907799631357, 0.06587824900634587, 0.06563416682183743, 0.06558517017401755, 0.06529250601306558, 0.0653676139190793, 0.06575650093145669, 0.06582394312135875, 0.06558749196119606, 0.0651419518981129, 0.0655698562040925, 0.06575204781256616, 0.06513111013919115, 0.06555918883532286, 0.06542421015910804, 0.06535031087696552, 0.06574332294985652, 0.06573476199992001, 0.06542768701910973, 0.06536396313458681, 0.06513982894830406, 0.06587063707411289, 0.06530343601480126, 0.06537824193947017, 0.06538204196840525, 0.06572151696309447, 0.06538101099431515, 0.06591817503795028, 0.06527311797253788, 0.06629961403086782, 0.06805627909488976, 0.06944216298870742, 0.06661869585514069, 0.0650338961277157, 0.06427168101072311, 0.06470153899863362, 0.06462150602601469, 0.06440454791299999, 0.0649994199629873, 0.0640638661570847, 0.06482408288866282, 0.06450435589067638, 0.06460428214631975, 0.06441573100164533, 0.06470661400817335, 0.06445019599050283, 0.06451550987549126, 0.06477812118828297, 0.06419346993789077, 0.06465152790769935, 0.06355798011645675, 0.0659298449754715, 0.06407350301742554, 0.06469785398803651, 0.06456616800278425, 0.06406449200585485, 0.06470226892270148, 0.06469198199920356, 0.06478797504678369, 0.06451412197202444, 0.06457524397410452, 0.064573606941849, 0.06485117808915675, 0.06415088498033583, 0.06426154496148229, 0.06572107993997633, 0.06594217312522233, 0.06964149605482817, 0.06601960491389036, 0.06386408512480557, 0.06405720394104719, 0.06399236898869276, 0.06363123888149858, 0.06397810904309154, 0.0635447520762682, 0.06380077288486063, 0.06390662118792534, 0.06388799985870719, 0.06366722006350756, 0.06406180397607386, 0.0641850030515343, 0.06334127695299685, 0.06366491201333702, 0.06376156699843705, 0.06357745197601616, 0.06436769710853696, 0.06442976486869156, 0.06362855597399175, 0.0637799771502614, 0.06395515101030469, 0.06384023278951645, 0.0640787361189723, 0.06344966287724674, 0.06369397602975368, 0.06433473504148424, 0.06381676299497485, 0.06457782094366848, 0.0788630701135844, 0.04726813896559179, 0.048742568120360374, 0.04851715196855366, 0.0468066509347409, 0.04722356493584812, 0.04816389107145369, 0.04677042691037059, 0.047100458992645144, 0.04765311116352677, 0.048826834885403514, 0.10668158112093806], [0.0793581590987742, 0.0733656610827893, 0.0714668978471309, 0.07210623216815293, 0.07029238692484796, 0.07008635206148028, 0.0713762859813869, 0.07088809995912015, 0.07019434589892626, 0.07044421206228435, 0.0703130210749805, 0.07111999788321555, 0.07079572905786335, 0.07041740510612726, 0.06986894086003304, 0.07006301102228463, 0.07119771791622043, 0.07026340416632593, 0.07083490584045649, 0.07100717118009925, 0.07028160593472421, 0.07101550698280334, 0.07174966507591307, 0.0712094169575721, 0.07051316904835403, 0.07048296881839633, 0.07081337505951524, 0.07093809801153839, 0.07076676492579281, 0.07092412817291915, 0.07018378796055913, 0.07085911999456584, 0.07097578886896372, 0.06992091611027718, 0.07087752199731767, 0.07097930391319096, 0.07153875404037535, 0.07086055912077427, 0.07023221091367304, 0.07024911209009588, 0.07092173490673304, 0.07005762099288404, 0.07010386697947979, 0.07067540613934398, 0.0704610999673605, 0.0711314519867301, 0.07058922899886966, 0.07024536398239434, 0.07064081495627761, 0.07138952100649476, 0.07055578287690878, 0.0701349601149559, 0.07143122795969248, 0.07040823809802532, 0.06944211502559483, 0.06977236596867442, 0.06936521106399596, 0.07000032998621464, 0.06989650591276586, 0.0692532688844949, 0.070037028985098, 0.06965815811417997, 0.06968100601807237, 0.06964761088602245, 0.06988191208802164, 0.06987085309810936, 0.07021457701921463, 0.07006962294690311, 0.06936453701928258, 0.06984215788543224, 0.06945087993517518, 0.07028956199064851, 0.06976282713003457, 0.0701849979814142, 0.06929958099499345, 0.0701976630371064, 0.06978563591837883, 0.0699880390893668, 0.06994048599153757, 0.06992896599695086, 0.06948634586296976, 0.06971546402201056, 0.07001057802699506, 0.06924770306795835, 0.06999948108568788, 0.07005725498311222, 0.07004204601980746, 0.07012236397713423, 0.06951563199982047, 0.06994253303855658, 0.07007145578972995, 0.06923041702248156, 0.06999275111593306, 0.06958972592838109, 0.0701517800334841, 0.07025565695948899, 0.06971025397069752, 0.07047069817781448, 0.06978065078146756, 0.070144402096048, 0.06925801304168999, 0.07008360303007066, 0.06993212597444654, 0.06965122604742646, 0.07005394995212555, 0.07017357693985105, 0.06976947700604796, 0.06996094295755029, 0.0698764140252024, 0.0697646529879421, 0.06970304413698614, 0.06984105380252004, 0.0695463300216943, 0.0700156781822443, 0.07006428786553442, 0.06986589613370597, 0.06994820898398757, 0.06960164499469101, 0.06963042495772243, 0.06935448804870248, 0.06892383494414389, 0.06890466995537281, 0.06904978910461068, 0.069392150035128, 0.06841025780886412, 0.0685978070832789, 0.06922569405287504, 0.06913783098571002, 0.0688353800214827, 0.06886159395799041, 0.06912339199334383, 0.0691957559902221, 0.06909915804862976, 0.06923908996395767, 0.06842644698917866, 0.06843633088283241, 0.0687712652143091, 0.06848065787926316, 0.06843965500593185, 0.0690901989582926, 0.068681585136801, 0.06833269190974534, 0.06781563092954457, 0.06787604815326631, 0.06786873680539429, 0.06761169200763106, 0.06800487008877099, 0.06787562509998679, 0.06835813401266932, 0.06829715380445123, 0.0675586611032486, 0.06791354389861226, 0.0678467492107302, 0.06752089178189635, 0.0680150871630758, 0.0679774119053036, 0.06763713411055505, 0.06834247196093202, 0.06811534985899925, 0.06716012302786112, 0.06880814302712679, 0.06792958197183907, 0.06782560609281063, 0.06762184808030725, 0.06806667894124985, 0.06753447907976806, 0.06771449884399772, 0.06794146099127829, 0.06770689995028079, 0.06771545810624957, 0.06775046302936971, 0.06742828991264105, 0.06803426216356456, 0.06763880397193134, 0.06817213096655905, 0.06765038706362247, 0.06785532692447305, 0.06761997402645648, 0.06747393403202295, 0.06806813878938556, 0.06765798921696842, 0.06768206390552223, 0.06801941804587841, 0.06822745595127344, 0.06858578091487288, 0.06693019717931747, 0.06797204399481416, 0.0676641408354044, 0.06797226914204657, 0.06739836395718157, 0.06762057705782354, 0.06744064390659332, 0.06780922506004572, 0.06816054484806955, 0.06896294304169714, 0.06754060601815581, 0.06814534799195826, 0.06790685304440558, 0.06748343701474369, 0.06763635808601975, 0.06752217281609774, 0.06944058719091117, 0.06559964083135128, 0.06752119306474924, 0.06787130096927285, 0.06713076401501894, 0.06754561606794596, 0.06739398092031479, 0.06768771191127598, 0.06740147504024208, 0.06721186102367938, 0.06740488111972809, 0.06853394396603107, 0.06635885895229876, 0.06730551598593593, 0.06736323703080416, 0.06771494401618838, 0.0673230008687824, 0.06725559500046074, 0.06729853618890047, 0.06693632598035038, 0.06740106595680118, 0.0674017509445548, 0.06728001614101231, 0.06719728698953986, 0.0671232589520514, 0.06710657593794167, 0.06772271706722677, 0.06699037295766175, 0.06702479207888246, 0.06719049299135804, 0.06735450588166714, 0.06788106798194349, 0.06688198400661349, 0.06714235316030681, 0.06732052192091942, 0.06713014491833746, 0.06750727095641196, 0.06734892702661455, 0.06721551110967994, 0.06743000284768641, 0.06700451322831213, 0.06744014797732234, 0.06745503284037113, 0.06732804700732231, 0.06721307500265539, 0.0669397790916264, 0.06736850296147168, 0.06764010991901159, 0.06688554701395333, 0.06564370496198535, 0.06591052818112075, 0.06541911698877811, 0.06517981295473874, 0.06471341103315353, 0.06552551384083927, 0.0653915680013597, 0.06524460902437568, 0.0648049630690366, 0.06541975610889494, 0.06488161301240325, 0.06482442491687834, 0.06549526192247868, 0.06476891809143126, 0.06568750692531466, 0.06464432715438306, 0.06506164092570543, 0.06467400793917477, 0.06491355691105127, 0.06545488210394979, 0.06529923691414297, 0.06553124403581023, 0.06549143907614052, 0.06541883503086865, 0.06553671392612159, 0.06550020491704345, 0.06524302600882947, 0.06532176514156163, 0.0700535539072007, 0.06672424403950572, 0.06549905193969607, 0.06546098599210382, 0.06498888018541038, 0.06478819297626615, 0.06524022785015404, 0.06500822212547064, 0.06437684502452612, 0.06522529595531523, 0.06468336400575936, 0.06453136703930795, 0.06459173699840903, 0.06499412981793284, 0.06467101513408124, 0.06449601496569812, 0.06482286704704165, 0.06456058006733656, 0.06465499894693494, 0.06455029104836285, 0.06430219393223524, 0.06492847297340631, 0.06474003894254565, 0.06436037505045533, 0.06439902004785836, 0.064709102967754, 0.06535330996848643, 0.06472267699427903, 0.06471772608347237, 0.0649841888807714, 0.06476264307275414, 0.06438357289880514, 0.06416354910470545, 0.06493063294328749, 0.06473553809337318, 0.06463587190955877, 0.06447421503253281, 0.06431202194653451, 0.06444338895380497, 0.06513194413855672, 0.06385338306427002, 0.06491786683909595, 0.06431177700869739, 0.06469260295853019, 0.06474376213736832, 0.06421777606010437, 0.06502814590930939, 0.06433788104914129, 0.06455726898275316, 0.0644249739125371, 0.06449016905389726, 0.06451210100203753, 0.06491822306998074, 0.06493121897801757, 0.0647452010307461, 0.06455775094218552, 0.06437245500274003, 0.06432501901872456, 0.06528260186314583, 0.06490950402803719, 0.06420002202503383, 0.0650730540510267, 0.06453252607025206, 0.06580624589696527, 0.06314738211221993, 0.06496918993070722, 0.06436037691310048, 0.06470414809882641, 0.06427777302451432, 0.0649345419369638, 0.0648476560600102, 0.06420137989334762, 0.06466270913369954, 0.06456454400904477, 0.06476342980749905, 0.06501523219048977, 0.06426706095226109, 0.06530030001886189, 0.06467800494283438, 0.06434088293462992, 0.06437862804159522, 0.06475470098666847, 0.06472774595022202, 0.06471090717241168, 0.06442440580576658, 0.06481462111696601, 0.0644317630212754, 0.06507565290667117, 0.0641206530854106, 0.06465043290518224, 0.06480908207595348, 0.06479721004143357, 0.06429088697768748, 0.06499413400888443, 0.0647924009244889, 0.0643754699267447, 0.0648072340991348, 0.06501445104368031, 0.06471156887710094, 0.06434311112388968, 0.06489014788530767, 0.06479894299991429, 0.06455409503541887, 0.06572622410021722, 0.0634331488981843, 0.06496569397859275, 0.06469415710307658, 0.06451543304137886, 0.06448325980454683, 0.06715826317667961, 0.07024744292721152, 0.05945443990640342, 0.06966573698446155, 0.06713579408824444, 0.06526302103884518, 0.06543153803795576, 0.06520039401948452, 0.06537313899025321, 0.06535683292895555, 0.0651952400803566, 0.06548459595069289, 0.06594818597659469, 0.06508952309377491, 0.0655620729085058, 0.06454625795595348, 0.06593897799029946, 0.06561904912814498, 0.06558087095618248, 0.06531453994102776, 0.06538758496753871, 0.06572860898450017, 0.0658007541205734, 0.06564306397922337, 0.06508998293429613, 0.06551978597417474, 0.06533682206645608, 0.06551971309818327, 0.06561735179275274, 0.06545392610132694, 0.0653298411052674, 0.0657396309543401, 0.06571319792419672, 0.0654461239464581, 0.06542786001227796, 0.06528335809707642, 0.06571108405478299, 0.06528082489967346, 0.0653638590592891, 0.06537736603058875, 0.06593678891658783, 0.06518130702897906, 0.06592613598331809, 0.06526618893258274, 0.06624242500402033, 0.06873684213496745, 0.0690116488840431, 0.06663473113439977, 0.06504680286161602, 0.06419995310716331, 0.06471653282642365, 0.0646246001124382, 0.0644473119173199, 0.06499749212525785, 0.06407045992091298, 0.06482181302271783, 0.06451321509666741, 0.06457517179660499, 0.06441845907829702, 0.0647931310813874, 0.06433966988697648, 0.06463896506465971, 0.06470870808698237, 0.06413511792197824, 0.0646717119961977, 0.06520609906874597, 0.06429577595554292, 0.06407946906983852, 0.06465890794061124, 0.06459647486917675, 0.06430354015901685, 0.0644547319971025, 0.06468518590554595, 0.06479694996960461, 0.0645241781603545, 0.06457008398137987, 0.0645503168925643, 0.06486018095165491, 0.0642924290150404, 0.0643509910441935, 0.06573367398232222, 0.06711094593629241, 0.06965072802267969, 0.06601716810837388, 0.06388801499269903, 0.06488977605476975, 0.06344692083075643, 0.06363861099816859, 0.06394518609158695, 0.06352555193006992, 0.063778585055843, 0.06392148300074041, 0.06386152608320117, 0.06365859298966825, 0.06407784600742161, 0.06416504783555865, 0.0633673002012074, 0.0636357949115336, 0.06377178896218538, 0.06358333909884095, 0.06435490283183753, 0.06468433397822082, 0.06340364809148014, 0.06358508998528123, 0.06415099208243191, 0.06381261092610657, 0.06410054909065366, 0.06343920482322574, 0.06369421607814729, 0.06435120897367597, 0.06380990496836603, 0.06484897108748555, 0.07789425202645361, 0.04744749702513218, 0.04854086902923882, 0.04817300383001566, 0.04715743800625205, 0.04721570806577802, 0.04828009591437876, 0.046676365192979574, 0.04704216495156288, 0.047695491928607225, 0.04884765110909939, 0.10608769790269434], [0.04066235292702913, 0.04013765696436167, 0.03989252890460193, 0.039111822145059705, 0.03855957998894155, 0.038204767974093556, 0.03797756205312908, 0.03876092005521059, 0.03817525994963944, 0.03774547600187361, 0.037821448873728514, 0.03789586410857737, 0.03795165498740971, 0.038018109975382686, 0.037860084092244506, 0.03789522987790406, 0.03798439214006066, 0.03792742779478431, 0.03801565617322922, 0.03799612098373473, 0.03789449296891689, 0.037959331879392266, 0.03798181703314185, 0.037991740042343736, 0.03785836906172335, 0.03795780404470861, 0.03814423782750964, 0.037916976027190685, 0.037876770133152604, 0.038276297971606255, 0.038000819040462375, 0.03780524292960763, 0.0379750479478389, 0.0380377599503845, 0.03789240797050297, 0.038027461152523756, 0.037898480892181396, 0.037955373991280794, 0.03810107405297458, 0.03793479804880917, 0.03807115298695862, 0.03800578904338181, 0.03804202005267143, 0.037935392931103706, 0.03796929703094065, 0.03812826401554048, 0.03794690687209368, 0.03795975795947015, 0.0379199031740427, 0.03801318095065653, 0.03801462100818753, 0.03792563686147332, 0.038058234145864844, 0.03786960802972317, 0.03821494081057608, 0.037832827074453235, 0.03796504298225045, 0.03786548599600792, 0.037924620090052485, 0.03794154687784612, 0.03794472897425294, 0.038088256027549505, 0.038019350031390786, 0.03788674110546708, 0.03814793494530022, 0.03798788390122354, 0.038059966173022985, 0.03816352994181216, 0.037915792083367705, 0.037938063964247704, 0.037937199929729104, 0.03791882190853357, 0.03793839900754392, 0.03811467601917684, 0.03796412609517574, 0.037934727035462856, 0.038156062830239534, 0.038018921157345176, 0.038002777844667435, 0.0379623151384294, 0.03792210086248815, 0.03806649008765817, 0.03785037505440414, 0.03812915692105889, 0.03791839792393148, 0.03832307318225503, 0.03856103797443211, 0.03821641695685685, 0.0378582919947803, 0.03790785907767713, 0.03798986691981554, 0.03802375798113644, 0.03801638912409544, 0.03798234695568681, 0.0380021499004215, 0.03791438112966716, 0.03794163488782942, 0.03793892404064536, 0.038048076909035444, 0.03808785113506019, 0.03797465795651078, 0.03796300897374749, 0.0378933590836823, 0.03802192001603544, 0.03786382800899446, 0.03796440688893199, 0.037988235941156745, 0.038015953032299876, 0.03792805899865925, 0.038063960149884224, 0.03944937093183398, 0.038060280960053205, 0.03788925311528146, 0.03797249682247639, 0.037928970996290445, 0.03792590997181833, 0.03792524803429842, 0.03814283711835742, 0.03785677487030625, 0.03817695006728172, 0.03862358792684972, 0.03851702902466059, 0.038214008091017604, 0.03793247207067907, 0.037951216800138354, 0.037934388034045696, 0.03799602808430791, 0.038011335069313645, 0.03790629399009049, 0.03797853086143732, 0.03808047506026924, 0.03831316204741597, 0.03797778394073248, 0.03804310201667249, 0.03830719692632556, 0.037735410034656525, 0.03787418408319354, 0.038075783057138324, 0.037880070973187685, 0.03799576289020479, 0.03806449891999364, 0.03797266120091081, 0.037910747807472944, 0.03819257509894669, 0.03816504590213299, 0.037885481026023626, 0.0379823031835258, 0.03804759681224823, 0.03793997201137245, 0.038016543025150895, 0.038335025077685714, 0.037772889947518706, 0.03787238197401166, 0.03792367107234895, 0.037972148042172194, 0.03806930687278509, 0.037926210090518, 0.03795104497112334, 0.03812267305329442, 0.03793900087475777, 0.038001750130206347, 0.03806080296635628, 0.03791223303414881, 0.03791950596496463, 0.03860878595151007, 0.038182457908988, 0.03820865415036678, 0.03801048989407718, 0.03803237294778228, 0.0380541339982301, 0.03803522209636867, 0.03794588311575353, 0.03787080687470734, 0.03852820512838662, 0.03764074295759201, 0.03807750600390136, 0.03787176194600761, 0.03793661808595061, 0.03796228696592152, 0.03817514004185796, 0.03781120595522225, 0.03813706897199154, 0.03789669997058809, 0.037956805899739265, 0.03790532611310482, 0.03810572694055736, 0.03779488313011825, 0.03799187601543963, 0.03805425297468901, 0.03806509496644139, 0.03802463295869529, 0.038170827087014914, 0.038114244816824794, 0.037884576013311744, 0.03828985616564751, 0.039452673867344856, 0.0380803591106087, 0.03821878903545439, 0.037906133802607656, 0.03814316401258111, 0.037944849114865065, 0.03825845895335078, 0.0378888959530741, 0.03811604296788573, 0.03780560404993594, 0.03810279001481831, 0.03809131309390068, 0.038018422899767756, 0.03819819702766836, 0.037886590929701924, 0.037821375066414475, 0.038126308005303144, 0.03785701701417565, 0.03819665010087192, 0.03794784680940211, 0.0392324801068753, 0.03764735488221049, 0.03810949204489589, 0.03795033413916826, 0.03816771297715604, 0.03795761591754854, 0.03794872690923512, 0.0381706110201776, 0.037883064011111856, 0.037928458070382476, 0.037977419095113873, 0.037939834874123335, 0.03807133203372359, 0.03796545811928809, 0.0381028838455677, 0.037901508156210184, 0.03808669280260801, 0.037908334052190185, 0.0382034529466182, 0.03793244203552604, 0.038061578990891576, 0.0380469779483974, 0.03803464910015464, 0.03794143092818558, 0.037999029038473964, 0.03835013112984598, 0.037753468845039606, 0.039225980173796415, 0.03804411180317402, 0.037989987060427666, 0.03802254912443459, 0.03801150992512703, 0.038073448929935694, 0.038194559048861265, 0.0381275899708271, 0.03783413302153349, 0.03798411809839308, 0.0379880559630692, 0.03808240592479706, 0.03794470801949501, 0.038528240052983165, 0.037954629864543676, 0.03813465405255556, 0.03822882496751845, 0.03810932906344533, 0.03782336902804673, 0.038094897055998445, 0.03806663095019758, 0.03817632491700351, 0.037933768006041646, 0.03814182709902525, 0.03808489884249866, 0.038039152044802904, 0.03781615803018212, 0.038065864937379956, 0.037843707017600536, 0.03819067496806383, 0.03775838203728199, 0.0380080530885607, 0.03792410693131387, 0.03803881211206317, 0.03793581295758486, 0.03807323705404997, 0.03801206103526056, 0.037915403954684734, 0.03795595187693834, 0.03819034807384014, 0.038037575082853436, 0.038076760014519095, 0.0379598259460181, 0.03809534292668104, 0.03804419003427029, 0.03809907287359238, 0.03805674915201962, 0.043514366960152984, 0.038471818901598454, 0.0416053521912545, 0.04166547884233296, 0.041783557971939445, 0.04168670508079231, 0.041787005960941315, 0.04166358197107911, 0.041632791981101036, 0.04224523017182946, 0.04165153484791517, 0.04210234014317393, 0.04192741587758064, 0.04184664902277291, 0.04164540092460811, 0.041676587192341685, 0.041553102899342775, 0.04204027005471289, 0.041601005010306835, 0.04164855694398284, 0.041760411113500595, 0.041620364878326654, 0.04176420206204057, 0.04173869104124606, 0.04168061399832368, 0.04153569298796356, 0.041811608942225575, 0.0416992730461061, 0.042310857912525535, 0.04126765113323927, 0.041714458959177136, 0.04177543194964528, 0.0416855679359287, 0.043430766090750694, 0.0403166429605335, 0.04168740310706198, 0.0419068387709558, 0.04155827220529318, 0.041596515802666545, 0.04168158816173673, 0.041647953912615776, 0.041597248055040836, 0.0415685479529202, 0.04161803494207561, 0.04164401814341545, 0.04167739301919937, 0.04144083382561803, 0.04166813800111413, 0.041788706090301275, 0.04171126591973007, 0.04165466409176588, 0.0416399659588933, 0.04148562112823129, 0.041582122910767794, 0.0417068290989846, 0.04177114390768111, 0.041671406012028456, 0.0446875870693475, 0.04160898202098906, 0.04173918999731541, 0.041624093893915415, 0.04150447202846408, 0.04149717092514038, 0.0416689720004797, 0.04138601804152131, 0.04156734305433929, 0.04152529686689377, 0.04167567007243633, 0.04154855199158192, 0.041589572094380856, 0.041590072913095355, 0.041529539972543716, 0.04217616911046207, 0.041845898842439055, 0.041430096141994, 0.042787483893334866, 0.04617389012128115, 0.044926143949851394, 0.047532368917018175, 0.047448833007365465, 0.047696194145828485, 0.04750157194212079, 0.047476059990003705, 0.047570958035066724, 0.04720116802491248, 0.04736845986917615, 0.04748241207562387, 0.047253239899873734, 0.04747057403437793, 0.04751722398214042, 0.04722110298462212, 0.04753114702180028, 0.0473251030780375, 0.047274623066186905, 0.04751829500310123, 0.04719581687822938, 0.04747386905364692, 0.04782545589841902, 0.04869192698970437, 0.04751812107861042, 0.04759011999703944, 0.047513776924461126, 0.04765545413829386, 0.04748362605459988, 0.04756287788040936, 0.04736301302909851, 0.04736637603491545, 0.047478029038757086, 0.04732403391972184, 0.04863562900573015, 0.04743915889412165, 0.047428609104827046, 0.04744689702056348, 0.04789514793083072, 0.048661583103239536, 0.04907680791802704, 0.046907237032428384, 0.04746616096235812, 0.047958164010196924, 0.04722351091913879, 0.047343265963718295, 0.04750203504227102, 0.0476118300575763, 0.04754598205909133, 0.04745044792070985, 0.047872297931462526, 0.04671731614507735, 0.0477424249984324, 0.0469011168461293, 0.047303991159424186, 0.04739404399879277, 0.04751290404237807, 0.0472836138214916, 0.04755112715065479, 0.04752045893110335, 0.0478607879485935, 0.047297134064137936, 0.047233151039108634, 0.04724734090268612, 0.04758933815173805, 0.047436332795768976, 0.047192515106871724, 0.047340833116322756, 0.04733873880468309, 0.04718922218307853, 0.04759951587766409, 0.04770064493641257, 0.048967912094667554, 0.047242785105481744, 0.04747443785890937, 0.0473511079326272, 0.0482473149895668, 0.047413786174729466, 0.04730500094592571, 0.04743905900977552, 0.04763850895687938, 0.04702157597057521, 0.0474855899810791, 0.04690498998388648, 0.04744618618860841, 0.047182197915390134, 0.0472244790289551, 0.047167225973680615, 0.04736928595229983, 0.047290106071159244, 0.04721671505831182, 0.04743316280655563, 0.04733041604049504, 0.0474314340390265, 0.04751459998078644, 0.04710160498507321, 0.04724973510019481, 0.04749529086984694, 0.04737389017827809, 0.04739613295532763, 0.04722041799686849, 0.04733340605162084, 0.047624449944123626, 0.05198037996888161, 0.06321106292307377, 0.06357296905480325, 0.06447525508701801, 0.06448195385746658, 0.06477050599642098, 0.06477564596571028, 0.06455377605743706, 0.06451970199123025, 0.0644456441514194, 0.06457781395874918, 0.06457432103343308, 0.06440248293802142, 0.06412750296294689, 0.06451567308977246, 0.06430620886385441, 0.06544984597712755, 0.06429854314774275, 0.0643756368663162, 0.06426560715772212, 0.06463773385621607, 0.06432860414497554, 0.06417443579994142, 0.06530735711567104, 0.0642709769308567, 0.06444301106967032, 0.06443281495012343, 0.06415672600269318, 0.0643960740417242, 0.06460357503965497, 0.06509646004997194, 0.06427099695429206, 0.06422623782418668, 0.06432502600364387, 0.06441788305528462, 0.06480196700431406, 0.06389273307286203, 0.06415789388120174, 0.06428364710882306, 0.063893310027197, 0.06429557199589908, 0.06431240402162075, 0.0649584939237684], [0.04160891380161047, 0.041647237027063966, 0.041775283170863986, 0.04169684601947665, 0.041776756988838315, 0.04167129984125495, 0.04163940506987274, 0.042242015013471246, 0.04166123387403786, 0.04209109302610159, 0.04192162700928748, 0.041877662064507604, 0.0416322520468384, 0.04167483886703849, 0.0415648790076375, 0.04203315917402506, 0.04161055292934179, 0.041655490873381495, 0.04173085605725646, 0.04162073996849358, 0.041764826979488134, 0.04177812999114394, 0.04165449505671859, 0.04153612907975912, 0.041805268032476306, 0.041897838935256004, 0.042213534004986286, 0.04117588489316404, 0.041714369086548686, 0.04176666494458914, 0.041691774036735296, 0.04341951897367835, 0.040316194063052535, 0.04167846706695855, 0.04192950902506709, 0.04155356180854142, 0.04159305221401155, 0.04167618998326361, 0.0416531078517437, 0.041587179992347956, 0.041563607985153794, 0.041615480091422796, 0.04164669290184975, 0.04168344917707145, 0.04143382399342954, 0.0416690509300679, 0.04178949701599777, 0.041707200929522514, 0.04165834607556462, 0.041641570860520005, 0.04148276709020138, 0.04152136202901602, 0.04170360695570707, 0.0418368570972234, 0.041705068899318576, 0.044717655051499605, 0.0415779419708997, 0.041730874916538596, 0.04164692712947726, 0.04147484805434942, 0.041483112843707204, 0.04168002912774682, 0.04138020193204284, 0.04157876200042665, 0.04151802998967469, 0.04167385399341583, 0.041547741973772645, 0.04159375396557152, 0.041578743141144514, 0.04154238593764603, 0.04216672899201512, 0.041856107069179416, 0.041410468984395266, 0.04288824088871479, 0.04611936118453741, 0.04501605383120477, 0.04752601310610771, 0.04742380604147911, 0.04771981085650623, 0.047438828041777015, 0.047497357008978724, 0.04753438290208578, 0.04725735215470195, 0.047372995875775814, 0.04747932916507125, 0.04725152486935258, 0.047472791047766805, 0.047522217966616154, 0.04720715503208339, 0.04753590398468077, 0.047356321942061186, 0.04726122505962849, 0.04751047398895025, 0.04720694711431861, 0.04746262985281646, 0.047869140980765224, 0.048663703026250005, 0.04751671012490988, 0.04760854784399271, 0.047507354989647865, 0.04764415114186704, 0.047472079982981086, 0.04763352801091969, 0.04731443687342107, 0.04734225501306355, 0.047482321970164776, 0.04733245610259473, 0.04864541604183614, 0.047414667904376984, 0.04744064505212009, 0.04746096604503691, 0.048042298993095756, 0.04857896687462926, 0.04902335302904248, 0.04697320400737226, 0.0474829759914428, 0.048049983102828264, 0.04711777297779918, 0.047320306999608874, 0.047513138968497515, 0.0475981249473989, 0.04754935693927109, 0.04744754801504314, 0.04786962014622986, 0.04672812391072512, 0.04773507406935096, 0.04693835508078337, 0.04728615586645901, 0.04739040392450988, 0.04910119599662721, 0.04569856310263276, 0.047550773015245795, 0.047534286975860596, 0.04788132896646857, 0.0472726549487561, 0.04723195009864867, 0.047321622958406806, 0.04754070704802871, 0.04741261689923704, 0.047202569199725986, 0.04734770185314119, 0.04731562500819564, 0.04720929404720664, 0.0476055140607059, 0.047684629913419485, 0.04897394706495106, 0.047252110904082656, 0.047479419969022274, 0.04733860515989363, 0.048311178805306554, 0.04736171499826014, 0.04730259906500578, 0.04745950014330447, 0.04769826587289572, 0.04685107199475169, 0.04755162005312741, 0.046912232879549265, 0.04746095719747245, 0.047187417978420854, 0.04721172200515866, 0.047164413845166564, 0.04737094114534557, 0.047287730034440756, 0.04722398589365184, 0.04743271600455046, 0.047318296041339636, 0.04744209093041718, 0.047524735098704696, 0.04709433694370091, 0.0472622150555253, 0.04750170698389411, 0.047375489957630634, 0.04739520791918039, 0.04721298906952143, 0.04733303491957486, 0.047733324114233255, 0.05180425988510251, 0.06266960501670837, 0.06395800714381039, 0.06459748884662986, 0.06449781102128327, 0.06481718714348972, 0.06477455398999155, 0.06454750685952604, 0.06446660705842078, 0.06453463109210134, 0.06455587781965733, 0.06451821816153824, 0.0643907729536295, 0.06420980487018824, 0.06452870019711554, 0.06428302079439163, 0.06548444903455675, 0.06422218307852745, 0.06446445290930569, 0.06426170305348933, 0.06464135902933776, 0.06432475498877466, 0.06418841308914125, 0.06530254683457315, 0.06425939314067364, 0.06446100398898125, 0.06442890199832618, 0.06414926401339471, 0.0643977930303663, 0.0645913309417665, 0.06510158791206777, 0.06425374606624246, 0.06424903287552297, 0.06433244701474905, 0.06490451004356146, 0.0642222489695996, 0.06399338808842003, 0.06417399295605719, 0.06426883209496737, 0.06388769298791885, 0.06430547195486724, 0.06431296304799616, 0.06489376304671168, 0.07562686991877854, 0.05229801102541387, 0.06891054986044765, 0.06572562409564853, 0.06468859990127385, 0.06482854718342423, 0.06458887993358076, 0.06456743297167122, 0.06469835294410586, 0.06457293615676463, 0.06489636399783194, 0.0649596219882369, 0.06422834494151175, 0.0646454649977386, 0.06473708408884704, 0.0649512279778719, 0.06455301702953875, 0.06457852781750262, 0.06482344097457826, 0.06467849900946021, 0.06467814208008349, 0.06488584307953715, 0.06478822091594338, 0.06461217091418803, 0.06472841207869351, 0.06429781601764262, 0.0648481419775635, 0.06471491907723248, 0.06456892285495996, 0.06561174499802291, 0.06470131408423185, 0.06473370804451406, 0.06469957297667861, 0.06435261596925557, 0.06502451305277646, 0.06473446288146079, 0.06464864313602448, 0.06463135406374931, 0.06490542180836201, 0.06472133006900549, 0.06509361905045807, 0.06479090289212763, 0.06461165007203817, 0.06530455499887466, 0.06508924392983317, 0.06493983301334083, 0.06504951696842909, 0.0647086959797889, 0.06481347605586052, 0.06512248795479536, 0.06468559615314007, 0.0645401009824127, 0.06497448892332613, 0.06492373114451766, 0.06465140590444207, 0.06454861396923661, 0.06462062895298004, 0.06478759017772973, 0.06491479580290616, 0.0651224059984088, 0.06491831410676241, 0.06475708587095141, 0.0649155389983207, 0.06462615518830717, 0.06512512592598796, 0.06432770402170718, 0.06459734798409045, 0.0648188809864223, 0.06477395095862448, 0.0646167250815779, 0.06520472094416618, 0.06462502898648381, 0.06460322998464108, 0.0644840111490339, 0.0648106848821044, 0.0647607899736613, 0.06503005814738572, 0.06465612398460507, 0.06450302293524146, 0.06444747000932693, 0.0647381090093404, 0.0647317860275507, 0.06495504383929074, 0.06444160803221166, 0.0642244340851903, 0.0651501880493015, 0.06487114704214036, 0.06584407296031713, 0.06506595481187105, 0.06469066021963954, 0.06475762790068984, 0.06472284789197147, 0.06462516402825713, 0.06454238598234951, 0.06496008206158876, 0.06475398014299572, 0.06497916183434427, 0.06440066709183156, 0.0646848778706044, 0.06477025710046291, 0.0655232248827815, 0.06424170918762684, 0.06470962590537965, 0.0646111040841788, 0.06460883584804833, 0.06459938711486757, 0.06514291907660663, 0.06468229996971786, 0.06453805486671627, 0.06445208913646638, 0.06532565085217357, 0.06479075597599149, 0.06501424103043973, 0.0658998959697783, 0.06567005114629865, 0.06331445300020278, 0.0646002369467169, 0.0650472310371697, 0.06424099090509117, 0.06446612603031099, 0.06411425699479878, 0.06446277699433267, 0.06431263498961926, 0.06391209503635764, 0.06439913506619632, 0.06439114501699805, 0.06406145496293902, 0.06466072704643011, 0.06421240698546171, 0.06436281185597181, 0.06461011315695941, 0.06427877093665302, 0.06478836899623275, 0.06414496805518866, 0.06382339494302869, 0.06435443600639701, 0.06434474000707269, 0.06429797690361738, 0.06421298813074827, 0.0641807068604976, 0.06452978006564081, 0.06361584109254181, 0.06440399191342294, 0.06444123503752053, 0.0652482679579407, 0.06495491904206574, 0.06537037901580334, 0.0649856049567461, 0.0654978770762682, 0.06477444502525032, 0.0643247808329761, 0.06417689705267549, 0.06443412089720368, 0.06398310209624469, 0.06465544295497239, 0.06396161811426282, 0.06449605198577046, 0.06409561401233077, 0.06418347405269742, 0.06379321590065956, 0.06448861910030246, 0.06413828581571579, 0.06442693201825023, 0.06429463601671159, 0.06408232799731195, 0.06430807011201978, 0.06456214701756835, 0.06397912884131074, 0.06447777710855007, 0.06372462399303913, 0.06397904898039997, 0.06397469597868621, 0.06434433697722852, 0.0639959410764277, 0.06421373900957406, 0.06400798098184168, 0.06408654595725238, 0.06407555099576712, 0.06413829606026411, 0.06402000086382031, 0.06430741818621755, 0.06430460792034864, 0.06406113691627979, 0.06382520613260567, 0.06448950082994998, 0.064128726022318, 0.06430039810948074, 0.0639932060148567, 0.06423838785849512, 0.06424107612110674, 0.06432224600575864, 0.06397710205055773, 0.06411297200247645, 0.06390277692116797, 0.06415516301058233, 0.06394602707587183, 0.0641895579174161, 0.06408052588813007, 0.06415617512539029, 0.06424499093554914, 0.06421566801145673, 0.06401697313413024, 0.06490472285076976, 0.06374969193711877, 0.06415538303554058, 0.06425499101169407, 0.0644565790425986, 0.06419264408759773, 0.0644432429689914, 0.06429682695306838, 0.06409722310490906, 0.06388188689015806, 0.06419643806293607, 0.06431530090048909, 0.06456289207562804, 0.06441658502444625, 0.06447030487470329, 0.0653825190383941, 0.06321160914376378, 0.06377852987498045, 0.0643831689376384, 0.06410395097918808, 0.06510532507672906, 0.063737713964656, 0.0642119231633842, 0.06401629280298948, 0.06443821010179818, 0.06413573608733714, 0.06422167900018394, 0.06384503794834018, 0.06469748495146632, 0.06394534395076334, 0.06419377401471138, 0.06440349109470844, 0.06389345601201057, 0.06420985283330083, 0.06377412611618638, 0.06399058108218014, 0.06433825194835663, 0.06469183205626905, 0.06335243885405362, 0.06379350693896413, 0.06396979000419378, 0.06414797506295145, 0.06413469603285193, 0.06475591589696705, 0.06418502307496965, 0.06408128305338323, 0.06391287199221551, 0.06410894799046218, 0.06423428491689265, 0.06430802610702813, 0.06445275107398629, 0.06388360587880015, 0.06422532396391034, 0.06396097200922668, 0.06431855214759707, 0.06427043792791665, 0.06419198191724718, 0.06374665512703359, 0.06439712992869318, 0.06400789297185838, 0.06420010211877525, 0.06421654089353979, 0.06419649301096797, 0.06365762208588421, 0.06421432294882834, 0.06401454890146852, 0.06420651101507246, 0.06506942608393729, 0.06414184300228953, 0.06474425387568772, 0.06372231501154602, 0.06402274500578642, 0.06317136716097593, 0.06422129482962191, 0.06470167706720531, 0.06440643197856843, 0.0638859469909221, 0.07072901516221464, 0.06567461998201907, 0.06439170381054282, 0.06423487118445337, 0.06442162790335715, 0.06440439401194453, 0.06403683801181614, 0.06408399203792214, 0.06459087901748717, 0.06408312893472612, 0.06428839988075197, 0.0655535280238837], [0.04753547301515937, 0.04742500279098749, 0.047877602046355605, 0.04729479504749179, 0.047491203993558884, 0.047661951975896955, 0.047126602148637176, 0.04736679187044501, 0.047483162954449654, 0.04724962613545358, 0.04747178195975721, 0.04752395092509687, 0.04720749915577471, 0.047535010846331716, 0.047354141948744655, 0.04726022505201399, 0.0475139650516212, 0.047202908899635077, 0.047466891119256616, 0.04785550502128899, 0.048669084906578064, 0.0475188831333071, 0.047605918953195214, 0.04750605090521276, 0.04764575813896954, 0.047479017870500684, 0.04761918797157705, 0.047314380994066596, 0.0473567470908165, 0.047481812071055174, 0.04732337384484708, 0.04865231807343662, 0.0474160120356828, 0.0474386690184474, 0.04744346998631954, 0.04799801600165665, 0.04861525003798306, 0.04903277987614274, 0.046918889973312616, 0.0474722501821816, 0.04809221997857094, 0.04708766588009894, 0.047329595079645514, 0.047506215050816536, 0.047602414852008224, 0.04755565710365772, 0.04744351003319025, 0.04787425487302244, 0.0467138250824064, 0.047772001940757036, 0.046898815082386136, 0.04729421203956008, 0.04738408490084112, 0.049069313099607825, 0.04574037599377334, 0.04754572594538331, 0.04754336201585829, 0.047864028019830585, 0.047279869904741645, 0.047233015997335315, 0.04731095512397587, 0.04754077293910086, 0.04742360091768205, 0.04720262996852398, 0.04734511696733534, 0.047323010163381696, 0.047204962000250816, 0.04756398592144251, 0.04772501112893224, 0.04897196893580258, 0.047246995847672224, 0.04747976711951196, 0.04733908409252763, 0.048289830796420574, 0.04737882921472192, 0.047308132983744144, 0.04744914500042796, 0.047707172809168696, 0.046913146041333675, 0.04748665913939476, 0.04691709787584841, 0.04745520604774356, 0.047179657965898514, 0.047220933018252254, 0.047162147937342525, 0.04736888618208468, 0.04729299689643085, 0.04722090787254274, 0.04743018001317978, 0.0473256791010499, 0.04743573092855513, 0.04752101702615619, 0.04709740704856813, 0.04726120689883828, 0.04750194400548935, 0.04736987315118313, 0.04739737999625504, 0.04721361701376736, 0.047335145995020866, 0.04773576697334647, 0.051869068993255496, 0.06285138800740242, 0.0639463618863374, 0.06464338907971978, 0.0645125200971961, 0.06479783379472792, 0.0647730550263077, 0.06454714899882674, 0.06449123797938228, 0.06451811618171632, 0.06455641891807318, 0.06452897191047668, 0.0643946691416204, 0.06420489796437323, 0.06451963190920651, 0.06428383802995086, 0.06548413611017168, 0.06424277392216027, 0.0644495920278132, 0.06426187604665756, 0.0646421208512038, 0.06432730215601623, 0.06418599095195532, 0.0653034879360348, 0.06427515903487802, 0.06439320789650083, 0.06447937316261232, 0.06415956886485219, 0.06439675018191338, 0.06459369300864637, 0.06509962282143533, 0.06426665210165083, 0.06423892290331423, 0.06434923713095486, 0.06492898194119334, 0.06419698800891638, 0.06397679192014039, 0.06417677109129727, 0.06427075294777751, 0.06387797696515918, 0.06425205408595502, 0.0643189288675785, 0.06496616010554135, 0.07563790795393288, 0.052287246100604534, 0.06884521385654807, 0.06578240916132927, 0.06468668091110885, 0.06476631807163358, 0.06466516805812716, 0.06454686284996569, 0.06470587896183133, 0.06457954808138311, 0.06489778496325016, 0.06497665005736053, 0.0641470190603286, 0.06471779383718967, 0.0647350731305778, 0.06494804006069899, 0.0645386828109622, 0.0645827769767493, 0.06482159020379186, 0.06468733097426593, 0.0646788408048451, 0.06488397205248475, 0.06503870408050716, 0.06455611088313162, 0.0645381840877235, 0.064419100061059, 0.06473053595982492, 0.06471764296293259, 0.06474781897850335, 0.06542232702486217, 0.06470314995385706, 0.06474046013318002, 0.06469810986891389, 0.06435804697684944, 0.06502626417204738, 0.06472799181938171, 0.06465998711064458, 0.06462439009919763, 0.06490799179300666, 0.06472152005881071, 0.06509235803969204, 0.06479559699073434, 0.06461457605473697, 0.06530238804407418, 0.06508327182382345, 0.06493402016349137, 0.06505929888226092, 0.06471469393000007, 0.06481647305190563, 0.06511661410331726, 0.06469585397280753, 0.06453640502877533, 0.06498220702633262, 0.0649415678344667, 0.06462369812652469, 0.06454533198848367, 0.06461920496076345, 0.06478973594494164, 0.06490419199690223, 0.06514210510067642, 0.06491375993937254, 0.06475542206317186, 0.06491482607088983, 0.06462447601370513, 0.06530883396044374, 0.06417335802689195, 0.06458287197165191, 0.06481546186842024, 0.06478063808754086, 0.06461192108690739, 0.06521004578098655, 0.06462079402990639, 0.0646033831872046, 0.0644751088693738, 0.06481361109763384, 0.06476320791989565, 0.06503495806828141, 0.06465632398612797, 0.06449160794727504, 0.06445908499881625, 0.06474267691373825, 0.06473503005690873, 0.06495173298753798, 0.06442981911823153, 0.06462229788303375, 0.06476468406617641, 0.06486272695474327, 0.06585122109390795, 0.06506701302714646, 0.06470224098302424, 0.06473987991921604, 0.06472458899952471, 0.06462811888195574, 0.06453774916008115, 0.06495672394521534, 0.06477337307296693, 0.06496834894642234, 0.0644003648776561, 0.06468754820525646, 0.06477250996977091, 0.0655962168239057, 0.064164815004915, 0.06471416004933417, 0.06461567501537502, 0.06459898105822504, 0.06460498995147645, 0.06522324308753014, 0.0646074500400573, 0.0645371398422867, 0.06445273500867188, 0.06531549314968288, 0.06480509485118091, 0.06500594597309828, 0.0659022661857307, 0.06596808787435293, 0.06315269297920167, 0.06479361304081976, 0.06477025197818875, 0.06421032198704779, 0.06451603211462498, 0.06404918385669589, 0.06464807502925396, 0.06426631403155625, 0.06378340511582792, 0.06439274293370545, 0.06439015385694802, 0.06405613501556218, 0.0646648311521858, 0.06421376904472709, 0.06435720482841134, 0.06460787216201425, 0.06427957699634135, 0.06488376786001027, 0.06408311799168587, 0.06379465595819056, 0.0643631829880178, 0.06436167214997113, 0.06427945895120502, 0.06421517790295184, 0.06416698009707034, 0.0645261590834707, 0.06362080085091293, 0.06440197816118598, 0.06444419687613845, 0.06532363593578339, 0.06488067819736898, 0.06537763099186122, 0.06497009890154004, 0.06549889710731804, 0.06477353279478848, 0.06442005513235927, 0.06437398097477853, 0.06417365605011582, 0.0639648528303951, 0.06466618506237864, 0.06394357816316187, 0.06452893489040434, 0.06407260405831039, 0.06417760392650962, 0.0637953020632267, 0.06449067406356335, 0.06413938896730542, 0.06442679790779948, 0.06429799692705274, 0.06411256408318877, 0.06428835890255868, 0.06458017509430647, 0.06403906107880175, 0.06443203589878976, 0.06369384494610131, 0.0639926299918443, 0.06395809911191463, 0.0643759360536933, 0.06396550685167313, 0.06420871196314692, 0.06401092722080648, 0.06412101280875504, 0.06404701597057283, 0.06413678103126585, 0.06401452096179128, 0.0643204590305686, 0.0643063741736114, 0.06405802886001766, 0.06381217413581908, 0.06449845596216619, 0.06412467500194907, 0.0644008528906852, 0.06389281107112765, 0.06424012687057257, 0.06424991716630757, 0.06433574901893735, 0.06396789802238345, 0.06410187482833862, 0.06392134609632194, 0.06413992401212454, 0.063963278895244, 0.06416581408120692, 0.06408069794997573, 0.06415709014981985, 0.06425526295788586, 0.06421946990303695, 0.06400393601506948, 0.06491515506058931, 0.06374759902246296, 0.06415773485787213, 0.06424883613362908, 0.06445426400750875, 0.06419044686481357, 0.06445655808784068, 0.06428860407322645, 0.06409217696636915, 0.06389176798984408, 0.06419625296257436, 0.0642152230720967, 0.06465872400440276, 0.06441986002027988, 0.06447238498367369, 0.0653788719791919, 0.06306646298617125, 0.06400911905802786, 0.06430619279854, 0.06417124904692173, 0.06531375995837152, 0.06358011113479733, 0.06435781088657677, 0.0639402661472559, 0.0651675530243665, 0.06329983798786998, 0.06431976286694407, 0.06391515699215233, 0.06471114209853113, 0.0639546390157193, 0.06414170586504042, 0.06431174208410084, 0.0639334989245981, 0.06408315105363727, 0.06396008608862758, 0.0638457618188113, 0.06436604401096702, 0.06476330710574985, 0.06318638799712062, 0.06379674188792706, 0.06407001498155296, 0.06415620911866426, 0.06412742007523775, 0.06464639795012772, 0.0643034118693322, 0.06408118503168225, 0.06414252100512385, 0.06387999514117837, 0.06430621584877372, 0.0642301810439676, 0.06444918201304972, 0.06397688505239785, 0.06413654796779156, 0.06411438807845116, 0.06418520398437977, 0.06425123894587159, 0.06419241288676858, 0.0639674011617899, 0.06394167803227901, 0.06435677479021251, 0.06417148117907345, 0.0641366618219763, 0.06421323609538376, 0.06365998205728829, 0.06419800594449043, 0.06409383891150355, 0.06415582518093288, 0.0649532899260521, 0.06423849309794605, 0.06492144288495183, 0.06358578312210739, 0.06436405680142343, 0.06297661014832556, 0.06392489583231509, 0.06469204416498542, 0.06440518703311682, 0.0638858019374311, 0.07080410607159138, 0.06560529698617756, 0.06438799388706684, 0.06423378805629909, 0.06451804703101516, 0.06430077995173633, 0.06416533188894391, 0.06396322115324438, 0.06458982196636498, 0.06409001699648798, 0.06436408893205225, 0.06552436901256442, 0.06702823308296502, 0.0612056830432266, 0.06902325199916959, 0.06632461887784302, 0.07272283593192697, 0.06576541415415704, 0.06514666695147753, 0.06518849497660995, 0.06525884405709803, 0.06527877692133188, 0.06507101911120117, 0.06537044094875455, 0.06520353304222226, 0.06518860999494791, 0.06518154288642108, 0.06509707006625831, 0.06511741504073143, 0.06528706196695566, 0.0651822560466826, 0.0653154100291431, 0.06535262684337795, 0.06537231104448438, 0.06535704107955098, 0.06518053985200822, 0.06525627919472754, 0.06532691093161702, 0.06529849697835743, 0.06508804694749415, 0.06571439001709223, 0.0648966059088707, 0.06528786313720047, 0.06533694989047945, 0.06538592115975916, 0.06549673294648528, 0.0648863110691309, 0.06520955683663487, 0.06513495906256139, 0.06513684010133147, 0.06540242186747491, 0.0651245079934597, 0.06508389208465815, 0.06507038488052785, 0.06505430117249489, 0.0660177138634026, 0.06567092495970428, 0.06719908397644758, 0.06346698012202978, 0.0651655700057745, 0.06509266095235944, 0.06556501192972064, 0.06525432202033699, 0.0652575611602515, 0.06508626998402178, 0.06495514488779008, 0.06523492303676903, 0.06551456707529724, 0.06528043700382113, 0.0654320060275495, 0.06509732198901474, 0.06506546190939844, 0.06509764911606908, 0.06549190287478268, 0.0653578129131347, 0.06518045114353299, 0.06516457907855511, 0.06577666196972132, 0.0649244268424809, 0.06547288317233324, 0.06553458492271602, 0.06515390588901937, 0.06500718113966286, 0.06499590491876006, 0.06562955607660115, 0.06548322504386306, 0.06536699691787362, 0.06675983709283173], [0.04749089595861733, 0.04820419708266854, 0.04695946606807411, 0.04732451098971069, 0.04751124698668718, 0.047597723081707954, 0.047553107840940356, 0.04744734801352024, 0.04787139198742807, 0.046720544109120965, 0.04776188591495156, 0.04691097512841225, 0.047288435976952314, 0.0473920488730073, 0.0490937780123204, 0.04570717108435929, 0.04754844098351896, 0.047540012979879975, 0.04787248489446938, 0.047276158118620515, 0.04723345208913088, 0.047316038981080055, 0.04753902484662831, 0.04741875105537474, 0.047202520072460175, 0.047346796840429306, 0.04731922899372876, 0.04720850917510688, 0.0475594038143754, 0.047728448174893856, 0.04897447000257671, 0.04724964592605829, 0.04747792799025774, 0.047341223107650876, 0.04829830280505121, 0.047372339060530066, 0.047304920153692365, 0.047451463993638754, 0.047879477962851524, 0.04682055092416704, 0.04742998001165688, 0.04689014796167612, 0.04745847498998046, 0.04718349198810756, 0.04721505404449999, 0.047163964016363025, 0.04736842610873282, 0.047291504917666316, 0.04722196701914072, 0.047433862928301096, 0.04732201201841235, 0.04743790696375072, 0.04752275510691106, 0.04709732695482671, 0.047259703977033496, 0.04750335309654474, 0.04737064498476684, 0.047394916880875826, 0.047215416096150875, 0.04733420698903501, 0.04792552697472274, 0.051765614887699485, 0.062116452027112246, 0.06398453912697732, 0.0643564909696579, 0.0644723060540855, 0.06475428282283247, 0.0647627820726484, 0.06457180506549776, 0.06451772898435593, 0.06442914088256657, 0.06460469611920416, 0.06456867791712284, 0.06444124015979469, 0.06405470496974885, 0.0645427939016372, 0.06427058205008507, 0.06544697703793645, 0.06429669982753694, 0.06432970403693616, 0.06429290515370667, 0.0646296557970345, 0.06431967508979142, 0.06422796007245779, 0.06524566886946559, 0.06425702618435025, 0.06443857494741678, 0.06445109890773892, 0.06416542013175786, 0.06438460689969361, 0.06461473205126822, 0.06507355789653957, 0.06424949201755226, 0.06424983614124358, 0.06435468187555671, 0.06438428396359086, 0.06476069404743612, 0.06394929508678615, 0.06418249988928437, 0.06425812095403671, 0.06390196620486677, 0.0643346079159528, 0.06427009287290275, 0.0647566020488739, 0.07597061293199658, 0.05236017610877752, 0.0689736979547888, 0.06570014311000705, 0.06471365201286972, 0.06473317602649331, 0.0646757478825748, 0.06451884913258255, 0.06475452496670187, 0.06454548495821655, 0.06489854888059199, 0.0649921630974859, 0.06408584606833756, 0.0647222138941288, 0.06475495407357812, 0.06492035184055567, 0.06455010105855763, 0.06460221414454281, 0.06478647491894662, 0.06468535494059324, 0.06471836706623435, 0.06491995393298566, 0.06492443406023085, 0.06450247787870467, 0.06461727502755821, 0.0643600809853524, 0.06481674406677485, 0.06471661408431828, 0.06495097698643804, 0.06519833905622363, 0.06473982101306319, 0.06472923583351076, 0.06469717808067799, 0.06436525098979473, 0.06497084209695458, 0.06471261382102966, 0.06467859307304025, 0.06465621409006417, 0.06489545595832169, 0.06472548400051892, 0.06504528690129519, 0.06483417912386358, 0.06460440182127059, 0.06530900998041034, 0.06506600114516914, 0.06498183286748827, 0.06503072311170399, 0.06469826004467905, 0.06481958902440965, 0.06513320002704859, 0.06466380297206342, 0.06456850888207555, 0.06496874894946814, 0.06496096518822014, 0.06459864601492882, 0.06453569582663476, 0.0646337519865483, 0.06479775602929294, 0.06491258111782372, 0.06512149493210018, 0.06493587791919708, 0.06476103002205491, 0.06490776804275811, 0.06461161491461098, 0.06477104499936104, 0.06469242018647492, 0.06455789599567652, 0.06483452301472425, 0.06478395382873714, 0.06461896398104727, 0.06519041210412979, 0.06463473290205002, 0.0646303182002157, 0.06443785782903433, 0.06484355800785124, 0.06474809208884835, 0.06502098101191223, 0.06468380498699844, 0.06449116789735854, 0.06442149495705962, 0.06472799414768815, 0.06476502493023872, 0.06493013794533908, 0.06446852907538414, 0.06477778195403516, 0.0645866000559181, 0.06484635593369603, 0.06585760600864887, 0.06505241198465228, 0.06470385403372347, 0.06476535904221237, 0.0647300819400698, 0.06462668115273118, 0.06454306887462735, 0.06494621792808175, 0.06472271913662553, 0.0649879879783839, 0.06443891092203557, 0.06463861302472651, 0.06479297811165452, 0.06504569901153445, 0.0648172318469733, 0.06459975009784102, 0.06462623202241957, 0.06460440088994801, 0.06460334802977741, 0.06508509791456163, 0.06475030910223722, 0.06456303899176419, 0.06442749709822237, 0.06531415786594152, 0.06475689704529941, 0.06501925503835082, 0.06592512386851013, 0.06495930603705347, 0.0641692120116204, 0.06469625816680491, 0.0647510897833854, 0.0641307202167809, 0.06451842794194818, 0.06414332590065897, 0.06443465710617602, 0.06445221789181232, 0.0637754169292748, 0.0643348281737417, 0.06441584997810423, 0.06403535092249513, 0.06473731808364391, 0.06419872189871967, 0.06433576997369528, 0.06462548114359379, 0.06425193091854453, 0.06491218111477792, 0.06405785586684942, 0.06378633202984929, 0.06436548894271255, 0.06433155504055321, 0.06429438106715679, 0.06422535493038595, 0.0641476281452924, 0.06422394281253219, 0.06391130411066115, 0.06447256985120475, 0.06440277118235826, 0.0652827899903059, 0.06492071785032749, 0.06539098615758121, 0.06498059397563338, 0.06549300299957395, 0.06474009389057755, 0.06429456011392176, 0.06438773893751204, 0.06424198998138309, 0.06402986100874841, 0.06461189896799624, 0.06397888716310263, 0.06448907684534788, 0.06410090206190944, 0.06415475788526237, 0.0638081741053611, 0.06448796903714538, 0.06417091889306903, 0.06442158995196223, 0.06425215303897858, 0.06409369315952063, 0.0642880629748106, 0.06453009299002588, 0.06406856304965913, 0.06448762188665569, 0.06374229304492474, 0.06400544894859195, 0.06387358694337308, 0.06436489196494222, 0.0639913841150701, 0.06414463510736823, 0.06406470597721636, 0.06408612383529544, 0.06402086303569376, 0.06422178307548165, 0.06401658896356821, 0.06435806397348642, 0.06422862410545349, 0.06404801993630826, 0.06383394589647651, 0.0644789740908891, 0.06413973704911768, 0.06428262987174094, 0.06401394517160952, 0.06426595081575215, 0.06425799219869077, 0.0642681629396975, 0.06397177488543093, 0.06407738500274718, 0.06390106095932424, 0.06420995318330824, 0.06389713799580932, 0.0642176738474518, 0.06408472708426416, 0.06415055599063635, 0.06424945406615734, 0.0641682029236108, 0.06412920006550848, 0.06484688306227326, 0.0637423568405211, 0.06415005004964769, 0.06422029691748321, 0.06452626711688936, 0.0641603460535407, 0.06447127694264054, 0.0642844180110842, 0.06413019285537302, 0.06384152709506452, 0.06420807889662683, 0.06415926502086222, 0.06471255305223167, 0.06439480814151466, 0.06453123781830072, 0.06528933509252965, 0.06346468487754464, 0.06357354717329144, 0.06439189682714641, 0.06407916103489697, 0.06461369316093624, 0.0638812929391861, 0.06411881791427732, 0.06413163803517818, 0.06421333807520568, 0.06420719902962446, 0.0641660369001329, 0.06413341406732798, 0.06441711401566863, 0.0639701730106026, 0.06436424097046256, 0.0643627920653671, 0.06417122879065573, 0.06422394514083862, 0.06386359804309905, 0.06395310186780989, 0.06438255100511014, 0.06385230901651084, 0.06408187095075846, 0.06397272297181189, 0.06394772906787694, 0.06413806392811239, 0.06411576410755515, 0.0646645650267601, 0.06429491890594363, 0.06409844616428018, 0.06405678694136441, 0.06399511592462659, 0.06424159510061145, 0.06430739001370966, 0.06440448295325041, 0.06390467588789761, 0.06420951010659337, 0.06406492693349719, 0.06418953812681139, 0.06426401995122433, 0.0641915719024837, 0.06392190395854414, 0.06419951003044844, 0.06410206598229706, 0.06416730000637472, 0.06419109203852713, 0.06419030204415321, 0.06366418697871268, 0.06421956792473793, 0.06404612399637699, 0.06417245115153491, 0.06503954390063882, 0.06415316904895008, 0.06460080202668905, 0.06384650804102421, 0.06386363483034074, 0.06376015511341393, 0.06398195587098598, 0.06441131420433521, 0.06441008998081088, 0.06391496490687132, 0.0707564780022949, 0.06564132194034755, 0.06438368209637702, 0.06422089505940676, 0.06447442900389433, 0.06440974096767604, 0.0640019690617919, 0.06412366987206042, 0.06459887512028217, 0.06407926394604146, 0.06432167091406882, 0.06520842993631959, 0.06751093408092856, 0.06114091910421848, 0.06913091382011771, 0.06622294802218676, 0.07272912398912013, 0.06589698209427297, 0.06515159807167947, 0.06519063981249928, 0.06526268203742802, 0.06528847012668848, 0.06506034499034286, 0.06525556184351444, 0.06521549704484642, 0.06515656714327633, 0.06534373201429844, 0.06507357489317656, 0.06511141289956868, 0.06527819903567433, 0.0651374461594969, 0.06539609981700778, 0.06527862814255059, 0.06539150094613433, 0.06524975108914077, 0.0653314278461039, 0.06522858003154397, 0.06532337912358344, 0.06527001084759831, 0.06508820899762213, 0.06576840206980705, 0.06487890100106597, 0.0653042970225215, 0.06532780500128865, 0.06553599401377141, 0.06534890597686172, 0.06487869005650282, 0.06521279690787196, 0.06506238807924092, 0.06516937306150794, 0.06539475987665355, 0.0651534479111433, 0.06507423403672874, 0.06507884105667472, 0.06505488092079759, 0.06601662700995803, 0.0656759450212121, 0.06518379994668067, 0.06544430600479245, 0.06516655720770359, 0.06521149189211428, 0.06547906808555126, 0.06521197594702244, 0.06524830590933561, 0.06509925797581673, 0.06496750400401652, 0.06527053797617555, 0.06547977519221604, 0.06530836992897093, 0.06540987291373312, 0.06506833899766207, 0.06511255400255322, 0.06509957113303244, 0.06546595203690231, 0.06535937287844718, 0.06519339699298143, 0.06513499096035957, 0.06582008209079504, 0.06489601195789874, 0.06547010806389153, 0.06552529684267938, 0.0651203510351479, 0.06504198908805847, 0.0649919060524553, 0.06562983989715576, 0.06545994500629604, 0.0654085329733789, 0.06598862796090543, 0.07508558104746044, 0.0745921500492841, 0.06427051802165806, 0.06467146496288478, 0.0643193970900029, 0.06449496396817267, 0.06435445393435657, 0.06477702502161264, 0.06454798090271652, 0.06458388501778245, 0.06433941703289747, 0.06468979199416935, 0.06432430609129369, 0.06456272699870169, 0.06450273701921105, 0.06459157494828105, 0.06466868589632213, 0.06445605610497296, 0.06455853208899498, 0.0645664818584919, 0.06440928601659834, 0.06458040000870824, 0.06450057798065245, 0.06472738203592598, 0.06430199788883328, 0.06450350116938353, 0.06429115403443575, 0.06465567392297089, 0.06459611793980002, 0.06428335094824433, 0.06439676810987294, 0.06443529692478478, 0.06453264714218676, 0.06442325189709663, 0.064406932098791, 0.06459765089675784, 0.06463044905103743, 0.0651954710483551, 0.0654473640024662], [0.06423419690690935, 0.06462356704287231, 0.06450683693401515, 0.06480712397024035, 0.06477194302715361, 0.06454838905483484, 0.06447956105694175, 0.06452600006014109, 0.06455680285580456, 0.06452452205121517, 0.06439214199781418, 0.06420487002469599, 0.06452532787807286, 0.06428409717045724, 0.0654806918464601, 0.06423834199085832, 0.06445579114370048, 0.06425951793789864, 0.06464369501918554, 0.0643244069069624, 0.06418704614043236, 0.06530314101837575, 0.06426275498233736, 0.06453551282174885, 0.0643494960386306, 0.0641550780273974, 0.06439858209341764, 0.06459226901642978, 0.06509832595475018, 0.06426383298821747, 0.06424431805498898, 0.0643302509561181, 0.06492446106858552, 0.06420612800866365, 0.0639885407872498, 0.06417664512991905, 0.06426810193806887, 0.06388331693597138, 0.06439077109098434, 0.06431996007449925, 0.0648141058627516, 0.07563440804369748, 0.052291525062173605, 0.06900645792484283, 0.06562616908922791, 0.06468814285472035, 0.06465385504998267, 0.06477110809646547, 0.06455614790320396, 0.06470351410098374, 0.06457568402402103, 0.06489689694717526, 0.06496563390828669, 0.06403343612328172, 0.06483746087178588, 0.0647351760417223, 0.06495236908085644, 0.0645442008972168, 0.06458104704506695, 0.06482283701188862, 0.06468119309283793, 0.06467817793600261, 0.0648843259550631, 0.06497000204399228, 0.06443787994794548, 0.06472389097325504, 0.06436624610796571, 0.06478385184891522, 0.06471624202094972, 0.06468216795474291, 0.06549258320592344, 0.06470012990757823, 0.06473905593156815, 0.06469871918670833, 0.06435481389053166, 0.06502840388566256, 0.06472777621820569, 0.06465458287857473, 0.06462956010363996, 0.06490669585764408, 0.06471977196633816, 0.06509381998330355, 0.06479357299394906, 0.06461565615609288, 0.0652991640381515, 0.06508520897477865, 0.06493939785286784, 0.065053875092417, 0.06471341499127448, 0.06481573591008782, 0.0651192800141871, 0.06469063507393003, 0.06453664205037057, 0.0649788249284029, 0.0649383480194956, 0.06462979991920292, 0.06454881210811436, 0.06462198588997126, 0.06478708004578948, 0.06490863999351859, 0.06513072201050818, 0.06491818302311003, 0.06475548492744565, 0.06491752900183201, 0.06462458800524473, 0.06528257415629923, 0.06418616999872029, 0.06459069903939962, 0.06481479294598103, 0.06477414490655065, 0.06461776699870825, 0.06520875310525298, 0.06462305900640786, 0.06460372498258948, 0.0644782679155469, 0.06481308303773403, 0.06476003397256136, 0.06503073708154261, 0.06465776287950575, 0.06449720705859363, 0.06445357901975513, 0.06474747904576361, 0.0647298579569906, 0.06495169107802212, 0.06443208991549909, 0.06461340095847845, 0.06476980308070779, 0.06486816704273224, 0.06584521499462426, 0.06506573082879186, 0.0646925091277808, 0.06475508189760149, 0.06472207396291196, 0.06462708907201886, 0.06454232102259994, 0.06495574302971363, 0.06476560793817043, 0.06497243512421846, 0.06440005195327103, 0.0646889319177717, 0.06477067898958921, 0.06557956105098128, 0.06418165704235435, 0.06471441104076803, 0.06461376883089542, 0.06460219714790583, 0.06460218201391399, 0.06514398287981749, 0.06468215794302523, 0.06453628418967128, 0.06445344793610275, 0.06532297190278769, 0.06479832110926509, 0.06500817090272903, 0.06590262800455093, 0.06588873104192317, 0.06322179199196398, 0.06469392706640065, 0.06486562988720834, 0.06421284703537822, 0.06449361494742334, 0.06407544715330005, 0.0644671160262078, 0.06444528885185719, 0.06378426705487072, 0.06439587287604809, 0.06438961322419345, 0.0640589389950037, 0.06466336478479207, 0.0642154491506517, 0.06435613590292633, 0.06461035599932075, 0.06427611713297665, 0.06488834298215806, 0.06406316999346018, 0.06380985304713249, 0.06436151079833508, 0.06433667405508459, 0.06430545006878674, 0.06421126192435622, 0.06417544395662844, 0.06452780705876648, 0.0636178229469806, 0.06440296303480864, 0.06444340315647423, 0.06524404091760516, 0.06495799403637648, 0.06537783099338412, 0.06497487891465425, 0.0654977101366967, 0.064774997998029, 0.06432574382051826, 0.06430543214082718, 0.0643227999098599, 0.063973484095186, 0.06465441593900323, 0.0639547579921782, 0.06451064301654696, 0.06408787705004215, 0.06418163585476577, 0.06379402801394463, 0.0644905490335077, 0.06413814797997475, 0.06442777696065605, 0.06429121200926602, 0.06408457714132965, 0.0643172028940171, 0.06457866402342916, 0.06395278801210225, 0.06449461495503783, 0.06371417408809066, 0.06397607107646763, 0.06397932791151106, 0.06434343405999243, 0.06399804982356727, 0.06420979299582541, 0.06400967505760491, 0.06408837414346635, 0.06407826184295118, 0.06413733796216547, 0.06401319219730794, 0.0643151979893446, 0.06430809781886637, 0.06405933015048504, 0.0638158458750695, 0.06449242704547942, 0.06412707711569965, 0.06404266296885908, 0.06425370485521853, 0.06423867703415453, 0.06424454413354397, 0.06434085592627525, 0.06396779790520668, 0.06410400499589741, 0.06391694000922143, 0.06414207210764289, 0.06395825301297009, 0.06417027302086353, 0.0640836728271097, 0.0641536870971322, 0.06425560708157718, 0.06421311991289258, 0.06401050393469632, 0.06490961718373, 0.06375044281594455, 0.06415467197075486, 0.0642528492026031, 0.06445605191402137, 0.06418880494311452, 0.06445816997438669, 0.06428556400351226, 0.06409591599367559, 0.06388801708817482, 0.06419375608675182, 0.06403821893036366, 0.06484112585894763, 0.0644113551825285, 0.06447394704446197, 0.06538178282789886, 0.06314681097865105, 0.06384464516304433, 0.06438469584099948, 0.06409781007096171, 0.06523713492788374, 0.06363498605787754, 0.06430457206442952, 0.06400419701822102, 0.06437245290726423, 0.06412932393141091, 0.064325146144256, 0.06386397685855627, 0.06467235903255641, 0.06396000110544264, 0.06413915194571018, 0.06433692295104265, 0.06389569211751223, 0.06421433086507022, 0.06387346400879323, 0.06387309497222304, 0.06432857713662088, 0.06473271502181888, 0.0633243469055742, 0.06414434709586203, 0.06372539489530027, 0.06415577814914286, 0.06412871787324548, 0.06464081001468003, 0.06430354993790388, 0.064085747115314, 0.06393537297844887, 0.06408281088806689, 0.06423028698191047, 0.06430634018033743, 0.06445456389337778, 0.06389052397571504, 0.06422272394411266, 0.06409032316878438, 0.06420503300614655, 0.06424976699054241, 0.06419361406005919, 0.0638876857701689, 0.06428817403502762, 0.06407985812984407, 0.06410435400903225, 0.06421342003159225, 0.06420865096151829, 0.06366418185643852, 0.06419790908694267, 0.06401152000762522, 0.06410608207806945, 0.06507026683539152, 0.06425071298144758, 0.06487562507390976, 0.06348498910665512, 0.06444851588457823, 0.06275472301058471, 0.06420055613853037, 0.06469774479046464, 0.06440574117004871, 0.0638839069288224, 0.07072956091724336, 0.06567790801636875, 0.06439022906124592, 0.06423350004479289, 0.0644298829138279, 0.06439451710321009, 0.06403940287418664, 0.06408207607455552, 0.06459527905099094, 0.06408059294335544, 0.06429096404463053, 0.06558079388923943, 0.06703785504214466, 0.060876568080857396, 0.06934520602226257, 0.06603303994052112, 0.07274751993827522, 0.06604883493855596, 0.06514377519488335, 0.06519152293913066, 0.0652521529700607, 0.06528352200984955, 0.06505745998583734, 0.0651170399505645, 0.06534633715637028, 0.06500175292603672, 0.0654816220048815, 0.06510774395428598, 0.06511345808394253, 0.06527996785007417, 0.06518293009139597, 0.06532471301034093, 0.06534750387072563, 0.06537246215157211, 0.06509410706348717, 0.06544000492431223, 0.06512921000830829, 0.06545914593152702, 0.06529977708123624, 0.065084557980299, 0.06571987085044384, 0.0648976971860975, 0.06528486195020378, 0.06533463089726865, 0.06569941504858434, 0.0651802709326148, 0.06488799420185387, 0.0652115079574287, 0.06513822986744344, 0.0651364210061729, 0.06540276110172272, 0.06512052589096129, 0.0650837121065706, 0.06506577390246093, 0.06505692610517144, 0.06601695204153657, 0.06567310285754502, 0.06708273012191057, 0.06357505498453975, 0.06516105704940856, 0.06508883880451322, 0.06557472608983517, 0.06524794292636216, 0.06527122808620334, 0.06508599896915257, 0.06495671207085252, 0.06522260000929236, 0.06552117387764156, 0.065281982999295, 0.06543177412822843, 0.06510005705058575, 0.06506394897587597, 0.06509733386337757, 0.06549223000183702, 0.06535593303851783, 0.0651791871059686, 0.06514649395830929, 0.06579800299368799, 0.06491636601276696, 0.06547581986524165, 0.06553450901992619, 0.06513985013589263, 0.06500775995664299, 0.06501301494427025, 0.06562584405764937, 0.06546367798000574, 0.06538820592686534, 0.06643898412585258, 0.07457297388464212, 0.07441765395924449, 0.06427690898999572, 0.06461895513348281, 0.0643489989452064, 0.0644904060754925, 0.06434524687938392, 0.06482030195184052, 0.06472879112698138, 0.0645880838856101, 0.06433552503585815, 0.06468621408566833, 0.06430240208283067, 0.06459673284552991, 0.06446789903566241, 0.0646042451262474, 0.06470075296238065, 0.06441890588030219, 0.06457195500843227, 0.06456347508355975, 0.06440133298747241, 0.06460888590663671, 0.06448934902437031, 0.06486130808480084, 0.06416184292174876, 0.06432671891525388, 0.06424974603578448, 0.06466469704173505, 0.06451060506515205, 0.06436057901009917, 0.06438896991312504, 0.06445379508659244, 0.06454335199669003, 0.0644071688875556, 0.06439781398512423, 0.06461908505298197, 0.06445000902749598, 0.06571831507608294, 0.06519096787087619, 0.06684140209108591, 0.07441556500270963, 0.07036598189733922, 0.0647182350512594, 0.06474573793821037, 0.06432275008410215, 0.06553118699230254, 0.06474390509538352, 0.06474877893924713, 0.06434156093746424, 0.064388289116323, 0.0646529698278755, 0.06457851198501885, 0.06443838216364384, 0.06449030386283994, 0.06438856897875667, 0.06454952200874686, 0.06449417606927454, 0.06448759394697845, 0.06482133199460804, 0.06459809397347271, 0.06424085516482592, 0.06440396490506828, 0.06450871005654335, 0.06480874889530241, 0.06533838715404272, 0.0637410779017955, 0.06501748505979776, 0.0645234058611095, 0.06448218715377152, 0.06501047406345606, 0.06446711393073201, 0.06449658586643636, 0.06428841198794544, 0.06435008603148162, 0.06464873114600778, 0.06476228591054678, 0.06443902710452676, 0.06456737383268774, 0.0647934500593692, 0.06425178609788418, 0.06467858096584678, 0.06462041405029595, 0.06472572684288025, 0.06447043200023472, 0.06429409305565059, 0.06462151696905494, 0.06484490307047963, 0.0645684979390353, 0.06448813294991851, 0.06452349107712507, 0.06487479410134256, 0.06445959396660328, 0.06450493703596294, 0.06473175995051861, 0.06459161383099854, 0.06460840813815594, 0.06446781288832426, 0.0649099771399051, 0.06443360284902155, 0.06478709308430552, 0.06446490902453661, 0.06558015989139676], [0.06455103191547096, 0.06438113818876445, 0.06493921997025609, 0.06477429787628353, 0.06454878905788064, 0.06436304096132517, 0.06463377317413688, 0.06455900287255645, 0.06440292694605887, 0.06439997116103768, 0.06430431082844734, 0.064533906057477, 0.06428612791933119, 0.06548502203077078, 0.06409579608589411, 0.06458317092619836, 0.06426035007461905, 0.06463963608257473, 0.06433056690730155, 0.06418836093507707, 0.06529553513973951, 0.06426264997571707, 0.06446294602937996, 0.06443039490841329, 0.06414484302513301, 0.0643973289988935, 0.06459570187143981, 0.06509652105160058, 0.06412782100960612, 0.0643733839970082, 0.0643365120049566, 0.06485024094581604, 0.06426415010355413, 0.06400207197293639, 0.06416860013268888, 0.06426992197521031, 0.06389082502573729, 0.06431034882552922, 0.06430524704046547, 0.06477014813572168, 0.07561969384551048, 0.05231725797057152, 0.06890815799124539, 0.06573467701673508, 0.06469711009413004, 0.06476083700545132, 0.06465206597931683, 0.06465606903657317, 0.06461102305911481, 0.06456437497399747, 0.06489338399842381, 0.06494084582664073, 0.06413908605463803, 0.06474240310490131, 0.06474285386502743, 0.06493996316567063, 0.06467683287337422, 0.06458091898821294, 0.06482096994295716, 0.06467312504537404, 0.06468873005360365, 0.06488411710597575, 0.06477730697952211, 0.06448619882576168, 0.06472611101344228, 0.06428024801425636, 0.06488666706718504, 0.06471611908636987, 0.06468579079955816, 0.06548073119483888, 0.0648299059830606, 0.06461870484054089, 0.0646923491731286, 0.0643582500051707, 0.06501204683445394, 0.06486383499577641, 0.06452953303232789, 0.06463481206446886, 0.06490260409191251, 0.0647096789907664, 0.06509486585855484, 0.06479329103603959, 0.06460771197453141, 0.065302963135764, 0.06521764281205833, 0.0649363670963794, 0.06504692300222814, 0.06457192194648087, 0.0648285390343517, 0.06512181810103357, 0.06468009599484503, 0.0645476789213717, 0.06496486207470298, 0.06490577501244843, 0.06467936397530138, 0.0645377750042826, 0.06463190889917314, 0.06478716991841793, 0.06502212909981608, 0.06501704291440547, 0.06492107617668808, 0.06475261482410133, 0.06492145406082273, 0.06461764709092677, 0.06520119495689869, 0.06422805599868298, 0.06459821993485093, 0.06495489994995296, 0.06464518513530493, 0.06461182399652898, 0.06521597690880299, 0.06462386203929782, 0.06460414710454643, 0.06449245288968086, 0.06481325393542647, 0.06475922418758273, 0.06502974382601678, 0.06466266815550625, 0.0646129660308361, 0.06432964792475104, 0.06473574694246054, 0.06469985307194293, 0.06497489591129124, 0.0645721850451082, 0.06474064709618688, 0.064508190844208, 0.06487283506430686, 0.06584395398385823, 0.06506765191443264, 0.0646954639814794, 0.06476222607307136, 0.06472393102012575, 0.0646179080940783, 0.06455204985104501, 0.0650656409561634, 0.06461620610207319, 0.06500314990989864, 0.0643895932007581, 0.06468912586569786, 0.06477956403978169, 0.06501526501961052, 0.06486285803839564, 0.06449898588471115, 0.06468793912790716, 0.06462219799868762, 0.0646031559444964, 0.06513763894326985, 0.06468929210677743, 0.0645312499254942, 0.06444916897453368, 0.06544895702973008, 0.06466962397098541, 0.06502147507853806, 0.06601260788738728, 0.06536688306368887, 0.06387247005477548, 0.0643183789215982, 0.06481451401486993, 0.06447916408069432, 0.06437265989370644, 0.06410413095727563, 0.064457027008757, 0.0644222900737077, 0.06392568699084222, 0.06438729888759553, 0.06438239011913538, 0.06408014590851963, 0.06466342299245298, 0.0641038971953094, 0.06447288999333978, 0.06461236299946904, 0.06428026081994176, 0.06502807419747114, 0.06389352795667946, 0.06381832994520664, 0.06436338205821812, 0.06433691293932498, 0.06428232695907354, 0.06424040696583688, 0.06417941302061081, 0.06453438615426421, 0.06361696496605873, 0.0644060019403696, 0.06433165306225419, 0.0652401999104768, 0.0649640669580549, 0.06537474109791219, 0.06498810090124607, 0.06549550616182387, 0.06476493994705379, 0.06443491391837597, 0.06416446599178016, 0.06430839304812253, 0.06411514687351882, 0.06464751809835434, 0.06397506012581289, 0.06446757493540645, 0.06412058300338686, 0.06418216787278652, 0.06379528413526714, 0.06448317202739418, 0.06414525490254164, 0.06442665494978428, 0.0642896390054375, 0.06407246016897261, 0.06417778390459716, 0.06458879890851676, 0.06390691408887506, 0.06453201686963439, 0.06373523315414786, 0.06398974894545972, 0.06408366491086781, 0.06433887407183647, 0.06399487308226526, 0.06420936691574752, 0.06401939014904201, 0.064064955804497, 0.064054872142151, 0.06417395104654133, 0.06401878781616688, 0.06430625496432185, 0.06430681818164885, 0.06405371404252946, 0.06383911380544305, 0.06447618710808456, 0.06413503992371261, 0.0641865381039679, 0.06409758306108415, 0.06425436097197235, 0.06422773282974958, 0.06432889914140105, 0.06398308603093028, 0.06407808000221848, 0.0639339848421514, 0.06412885710597038, 0.06393470684997737, 0.06422523618675768, 0.06408115802332759, 0.06415353482589126, 0.06412910018116236, 0.0642074488569051, 0.06414621113799512, 0.0648976459633559, 0.06375322584062815, 0.06414403510279953, 0.06426348793320358, 0.06445850897580385, 0.06419730512425303, 0.06442609801888466, 0.06430783588439226, 0.06410285807214677, 0.06387489102780819, 0.0641957400366664, 0.06407989794388413, 0.06468925392255187, 0.0643988330848515, 0.06447928608395159, 0.0653800149448216, 0.06335753202438354, 0.06363592599518597, 0.06437683897092938, 0.06411501788534224, 0.06481833709403872, 0.06389532401226461, 0.06410372606478631, 0.06426515197381377, 0.06437835702672601, 0.06422755378298461, 0.06411280715838075, 0.0639051990583539, 0.06467798282392323, 0.06392477499321103, 0.0642353140283376, 0.06448958301916718, 0.06389828398823738, 0.06418545613996685, 0.06378429499454796, 0.06387519487179816, 0.06434653513133526, 0.06456877989694476, 0.06348054599948227, 0.06422452908009291, 0.06395830283872783, 0.06415351899340749, 0.06413212604820728, 0.06447143107652664, 0.06447147694416344, 0.06408299691975117, 0.06392395915463567, 0.06409546500071883, 0.06423066696152091, 0.06430839188396931, 0.06445445609278977, 0.06388627295382321, 0.06422284315340221, 0.06409254693426192, 0.06418319791555405, 0.06427234201692045, 0.06419381010346115, 0.0638859118334949, 0.06452963408082724, 0.06373853608965874, 0.0642017018981278, 0.06421325192786753, 0.06420588120818138, 0.06365872384048998, 0.06420717714354396, 0.06401097681373358, 0.06435879599303007, 0.0652282100636512, 0.06383681506849825, 0.06480767205357552, 0.06381804402917624, 0.06410997989587486, 0.06309765507467091, 0.06359382788650692, 0.06471800501458347, 0.0644163100514561, 0.06388042191974819, 0.07073444500565529, 0.0658061180729419, 0.0642371759749949, 0.06439530802890658, 0.06425362499430776, 0.06456612003967166, 0.06390765286050737, 0.06420994317159057, 0.06446093390695751, 0.06421736790798604, 0.06415283819660544, 0.0656729219481349, 0.0672322700265795, 0.060878464952111244, 0.06919944891706109, 0.06617458094842732, 0.0727538620121777, 0.06605599517934024, 0.06514224782586098, 0.06519525218755007, 0.06515206489712, 0.06538551393896341, 0.06504718610085547, 0.06511765089817345, 0.0652178491000086, 0.06513247196562588, 0.0654815910384059, 0.06497146305628121, 0.06525091594085097, 0.06527189095504582, 0.06518040294758976, 0.06533850100822747, 0.06534301699139178, 0.06537091895006597, 0.0650931061245501, 0.06544363196007907, 0.06511660502292216, 0.0654748729430139, 0.06527579412795603, 0.06508216192014515, 0.06574050593189895, 0.06490113004110754, 0.06528325704857707, 0.06533352006226778, 0.0655280549544841, 0.06534566706977785, 0.06489125289954245, 0.06521618901751935, 0.06513519398868084, 0.06514509487897158, 0.06538808206096292, 0.06513330899178982, 0.06507950904779136, 0.06505745812319219, 0.06505985790863633, 0.06601970596238971, 0.06566891609691083, 0.06673379801213741, 0.0639184929896146, 0.06506516481749713, 0.06519401911646128, 0.06557314586825669, 0.06523841712623835, 0.06526192394085228, 0.06510719913057983, 0.06495647481642663, 0.06521404208615422, 0.06551625393331051, 0.06528794812038541, 0.06543281394988298, 0.06510111107490957, 0.06506418786011636, 0.06509914202615619, 0.06548908003605902, 0.06535280891694129, 0.06518534407950938, 0.06514050997793674, 0.06565495510585606, 0.06505920900963247, 0.06547518982551992, 0.06553693115711212, 0.06512283091433346, 0.06502368906512856, 0.0650146899279207, 0.06562118302099407, 0.06546156504191458, 0.06539582693949342, 0.06615650514140725, 0.07487174379639328, 0.07445412199012935, 0.06427760701626539, 0.06461888109333813, 0.0643516790587455, 0.06449117697775364, 0.0643446040339768, 0.06475444789975882, 0.06472696107812226, 0.06458743405528367, 0.06433861190453172, 0.06468687392771244, 0.06430632201954722, 0.0645864880643785, 0.0644841119647026, 0.06459752190858126, 0.0646952569950372, 0.06442451220937073, 0.06456748000346124, 0.06456878292374313, 0.0643972980324179, 0.06460666889324784, 0.06448259111493826, 0.06476343586109579, 0.0642690050881356, 0.06437678309157491, 0.06426554592326283, 0.06465962086804211, 0.06470965407788754, 0.06416175700724125, 0.06439223699271679, 0.06444904091767967, 0.06454723607748747, 0.06440310599282384, 0.06440053996630013, 0.06461723195388913, 0.06440723803825676, 0.0655896810349077, 0.06566241709515452, 0.06739085889421403, 0.07438103901222348, 0.07029197714291513, 0.06390841188840568, 0.06474813795648515, 0.0643208110705018, 0.06553765293210745, 0.06474154512397945, 0.06457590591162443, 0.06450890004634857, 0.06437670602463186, 0.0646651869174093, 0.0645067139994353, 0.06451633595861495, 0.06448871712200344, 0.06438871193677187, 0.06454170891083777, 0.06449316604994237, 0.06449435208924115, 0.0648254209663719, 0.06444074492901564, 0.06438375916332006, 0.06416953098960221, 0.06467504194006324, 0.06481132004410028, 0.06539933197200298, 0.0640953159891069, 0.06466370704583824, 0.06452619493938982, 0.06449230900034308, 0.0650005079805851, 0.0644650470931083, 0.06431611301377416, 0.06447143782861531, 0.06435889913700521, 0.06463751196861267, 0.06477538286708295, 0.06443508411757648, 0.06456817593425512, 0.0646280120126903, 0.06441345694474876, 0.06467503914609551, 0.06461929203942418, 0.06465592398308218, 0.06452764896675944, 0.06424135202541947, 0.06467854301445186, 0.06485918094404042, 0.06456174096092582, 0.06447738292627037, 0.06453231512568891, 0.06488205399364233, 0.06445662304759026, 0.06450508395209908, 0.06452834489755332, 0.06479714810848236, 0.0646028290502727, 0.0643994368147105, 0.06497946218587458, 0.0644321059808135, 0.06478714197874069, 0.06446677795611322, 0.06533481902442873, 0.06782844290137291], [0.06445798114873469, 0.06483073788695037, 0.06477208505384624, 0.064549206988886, 0.0644583790563047, 0.0645526668522507, 0.06455775001086295, 0.06448963400907815, 0.06439750106073916, 0.06420313706621528, 0.06456570699810982, 0.06427476787939668, 0.06548541714437306, 0.06420821091160178, 0.06444914010353386, 0.06426165392622352, 0.06463939696550369, 0.06433122395537794, 0.06423734012059867, 0.06525189406238496, 0.06427883286960423, 0.06444395892322063, 0.06442796718329191, 0.06416278192773461, 0.06439561187289655, 0.06459246622398496, 0.06509594479575753, 0.06427013617940247, 0.06423867191188037, 0.06434951908886433, 0.06501925690099597, 0.06411210191436112, 0.06397688714787364, 0.06417186884209514, 0.06427206215448678, 0.0638763690367341, 0.06431560800410807, 0.06431593792513013, 0.06491272198036313, 0.07564077689312398, 0.05228332616388798, 0.06891581998206675, 0.06570959696546197, 0.06468514492735267, 0.06476741610094905, 0.06466845399700105, 0.06454595597460866, 0.0647038440220058, 0.06458253599703312, 0.06489454791881144, 0.06498388200998306, 0.06414300506003201, 0.06472055800259113, 0.06473702401854098, 0.06494224886409938, 0.06453768606297672, 0.06458274694159627, 0.06482309708371758, 0.06469080504029989, 0.06472363299690187, 0.06488504889421165, 0.06483859498985112, 0.06450339802540839, 0.06474136002361774, 0.06427369709126651, 0.06487785605713725, 0.0647189209703356, 0.06459372583776712, 0.06558012100867927, 0.06470190919935703, 0.06473808898590505, 0.06469728099182248, 0.06436181883327663, 0.06502399314194918, 0.0647275319788605, 0.06466565793380141, 0.0646248459815979, 0.06490365997888148, 0.06472249794751406, 0.06509064510464668, 0.06479865312576294, 0.06461532297544181, 0.0653028218075633, 0.06508285901509225, 0.06493208999745548, 0.06506333313882351, 0.064718808978796, 0.06481388094834983, 0.0651153470389545, 0.06469574198126793, 0.06453804392367601, 0.064983778167516, 0.0649470139760524, 0.06461568083614111, 0.06454553618095815, 0.06461738096550107, 0.06479138694703579, 0.06490097590722144, 0.0651456390041858, 0.06491375900804996, 0.06475745304487646, 0.06491461815312505, 0.06463856692425907, 0.06531099998392165, 0.06415961403399706, 0.06458133389241993, 0.06481640902347863, 0.06478603510186076, 0.06460721092298627, 0.06521080201491714, 0.06461954908445477, 0.06460252380929887, 0.06447622110135853, 0.06482665194198489, 0.06474720314145088, 0.06503874901682138, 0.06466302694752812, 0.06448593502864242, 0.06445627799257636, 0.06474601500667632, 0.0647384359035641, 0.06494895298965275, 0.06442990899085999, 0.06462317192927003, 0.06476863706484437, 0.06485732807777822, 0.06585463602095842, 0.0650640670210123, 0.06470460002310574, 0.06473813392221928, 0.06472442205995321, 0.06463170796632767, 0.06453660898841918, 0.06495502800680697, 0.06477143499068916, 0.06496953102760017, 0.06440164102241397, 0.0646873158402741, 0.0647746161557734, 0.06560852099210024, 0.06415627291426063, 0.06471133092418313, 0.06461574812419713, 0.06459436495788395, 0.06460669101215899, 0.06503729801625013, 0.06479711690917611, 0.06453712400980294, 0.06445980304852128, 0.06530731893144548, 0.06480777100659907, 0.06500412407331169, 0.0659039809834212, 0.06601741001941264, 0.06311314809136093, 0.06457544583827257, 0.06498742708936334, 0.06422678497619927, 0.06450400990433991, 0.0641294561792165, 0.06434746296145022, 0.06448201206512749, 0.06377891381271183, 0.06441380199976265, 0.06436734297312796, 0.06405525607988238, 0.06466669193468988, 0.0642157350666821, 0.06435548607259989, 0.06461243890225887, 0.06427483097650111, 0.06489120214246213, 0.06407998991198838, 0.0637941260356456, 0.06436403095722198, 0.0643635441083461, 0.06427642796188593, 0.06421922892332077, 0.06416667392477393, 0.0645216591656208, 0.06361829303205013, 0.06445606192573905, 0.06444315402768552, 0.06528266496025026, 0.06494301301427186, 0.06535681989043951, 0.0649204021319747, 0.06554629607126117, 0.06477448786608875, 0.06416763295419514, 0.06430815113708377, 0.06444613402709365, 0.06396245793439448, 0.06466807285323739, 0.0639431681483984, 0.06454182602465153, 0.06405954994261265, 0.06418132595717907, 0.06380138802342117, 0.06447847397066653, 0.06414241716265678, 0.0644756609108299, 0.06427244399674237, 0.06415345799177885, 0.06427518394775689, 0.06453090510331094, 0.06405859882943332, 0.06447766604833305, 0.06368790310807526, 0.06399030005559325, 0.06389972078613937, 0.0643865130841732, 0.06397843197919428, 0.0641876719892025, 0.06401093094609678, 0.06413585203699768, 0.06403188896365464, 0.06414150912314653, 0.06406119093298912, 0.06433259998448193, 0.06424968410283327, 0.06405777600593865, 0.06387030286714435, 0.06444059312343597, 0.06412448105402291, 0.06416633888147771, 0.06418007495813072, 0.06423580017872155, 0.06420346186496317, 0.06433091009967029, 0.06396949500776827, 0.06409887899644673, 0.06392445601522923, 0.06418976187705994, 0.06391629693098366, 0.06416797800920904, 0.06407808000221848, 0.06415344798006117, 0.06425601220689714, 0.0642274518031627, 0.06404722621664405, 0.06486831791698933, 0.0637462108861655, 0.0641572221647948, 0.06425400986336172, 0.06445387308485806, 0.06424009893089533, 0.06440437212586403, 0.06428988790139556, 0.06414358504116535, 0.0638406400103122, 0.0641979540232569, 0.06421348988078535, 0.06465751002542675, 0.06442166212946177, 0.06452986178919673, 0.06536877015605569, 0.06295224698260427, 0.06408793199807405, 0.06434589787386358, 0.06412457209080458, 0.06546149402856827, 0.06353584607131779, 0.06428811186924577, 0.0639570290222764, 0.06546148401685059, 0.06299637607298791, 0.06429657503031194, 0.06396445678547025, 0.06475242599844933, 0.06396445515565574, 0.06409024400636554, 0.06434496794827282, 0.0635632649064064, 0.06438201805576682, 0.06394557701423764, 0.06385455094277859, 0.06435891101136804, 0.06478858296759427, 0.06315083708614111, 0.0640100589953363, 0.06389691401273012, 0.06416706810705364, 0.06406965479254723, 0.06469887401908636, 0.06430510617792606, 0.06407823786139488, 0.06410270603373647, 0.06392162293195724, 0.06426888308487833, 0.06426310911774635, 0.06447117798961699, 0.0639217880088836, 0.06419312278740108, 0.06404371513053775, 0.06419112207368016, 0.06430053198710084, 0.06419851584360003, 0.06391865899786353, 0.06417400995269418, 0.06412008916959167, 0.06417250097729266, 0.06418342795222998, 0.06416255212388933, 0.06365972990170121, 0.0642469939775765, 0.06406265590339899, 0.06414143205620348, 0.06508223596028984, 0.06411033309996128, 0.06497219810262322, 0.06353510986082256, 0.06459279707632959, 0.06246917485259473, 0.06420955318026245, 0.06473738583736122, 0.06440411997027695, 0.0638879460748285, 0.07075924915261567, 0.06566555798053741, 0.06436807685531676, 0.06423964002169669, 0.06448046397417784, 0.0643376971129328, 0.06383947492577136, 0.06428501498885453, 0.06459562107920647, 0.06408547889441252, 0.06432574614882469, 0.06558126397430897, 0.06696832901798189, 0.06086638499982655, 0.06936352280899882, 0.06586021999828517, 0.07290452416054904, 0.0660385109949857, 0.06515422300435603, 0.06518567004241049, 0.06526059587486088, 0.06527651892974973, 0.06507193902507424, 0.06508535798639059, 0.06520807300694287, 0.0651380370836705, 0.06551355798728764, 0.06509093497879803, 0.06512428307905793, 0.06528955791145563, 0.0651799060869962, 0.06531412084586918, 0.06535507598891854, 0.06537214713171124, 0.06506781396456063, 0.06547061097808182, 0.06525505101308227, 0.06532554002478719, 0.06530139991082251, 0.06508820201270282, 0.06571241514757276, 0.06491571688093245, 0.0652702460065484, 0.06533541902899742, 0.06554075097665191, 0.06534564704634249, 0.06488407007418573, 0.06521065393462777, 0.06513504195027053, 0.06513218302279711, 0.06540574110113084, 0.06512309284880757, 0.06508577708154917, 0.06507561798207462, 0.06505111092701554, 0.0660162961576134, 0.0656714579090476, 0.0672657189425081, 0.0634086960926652, 0.06516300304792821, 0.06514272792264819, 0.06550961500033736, 0.06526333489455283, 0.0652488621417433, 0.06508476100862026, 0.06495678098872304, 0.06523699196986854, 0.06551410793326795, 0.06528253410942852, 0.06543103791773319, 0.06509831105358899, 0.06506469706073403, 0.06509755784645677, 0.06549303117208183, 0.06535852188244462, 0.06517882691696286, 0.06516660004854202, 0.06577424402348697, 0.06492914608679712, 0.06547205592505634, 0.06553543591871858, 0.06515554315410554, 0.0650141378864646, 0.06498469016514719, 0.06562775699421763, 0.06548793381080031, 0.06536745210178196, 0.06633241288363934, 0.07449382217600942, 0.07438542391173542, 0.06427717604674399, 0.06461853696964681, 0.06434823991730809, 0.06448938697576523, 0.06434482405893505, 0.06503027095459402, 0.06458260398358107, 0.06458299001678824, 0.06433422002010047, 0.06468395702540874, 0.06430830503813922, 0.06459287204779685, 0.0644647809676826, 0.06462110602296889, 0.06469212588854134, 0.06441376218572259, 0.06457678298465908, 0.06455800984986126, 0.06440365011803806, 0.06461286102421582, 0.06449539889581501, 0.06519921310245991, 0.06381580187007785, 0.06428131996653974, 0.06423701113089919, 0.06466777203604579, 0.06460522394627333, 0.06426571705378592, 0.06441790796816349, 0.06442491197958589, 0.06454188097268343, 0.06440819101408124, 0.0643986479844898, 0.06461924407631159, 0.06441914383322, 0.06592402909882367, 0.06535166595131159, 0.06682791002094746, 0.07453999202698469, 0.07022899203002453, 0.06446894886903465, 0.06474803411401808, 0.06432138406671584, 0.06554008880630136, 0.06473238207399845, 0.06474224594421685, 0.06434998102486134, 0.06456242105923593, 0.06448452803306282, 0.06457700091414154, 0.06445694994181395, 0.06446507107466459, 0.06438840692862868, 0.06455147103406489, 0.0645058851223439, 0.06447110883891582, 0.06482497416436672, 0.0642788780387491, 0.06457894993945956, 0.06440117605961859, 0.06450261687859893, 0.06480748602189124, 0.06533847702667117, 0.06391248502768576, 0.06483878684230149, 0.0645243851467967, 0.06448019295930862, 0.06502053304575384, 0.06446918891742826, 0.06448735808953643, 0.0642947698943317, 0.06434316001832485, 0.06465236796066165, 0.06476011499762535, 0.06444308999925852, 0.06456233817152679, 0.06446763500571251, 0.06458385102450848, 0.06467968178912997, 0.06461834907531738, 0.06473676394671202, 0.0644566579721868, 0.06429858016781509, 0.06462250486947596, 0.06484143016859889, 0.06457036198116839, 0.06448886101134121, 0.06452367198653519, 0.06487324996851385, 0.06446262286044657, 0.06450968701392412, 0.06473038205876946, 0.06459637102670968, 0.06460009305737913, 0.0644772530067712, 0.06489878101274371, 0.06443215790204704, 0.0647935951128602, 0.06445907498709857, 0.06587776890955865, 0.06671375501900911, 0.0866124420426786], [0.06447948981076479, 0.06441117008216679, 0.06477263895794749, 0.06454942096024752, 0.06489682709798217, 0.06409631902351975, 0.06455846899189055, 0.064928880892694, 0.0644016731530428, 0.06365848099812865, 0.0646472629159689, 0.06428963202051818, 0.06548406393267214, 0.06465847813524306, 0.06387213291600347, 0.0642641659360379, 0.0646405250299722, 0.06433015898801386, 0.06432573404163122, 0.06515970802865922, 0.06426312704570591, 0.06446031085215509, 0.06442185607738793, 0.06415425101295114, 0.06439508008770645, 0.06460366095416248, 0.06509735086001456, 0.06426986819133162, 0.06422480684705079, 0.06433267798274755, 0.06491877208463848, 0.0643308931030333, 0.06386525393463671, 0.06415723403915763, 0.06428260402753949, 0.06389002478681505, 0.06445311801508069, 0.064161746064201, 0.06555172707885504, 0.07497035688720644, 0.05231355410069227, 0.06891593406908214, 0.06573922792449594, 0.06469913106411695, 0.06476065097376704, 0.06464842683635652, 0.06454660999588668, 0.06472441297955811, 0.0645637041889131, 0.06489180284552276, 0.06492242310196161, 0.06415779097005725, 0.06474345899187028, 0.06473963800817728, 0.06493928493000567, 0.06451168819330633, 0.06460332591086626, 0.06479864800348878, 0.06469596899114549, 0.06469915504567325, 0.0648964629508555, 0.0651656030677259, 0.06468261987902224, 0.06426167394965887, 0.06469299411401153, 0.06447737803682685, 0.0647065918892622, 0.06528673111461103, 0.06487917900085449, 0.06469031982123852, 0.06476218020543456, 0.06469153799116611, 0.06435419782064855, 0.06501338607631624, 0.06471637496724725, 0.0646759511437267, 0.06462646392174065, 0.06491029192693532, 0.06471492000855505, 0.06508705299347639, 0.06479178904555738, 0.06461085611954331, 0.06530362693592906, 0.06506377086043358, 0.06495004519820213, 0.06503860093653202, 0.06472158594988286, 0.06482264189980924, 0.06512367608956993, 0.06468515610322356, 0.0645429149735719, 0.06496288091875613, 0.0648985430598259, 0.0646886769682169, 0.06453620688989758, 0.06463132915087044, 0.06478860205970705, 0.06486664479598403, 0.0651708641089499, 0.06492633209563792, 0.06475462182424963, 0.06491222698241472, 0.06461793603375554, 0.06512797810137272, 0.0642937810625881, 0.06460440182127059, 0.06480152695439756, 0.06479869410395622, 0.06461264891549945, 0.06520927604287863, 0.06462971400469542, 0.06459980295039713, 0.06449363217689097, 0.0648185620084405, 0.06475884281098843, 0.0650253661442548, 0.06466800183989108, 0.0644607380963862, 0.06447738199494779, 0.0647338789422065, 0.06468611606396735, 0.06499305507168174, 0.0644244069699198, 0.06488052289932966, 0.06451201508753002, 0.06485942704603076, 0.0658496969845146, 0.06507655302993953, 0.06469259387813509, 0.06475999997928739, 0.06472940114326775, 0.06461696187034249, 0.06455060816369951, 0.06490781786851585, 0.06476965104229748, 0.06500858091749251, 0.06438340409658849, 0.06469441787339747, 0.06477857008576393, 0.06500917696394026, 0.06471758312545717, 0.06483218097127974, 0.064496710896492, 0.06462862715125084, 0.06460284488275647, 0.06513729505240917, 0.0646936078555882, 0.06453192001208663, 0.06444188416935503, 0.06530860788188875, 0.06479576998390257, 0.06503681116737425, 0.06585599994286895, 0.06534674600698054, 0.06367713492363691, 0.06506161903962493, 0.06465126108378172, 0.06409931182861328, 0.06475353497080505, 0.06411425420083106, 0.06446265080012381, 0.0645630091894418, 0.06339135393500328, 0.06440127594396472, 0.06438402808271348, 0.0640433318912983, 0.06470355507917702, 0.06419869605451822, 0.0643660849891603, 0.06462100287899375, 0.0642732479609549, 0.06489517213776708, 0.06400347594171762, 0.06384182092733681, 0.06435705511830747, 0.06434877403080463, 0.0642889488954097, 0.06420780811458826, 0.06421126797795296, 0.06467381888069212, 0.06346884509548545, 0.06455250503495336, 0.06454343302175403, 0.06524037080816925, 0.06496530817821622, 0.06537251989357173, 0.06498404406011105, 0.06549970200285316, 0.0647748839110136, 0.064216751139611, 0.06453087297268212, 0.0638698460534215, 0.06403131084516644, 0.06464029313065112, 0.06413625692948699, 0.06428210902959108, 0.06430124398320913, 0.06402662489563227, 0.06380240712314844, 0.06448247004300356, 0.06414860882796347, 0.06443833303637803, 0.06427997094579041, 0.06421759305521846, 0.06439909315668046, 0.06459217378869653, 0.06398207601159811, 0.06447620317339897, 0.06371934688650072, 0.06398408906534314, 0.06370770605280995, 0.0643330488819629, 0.06399644399061799, 0.06435835501179099, 0.06387504911981523, 0.0640587080270052, 0.06404723995365202, 0.06418765895068645, 0.06416557496413589, 0.0643031729850918, 0.06416469695977867, 0.06404896709136665, 0.06399343791417778, 0.06432197615504265, 0.06413531792350113, 0.06482594297267497, 0.06360400398261845, 0.06425919197499752, 0.06405545910820365, 0.06434912700206041, 0.0639874238986522, 0.064070989144966, 0.0639311308041215, 0.06428545410744846, 0.06378689804114401, 0.06421226798556745, 0.06408328306861222, 0.06416128179989755, 0.0642864580731839, 0.06420428398996592, 0.06414065603166819, 0.06473505799658597, 0.06376087991520762, 0.06414621300064027, 0.06425804016180336, 0.06447250698693097, 0.06434139097109437, 0.06427840795367956, 0.06430297996848822, 0.06425252999179065, 0.06372476206161082, 0.0641977689228952, 0.06446746713481843, 0.06468946090899408, 0.06441809702664614, 0.06446538493037224, 0.06538032600656152, 0.06318714516237378, 0.06380474101752043, 0.06438105087727308, 0.0641098041087389, 0.06493563996627927, 0.06386531284078956, 0.06411198107525706, 0.0637927579227835, 0.06438725511543453, 0.06422320706769824, 0.06412829388864338, 0.06399584305472672, 0.06458182889036834, 0.06389880599454045, 0.06426699017174542, 0.06453739781863987, 0.06393444212153554, 0.06417611800134182, 0.06426859204657376, 0.06370127503760159, 0.06432689889334142, 0.06464237510226667, 0.06341374199837446, 0.0637807659804821, 0.06397873884998262, 0.0641198621597141, 0.06416283501312137, 0.06463018897920847, 0.06430387194268405, 0.06408334895968437, 0.06448307004757226, 0.06353984191082418, 0.06424023909494281, 0.06430811993777752, 0.06444714008830488, 0.06388087500818074, 0.06423237384296954, 0.06425812700763345, 0.06402216502465308, 0.06424530711956322, 0.06420877296477556, 0.06405141903087497, 0.06409775791689754, 0.06400452996604145, 0.06420468003489077, 0.0642045671120286, 0.06405581790022552, 0.06366223399527371, 0.06435911799781024, 0.0640155989676714, 0.06403805199079216, 0.06507494510151446, 0.0641509909182787, 0.06480553792789578, 0.06376935518346727, 0.06382338795810938, 0.06440997589379549, 0.06322210095822811, 0.06470618001185358, 0.06441336404532194, 0.06387961097061634, 0.07073242100887, 0.06553862011060119, 0.06450721086002886, 0.06412527221255004, 0.06452743778936565, 0.06429308606311679, 0.06417693803086877, 0.06393164210021496, 0.06473636091686785, 0.06394945806823671, 0.06442192289978266, 0.06540351011790335, 0.06721578887663782, 0.06132116913795471, 0.06905668787658215, 0.06644030311144888, 0.07264800695702434, 0.06560758501291275, 0.06502555985935032, 0.06517203408293426, 0.06519567500799894, 0.0653631598688662, 0.06503597204573452, 0.06567137711681426, 0.06507351901382208, 0.06527471193112433, 0.06492264196276665, 0.06538974912837148, 0.06483793398365378, 0.06524741183966398, 0.06518302508629858, 0.06548835011199117, 0.06520248879678547, 0.06538607808761299, 0.06564787100069225, 0.06499791797250509, 0.06557347113266587, 0.06503253895789385, 0.06527525396086276, 0.06496061105281115, 0.06574319489300251, 0.06502298405393958, 0.06516149686649442, 0.06533327605575323, 0.06553083099424839, 0.06533707096241415, 0.06489764619618654, 0.06522142700850964, 0.06512228678911924, 0.06526980712078512, 0.06527348700910807, 0.0652423519641161, 0.06508412305265665, 0.06493914406746626, 0.06506270682439208, 0.06601475412026048, 0.06566673284396529, 0.06519538816064596, 0.06557405204512179, 0.06495554093271494, 0.06531125702895224, 0.06545037683099508, 0.06522797513753176, 0.06523921503685415, 0.06526121497154236, 0.06483080494217575, 0.06519954395480454, 0.06552773714065552, 0.06529548391699791, 0.06543696206063032, 0.06509121600538492, 0.06506393197923899, 0.06522528198547661, 0.06547152786515653, 0.06524442718364298, 0.06519384495913982, 0.06513547385111451, 0.06607814412564039, 0.06463545397855341, 0.06545350002124906, 0.06555791688151658, 0.06512048817239702, 0.06502183387055993, 0.06502021895721555, 0.0656172560993582, 0.06546127609908581, 0.0653978269547224, 0.06664922297932208, 0.07443143404088914, 0.07450483599677682, 0.06426601787097752, 0.06463765096850693, 0.06435216520912945, 0.06449049687944353, 0.06434404011815786, 0.06523506599478424, 0.06413799780420959, 0.0645986502058804, 0.0643486708868295, 0.06468611396849155, 0.06429900391958654, 0.06458721216768026, 0.06449302984401584, 0.06459267414174974, 0.0646906120236963, 0.06443470995873213, 0.06456121290102601, 0.06457068817690015, 0.06440294091589749, 0.06459687394089997, 0.06448240811005235, 0.06554670701734722, 0.06348989601247013, 0.06444234983064234, 0.06435635196976364, 0.064588240114972, 0.06460462091490626, 0.06426684209145606, 0.0644616570789367, 0.0643838478717953, 0.06453128112480044, 0.06441481993533671, 0.0643992938566953, 0.06462043919600546, 0.06464756280183792, 0.06631452520377934, 0.06449485896155238, 0.06792704900726676, 0.07437516492791474, 0.07029049191623926, 0.06348472204990685, 0.06473895395174623, 0.06432886305265129, 0.06553881312720478, 0.06474260985851288, 0.06501060794107616, 0.0648179380223155, 0.06425326108001173, 0.0641354639083147, 0.06451360904611647, 0.06452019209973514, 0.06448145303875208, 0.06439771200530231, 0.06453147693537176, 0.06439094408415258, 0.06459814286790788, 0.06482956791296601, 0.06444094702601433, 0.06427267612889409, 0.06427311594597995, 0.06467544008046389, 0.06468536099418998, 0.06541609694249928, 0.06469720299355686, 0.06417186907492578, 0.06441215798258781, 0.06461043003946543, 0.06498242286033928, 0.06436757300980389, 0.06478675105609, 0.06400803290307522, 0.06446152413263917, 0.06464015087112784, 0.06467635603621602, 0.0644308680202812, 0.06466972013004124, 0.06452290178276598, 0.06441490002907813, 0.06467995001003146, 0.06472126813605428, 0.06452266895212233, 0.06455001188442111, 0.06435297220014036, 0.06457799091003835, 0.06486147292889655, 0.064558160956949, 0.06447685603052378, 0.06452742917463183, 0.06488781492225826, 0.06444653100334108, 0.06451268796809018, 0.06462249299511313, 0.06470473599620163, 0.06457072193734348, 0.06453325995244086, 0.06486955005675554, 0.06443500495515764, 0.06478645303286612, 0.06447340012528002, 0.06543455203063786, 0.06846754299476743, 0.0861419418361038], [0.06442709197290242, 0.06417318503372371, 0.06445694388821721, 0.0643778401426971, 0.06385867786593735, 0.06441003503277898, 0.06438534404151142, 0.06404838687740266, 0.06470320420339704, 0.06436353479512036, 0.06420444999821484, 0.0646199129987508, 0.06426624301820993, 0.06489361100830138, 0.06400395301170647, 0.06384364701807499, 0.06434521405026317, 0.06435194704681635, 0.064304003957659, 0.06419910001568496, 0.06419131206348538, 0.06540331593714654, 0.06276093400083482, 0.06439968198537827, 0.06444125901907682, 0.06526163499802351, 0.0649272589944303, 0.06538416398689151, 0.06497080205008388, 0.06550939497537911, 0.06474141404032707, 0.06432626000605524, 0.06428466085344553, 0.06448449101299047, 0.06384949199855328, 0.06463749799877405, 0.06396183697506785, 0.06445157318376005, 0.06411411985754967, 0.06421920098364353, 0.06379503500647843, 0.06449174601584673, 0.06415132898837328, 0.06442039692774415, 0.06427546008490026, 0.06406228500418365, 0.06430159811861813, 0.0645417959894985, 0.06430581491440535, 0.0641677810344845, 0.06372454785741866, 0.0640045350883156, 0.0639777691103518, 0.06433433783240616, 0.06399559509009123, 0.06414926494471729, 0.06408436899073422, 0.06405394314788282, 0.06402760790660977, 0.06420730403624475, 0.06402179202996194, 0.0642695790156722, 0.06435317196883261, 0.06404302595183253, 0.06381946802139282, 0.06450260896235704, 0.06412790506146848, 0.06426975596696138, 0.06399200903251767, 0.06427408684976399, 0.0642013989854604, 0.06435984303243458, 0.063980728155002, 0.0640715709887445, 0.0639347960241139, 0.06412625289522111, 0.06394363101571798, 0.06421364890411496, 0.06408594502136111, 0.06414666213095188, 0.06429536291398108, 0.06419823202304542, 0.06397709902375937, 0.06491061206907034, 0.0637444609310478, 0.06416070694103837, 0.06425351114012301, 0.06447923183441162, 0.06414276314899325, 0.06447499990463257, 0.06429968401789665, 0.06409927806816995, 0.06388050294481218, 0.06419297307729721, 0.06416809489019215, 0.06467884290032089, 0.06441654311493039, 0.06449341401457787, 0.0653784740716219, 0.06350413803011179, 0.06351512693800032, 0.06433510687202215, 0.06416257796809077, 0.06481032818555832, 0.06389898783527315, 0.06409599515609443, 0.06402606703341007, 0.06437275186181068, 0.0642049650195986, 0.06414317199960351, 0.06408891500905156, 0.06448489101603627, 0.0639155909884721, 0.06425059004686773, 0.06453005806542933, 0.06403490994125605, 0.06418027495965362, 0.06383300409652293, 0.06389327486976981, 0.06436799210496247, 0.06389286695048213, 0.06407277495600283, 0.06396730802953243, 0.0639690940733999, 0.06410415982827544, 0.0641577651258558, 0.06464735488407314, 0.0643147740047425, 0.06407853495329618, 0.06406901217997074, 0.06396406702697277, 0.0642595759127289, 0.06427746405825019, 0.06444693100638688, 0.0639123700093478, 0.06419812701642513, 0.06406226800754666, 0.06420047278515995, 0.06425869301892817, 0.06418077205307782, 0.06392126204445958, 0.06421901006251574, 0.06410022685304284, 0.06417462206445634, 0.06417938810773194, 0.06421141000464559, 0.06366838491521776, 0.0641962350346148, 0.0640404939185828, 0.06416403199546039, 0.06507227313704789, 0.06416078982874751, 0.06469320016913116, 0.06365742697380483, 0.06384575390256941, 0.06461252807639539, 0.06404743785969913, 0.06356536317616701, 0.0644229669123888, 0.06388589809648693, 0.070753606967628, 0.06564730894751847, 0.06437872792594135, 0.06424656696617603, 0.06439815019257367, 0.06443580496124923, 0.06401719385758042, 0.06409056019037962, 0.06460054684430361, 0.06407561898231506, 0.06432178406976163, 0.06550792208872736, 0.067172905895859, 0.061084178974851966, 0.06914603314362466, 0.06623596395365894, 0.07272895891219378, 0.06592674111016095, 0.06517908303067088, 0.06517674378119409, 0.06544207409024239, 0.06512112403288484, 0.06503049493767321, 0.06523181800730526, 0.06522060604766011, 0.0651581019628793, 0.0653248131275177, 0.06510420795530081, 0.06512696202844381, 0.06523294001817703, 0.06519383378326893, 0.06534445518627763, 0.06534676090814173, 0.06538227805867791, 0.0652148260269314, 0.06525261094793677, 0.06530048907734454, 0.06533540692180395, 0.06527219200506806, 0.06511874590069056, 0.06574545102193952, 0.06487383414059877, 0.06531126797199249, 0.06533235195092857, 0.06553054694086313, 0.06532835797406733, 0.06490613799542189, 0.06522218114696443, 0.06510126101784408, 0.06513307499699295, 0.06542806699872017, 0.06509165186434984, 0.06507912906818092, 0.06509342906065285, 0.06506213289685547, 0.06601257203146815, 0.06566592305898666, 0.06519011803902686, 0.06538049899972975, 0.06542137498036027, 0.06503377901390195, 0.06546164001338184, 0.0652186837978661, 0.06524613103829324, 0.06510835513472557, 0.06498331599868834, 0.06519264495000243, 0.06554079405032098, 0.06529401382431388, 0.06543011520989239, 0.06508695078082383, 0.06507621216587722, 0.06506324000656605, 0.06547861197032034, 0.0653921109624207, 0.06520116096362472, 0.06513290200382471, 0.06577965407632291, 0.06493163388222456, 0.065453649032861, 0.06555539602413774, 0.06512175197713077, 0.06501892814412713, 0.06502396101132035, 0.06560977292247117, 0.06546335201710463, 0.06540187005884945, 0.06618498684838414, 0.07490569306537509, 0.07454983307980001, 0.06426768680103123, 0.06463831104338169, 0.06435082596726716, 0.06449219700880349, 0.06434581708163023, 0.0647197540383786, 0.06459042406640947, 0.06459380383603275, 0.06434920895844698, 0.06469273008406162, 0.06429328792728484, 0.06459262710995972, 0.06449463800527155, 0.06459150603041053, 0.06469309399835765, 0.06443298887461424, 0.06454886496067047, 0.064579077064991, 0.06440406502224505, 0.06459525413811207, 0.06448110286146402, 0.06491655204445124, 0.06412240397185087, 0.06448435201309621, 0.06428014510311186, 0.06468124291859567, 0.06460503092966974, 0.06426674104295671, 0.06438997201621532, 0.0644547960255295, 0.06453416496515274, 0.06441373704001307, 0.06439419183880091, 0.06462543411180377, 0.06450225203298032, 0.06549008307047188, 0.06538217491470277, 0.0677685730624944, 0.07409466081298888, 0.07035259203985333, 0.06387068214826286, 0.06474584201350808, 0.06432133098132908, 0.06554450886324048, 0.06473652203567326, 0.06471102708019316, 0.06450564786791801, 0.06435833219438791, 0.06448436598293483, 0.06467178394086659, 0.06439204793423414, 0.06444609700702131, 0.06443887203931808, 0.06449050991795957, 0.06455328012816608, 0.06445678300224245, 0.06483518285676837, 0.06443937798030674, 0.06439685099758208, 0.0647794350516051, 0.06417976203374565, 0.0649986460339278, 0.06508991308510303, 0.06505352398380637, 0.06370003800839186, 0.06454889895394444, 0.06447465089149773, 0.06507051805965602, 0.0644037020392716, 0.06445995601825416, 0.06434244685806334, 0.06433075107634068, 0.0646521260496229, 0.06479376996867359, 0.06442909291945398, 0.06454263208433986, 0.06465364503674209, 0.06440884293988347, 0.06468685902655125, 0.06458975397981703, 0.06496183690614998, 0.06423467909917235, 0.06434955098666251, 0.06458220700733364, 0.06486026896163821, 0.06456233910284936, 0.06447568186558783, 0.0645257590804249, 0.06488696206361055, 0.06444940599612892, 0.0645070148166269, 0.0651103740092367, 0.06421986920759082, 0.06457372498698533, 0.06452906783670187, 0.06487021711654961, 0.06442988105118275, 0.0647958479821682, 0.06446874188259244, 0.06543272593989968, 0.06839251704514027, 0.08537586499005556, 0.10731222317554057, 0.0004774569533765316, 0.04266306897625327, 0.054884752025827765, 0.04559539584442973, 0.04448532615788281, 0.043720305897295475, 0.04268198902718723, 0.04205973097123206, 0.04171736491844058, 0.04180817911401391, 0.04156644595786929, 0.042095065116882324, 0.04155605402775109, 0.04151060385629535, 0.041689016157761216, 0.04146298486739397, 0.04158017295412719, 0.04166741715744138, 0.041597333969548345, 0.04168442799709737, 0.04150190087966621, 0.04158222395926714, 0.04141499404795468, 0.041543019004166126, 0.04174490203149617, 0.04158832295797765, 0.04157091514207423, 0.041690350975841284, 0.04207752691581845, 0.04155159997753799, 0.04143713694065809, 0.04140430106781423, 0.04150246595963836, 0.0417413841933012, 0.04162088385783136, 0.041501509957015514, 0.041463208151981235, 0.04167453199625015, 0.04145218385383487, 0.04150545410811901, 0.04205769393593073, 0.04179509705863893, 0.0415307788643986, 0.04134472203440964, 0.04152263607829809, 0.041765548987314105, 0.041547039058059454, 0.041370474034920335, 0.04153956682421267, 0.0417063650675118, 0.04150692303664982, 0.04150090692564845, 0.04146940400823951, 0.0416381701361388, 0.041483179898932576, 0.04133944911882281, 0.041611453983932734, 0.04149447800591588, 0.041642292868345976, 0.0414893280249089, 0.04155679699033499, 0.04163948795758188, 0.04142038617283106, 0.04138679592870176, 0.0419125419575721, 0.04134249989874661, 0.041597551200538874, 0.04132334794849157, 0.042034168960526586, 0.04158749897032976, 0.041438769083470106, 0.0414636239875108, 0.04155316203832626, 0.04188174894079566, 0.04130069608800113, 0.04164154385216534, 0.04157988098450005, 0.04156820406205952, 0.041401162976399064, 0.041456713108345866, 0.04173975298181176, 0.04137279302813113, 0.0415488569997251, 0.041538799880072474, 0.042160304030403495, 0.041678682900965214, 0.041526989080011845, 0.04141010600142181, 0.041586400009691715, 0.04188962001353502, 0.0413800950627774, 0.041360341012477875, 0.041617120848968625, 0.04148647794499993, 0.04123756103217602, 0.04153700196184218, 0.04185972618870437, 0.04155291197821498, 0.041599195916205645, 0.041516531025990844, 0.04187130811624229, 0.041737522929906845, 0.041353347012773156, 0.04140405496582389, 0.04160909587517381, 0.0415221550501883, 0.04155226214788854, 0.04138101986609399, 0.04163127392530441, 0.04169612401165068, 0.04164112498983741, 0.04133041016757488, 0.041607247898355126, 0.0416899099946022, 0.04192930995486677, 0.041403996059671044, 0.04162556095980108, 0.04155429406091571, 0.04151711403392255, 0.041394968051463366, 0.04155243583954871, 0.04187384503893554, 0.041485944064334035, 0.04145359597168863, 0.04159043496474624, 0.04168466315604746, 0.04155487078242004, 0.04141745809465647, 0.04188655805774033, 0.04130809800699353, 0.04155476903542876, 0.04134089592844248, 0.04160402691923082, 0.04155618418008089, 0.04147681198082864, 0.04143105284310877, 0.04176405700854957, 0.04169648000970483, 0.041541069047525525, 0.041507424088194966, 0.04151220479980111, 0.041640193201601505, 0.041389898862689734, 0.041621840093284845, 0.04161682794801891, 0.04168673697859049, 0.04164418508298695, 0.04130346793681383, 0.04162933211773634, 0.041517878882586956, 0.041626122081652284, 0.04130394896492362, 0.04157233890146017, 0.041433627950027585, 0.04145708819851279, 0.04153016488999128, 0.042402483988553286], [0.06878908001817763, 0.06474168412387371, 0.06432997900992632, 0.06553481495939195, 0.06474307202734053, 0.06473298883065581, 0.06453685602173209, 0.06439227517694235, 0.06449169898405671, 0.06453914591111243, 0.0644649260211736, 0.06445349100977182, 0.0644217450171709, 0.06452939985319972, 0.06450042710639536, 0.06449855910614133, 0.06485871598124504, 0.06442812993191183, 0.06436936208046973, 0.06436258601024747, 0.0645318659953773, 0.0648292328696698, 0.06534051802009344, 0.06441372097469866, 0.06433284608647227, 0.06450475100427866, 0.0645081598777324, 0.06497256108559668, 0.06445451499894261, 0.06450536102056503, 0.06431351695209742, 0.0643531740643084, 0.06466660089790821, 0.06474263593554497, 0.06441521109081805, 0.06460374407470226, 0.06462591583840549, 0.06440678401850164, 0.06467611319385469, 0.06463049678131938, 0.06468706903979182, 0.06445421208627522, 0.06433491688221693, 0.06460759718902409, 0.06484565092250705, 0.06456552189774811, 0.06448676204308867, 0.06451723002828658, 0.06489323801361024, 0.06448482000268996, 0.06448033102788031, 0.06473441096022725, 0.06460809311829507, 0.06456265901215374, 0.06452235300093889, 0.06488772295415401, 0.06444428884424269, 0.06477244198322296, 0.06447838409803808, 0.06513801799155772, 0.06790719600394368, 0.08706224896013737, 0.1070157471112907, 7.744901813566685e-05, 0.04300907487049699, 0.05488677299581468, 0.045571945142000914, 0.04449622589163482, 0.04372251103632152, 0.04269413207657635, 0.042064786888659, 0.041686119977384806, 0.041822987142950296, 0.04157786490395665, 0.0420651740860194, 0.0415751738473773, 0.04150527901947498, 0.04168615001253784, 0.04145951406098902, 0.04159247595816851, 0.0416574589908123, 0.041597686940804124, 0.04168664198368788, 0.04150350298732519, 0.041585133178159595, 0.04142304789274931, 0.04155199904926121, 0.04173677391372621, 0.04158221906982362, 0.04157241806387901, 0.0416832969058305, 0.042078258004039526, 0.04155712900683284, 0.04142784606665373, 0.04140848899260163, 0.04150450602173805, 0.041739052860066295, 0.04162019211798906, 0.04150218586437404, 0.04147118120454252, 0.04167147586122155, 0.04145919601432979, 0.04149685800075531, 0.042053553042933345, 0.04180096602067351, 0.04152331198565662, 0.04134773393161595, 0.041535126976668835, 0.041760656982660294, 0.04154967097565532, 0.04136648401618004, 0.04156429204158485, 0.04168418701738119, 0.041500816121697426, 0.0415008079726249, 0.04147534188814461, 0.04163310513831675, 0.0414873359259218, 0.04133849102072418, 0.04160806396976113, 0.041498323902487755, 0.04164024000056088, 0.04149450804106891, 0.04155429801903665, 0.041638378985226154, 0.041417275089770555, 0.04139186884276569, 0.041910559171810746, 0.04136028094217181, 0.041586210019886494, 0.04132489487528801, 0.04198925313539803, 0.04161978093907237, 0.04145053611136973, 0.04146685893647373, 0.04155084094963968, 0.041942913085222244, 0.04122881987132132, 0.041608378989622, 0.04162175813689828, 0.04157257406041026, 0.04139186698012054, 0.04147457890212536, 0.04172989702783525, 0.041366924066096544, 0.04154349002055824, 0.04154371586628258, 0.04216285492293537, 0.04167421814054251, 0.041536936070770025, 0.04140147380530834, 0.04159431206062436, 0.0418838239274919, 0.04137889598496258, 0.04135689022950828, 0.04161194385960698, 0.041501855943351984, 0.04118801816366613, 0.04158401000313461, 0.04186649085022509, 0.041546470019966364, 0.0415993039496243, 0.04151120805181563, 0.0418783170171082, 0.04173631011508405, 0.041351270861923695, 0.04141000402159989, 0.04161191498860717, 0.041515331948176026, 0.04155300813727081, 0.04139891802333295, 0.04161374783143401, 0.04169595707207918, 0.04164557904005051, 0.04132816498167813, 0.041622488060966134, 0.041673945961520076, 0.04192625195719302, 0.04140494600869715, 0.04163120500743389, 0.04154981090687215, 0.04152694810181856, 0.04139014892280102, 0.041558271972462535, 0.04186504799872637, 0.041492137126624584, 0.041449307929724455, 0.04159386199899018, 0.04167748801410198, 0.04155913204886019, 0.04141909210011363, 0.04188925097696483, 0.041308393934741616, 0.04155311593785882, 0.04133986495435238, 0.04161160113289952, 0.041553903836756945, 0.041471915086731315, 0.04143641307018697, 0.04175695194862783, 0.041701070964336395, 0.04154313309118152, 0.041505802888423204, 0.041516944067552686, 0.04163585091009736, 0.041394263971596956, 0.04161907802335918, 0.04161582514643669, 0.04168604896403849, 0.041645613964647055, 0.04130034102126956, 0.04163072491064668, 0.04151591402478516, 0.0416279979981482, 0.04130396316759288, 0.041572160786017776, 0.04143572715111077, 0.04145886888727546, 0.04153158701956272, 0.04244779911823571, 0.06269504502415657, 0.022745323833078146, 0.04776330105960369, 0.03788264491595328, 0.038335368037223816, 0.037922236137092113, 0.03786754794418812, 0.03786067804321647, 0.04150907485745847, 0.04126629116944969, 0.04176882398314774, 0.04165383893996477, 0.04176320694386959, 0.04170731897465885, 0.041761147091165185, 0.041582670994102955, 0.041631102096289396, 0.041643753880634904, 0.04176461789757013, 0.04174136300571263, 0.041610377142205834, 0.04168879194185138, 0.04170972993597388, 0.04158292803913355, 0.0417961070779711, 0.04157999996095896, 0.04156673699617386, 0.041752055985853076, 0.04151505697518587, 0.04168777703307569, 0.04163103993050754, 0.0422355851624161, 0.041696310974657536, 0.04146363399922848, 0.04165594885125756, 0.0417623370885849, 0.04158627497963607, 0.04159137397073209, 0.041662957053631544, 0.04175053001381457, 0.04183953907340765, 0.04169317986816168, 0.04201258812099695, 0.04174600983969867, 0.04202417517080903, 0.042005933821201324, 0.041564874118193984, 0.041585620027035475, 0.041661125840619206, 0.041499832179397345, 0.041677623987197876, 0.04153931490145624, 0.04172873101197183, 0.04158786893822253, 0.041502908105030656, 0.041541945887729526, 0.04181450209580362, 0.04155829502269626, 0.04149688803590834, 0.04170386283658445, 0.04169174307025969, 0.041733792051672935, 0.04200466210022569, 0.04166794288903475, 0.0416123210452497, 0.041704789036884904, 0.041463566944003105, 0.041751704877242446, 0.04170164908282459, 0.04152901191264391, 0.04157502809539437, 0.04151491611264646, 0.04165435186587274, 0.04153885995037854, 0.04150421801023185, 0.0416566701605916, 0.04155853181146085, 0.041705242125317454, 0.041439526015892625, 0.04165557189844549, 0.041586495004594326, 0.041641257936134934, 0.041549027198925614, 0.04150673095136881, 0.041700847912579775, 0.04213389800861478, 0.0420590010471642, 0.04164995695464313, 0.041502945125103, 0.04174787690863013, 0.04170313593931496, 0.04181678616441786, 0.041588729014620185, 0.04154711798764765, 0.0415497028734535, 0.04163752612657845, 0.04155342187732458, 0.041495193960145116, 0.041722106048837304, 0.04159070109017193, 0.041748143965378404, 0.041612060042098165, 0.04223477793857455, 0.04160204203799367, 0.041652098996564746, 0.041447178926318884, 0.04172006412409246, 0.04188791988417506, 0.04149539791978896, 0.04141973005607724, 0.04178404505364597, 0.04165137489326298, 0.041672711027786136, 0.041651356033980846, 0.041442218935117126, 0.04167303116992116, 0.04174741497263312, 0.04160053003579378, 0.04160570097155869, 0.041600725846365094, 0.04185675899498165, 0.04171995515935123, 0.041473775869235396, 0.04166682413779199, 0.04183862381614745, 0.04167814808897674, 0.041361453011631966, 0.04163062199950218, 0.04162303893826902, 0.04144123010337353, 0.0415863529779017, 0.041943693067878485, 0.04141381103545427, 0.04149380000308156, 0.041424708906561136, 0.041791200870648026, 0.041734131053090096, 0.041507303016260266, 0.04142400110140443, 0.04170782002620399, 0.041568046901375055, 0.04162502405233681, 0.0414553249720484, 0.04163010301999748, 0.04185174801386893, 0.041396963875740767, 0.04138734610751271, 0.04159049899317324, 0.041755069978535175, 0.04172279802151024, 0.04161376692354679, 0.041559385135769844, 0.04145093681290746, 0.0417831891681999, 0.04162341193296015, 0.041466106893494725, 0.041637483052909374, 0.04159608599729836, 0.04159395815804601, 0.04173623980022967, 0.04170535202138126, 0.04167384817264974, 0.041564777959138155, 0.04156640893779695, 0.041614802088588476, 0.041523272870108485, 0.04155285097658634, 0.041649049147963524, 0.041621899930760264, 0.041467624017968774, 0.04155108891427517, 0.04144785599783063, 0.04161776998080313, 0.04175779898650944, 0.041220579063519835, 0.04174314113333821, 0.041411213809624314, 0.04157537710852921, 0.041450324933975935, 0.04172382201068103, 0.04160651215352118, 0.041477539809420705, 0.04157669213600457, 0.041488882852718234, 0.04155254201032221, 0.04144168901257217, 0.04146452317945659, 0.04153234278783202, 0.041576637187972665, 0.04145075683481991, 0.041569276014342904, 0.04142016312107444, 0.04162121098488569, 0.04167897789739072, 0.04148992104455829, 0.0425005869474262, 0.04164471011608839, 0.041353822918608785, 0.041463022120296955, 0.04194899182766676, 0.04167780512943864, 0.04161097691394389, 0.041577238123863935, 0.041419144021347165, 0.04158658185042441, 0.04162715608254075, 0.04146835091523826, 0.04155505308881402, 0.04151751194149256, 0.041559390956535935, 0.04169903811998665, 0.04153885296545923, 0.04185604699887335, 0.0423772819340229, 0.04127059015445411, 0.041511372895911336, 0.04194398201070726, 0.0414729209151119, 0.04152903403155506, 0.04165317094884813, 0.04220215207897127, 0.041304884012788534, 0.04158374387770891, 0.0418401260394603, 0.04121165908873081, 0.04141922504641116, 0.04145533894188702, 0.04161469801329076, 0.041794324992224574, 0.041466349037364125, 0.04143965500406921, 0.041698980843648314, 0.04164971713908017, 0.041350852930918336, 0.04136879206635058, 0.04134290502406657, 0.04161184304393828, 0.041537833865731955, 0.041438905056566, 0.04177844803780317, 0.041680393973365426, 0.04160261107608676, 0.04241418791934848, 0.04306622198782861, 0.04485265305265784, 0.047011938877403736, 0.04730465612374246, 0.047746479976922274, 0.04769675596617162, 0.04792857589200139, 0.04751435201615095, 0.04786511207930744, 0.04835448507219553, 0.046892168931663036, 0.04748192708939314, 0.04755275486968458, 0.047648495994508266, 0.047846920089796185, 0.047882060054689646, 0.04811804392375052, 0.0476247810292989, 0.04774269298650324, 0.04762488487176597, 0.048206472070887685, 0.04717515711672604, 0.04756849282421172, 0.04747603018768132, 0.04746037279255688, 0.047714166110381484, 0.047825061017647386, 0.04763357900083065, 0.04753437195904553, 0.04930364899337292, 0.04827293008565903, 0.04791387286968529, 0.04764887900091708, 0.04760868614539504, 0.047529720002785325, 0.04764098790474236, 0.047584549989551306, 0.0475498519372195, 0.048077818006277084, 0.04781814105808735, 0.04743001307360828, 0.04725264594890177, 0.047881179954856634, 0.047187244053930044, 0.04751154291443527, 0.04785702517256141, 0.04756325180642307, 0.04836827307008207], [0.041771171148866415, 0.041660378919914365, 0.041757066966965795, 0.041707725962623954, 0.04176795110106468, 0.041583220940083265, 0.041630862979218364, 0.04164087609387934, 0.04176745191216469, 0.04174257209524512, 0.04160823905840516, 0.04168871301226318, 0.041705401847139, 0.04158328613266349, 0.04179196688346565, 0.04157476196996868, 0.041580650955438614, 0.041748258052393794, 0.041520789032801986, 0.041688137920573354, 0.04162601800635457, 0.04223739611916244, 0.04169772705063224, 0.041463991859927773, 0.04165269900113344, 0.0417597780469805, 0.041573175927624106, 0.041597980074584484, 0.04166780295781791, 0.04173927195370197, 0.04183631204068661, 0.04170424514450133, 0.042014044942334294, 0.041739296866580844, 0.042023408226668835, 0.04201340000145137, 0.04156765597872436, 0.041584780905395746, 0.04166007391177118, 0.0414952440187335, 0.04166969517245889, 0.041548619978129864, 0.041727122850716114, 0.04159035300835967, 0.04150568204931915, 0.041536835953593254, 0.04181861015968025, 0.04154656291939318, 0.0415020058862865, 0.04171117302030325, 0.041663700016215444, 0.04175353096798062, 0.04200328607112169, 0.04164501209743321, 0.04164717788808048, 0.041679081972688437, 0.04148086812347174, 0.041753680910915136, 0.04170249309390783, 0.04151808889582753, 0.04158367612399161, 0.04151000385172665, 0.04165928694419563, 0.04152903216890991, 0.04149026796221733, 0.04168151807971299, 0.04155396483838558, 0.04170966916717589, 0.04144025780260563, 0.04161340417340398, 0.041631539817899466, 0.041641312185674906, 0.041543025989085436, 0.04148468794301152, 0.04168985993601382, 0.04213414201512933, 0.0420917090959847, 0.0416143408510834, 0.04151346301659942, 0.04174361610785127, 0.04170165490359068, 0.0418291911482811, 0.041587743908166885, 0.04155268194153905, 0.04154931311495602, 0.041628323029726744, 0.041563216829672456, 0.04150419100187719, 0.04170510312542319, 0.04159857286140323, 0.041755644138902426, 0.04158925195224583, 0.042213588021695614, 0.04161734809167683, 0.04166810098104179, 0.04144583782181144, 0.041717439191415906, 0.0418582409620285, 0.04154580691829324, 0.04148180293850601, 0.04167853808030486, 0.04167835507541895, 0.04167300695553422, 0.04164025606587529, 0.041451707016676664, 0.04167439881712198, 0.041719789151102304, 0.041618483839556575, 0.04161635017953813, 0.04160563298501074, 0.041850554989650846, 0.04171988693997264, 0.041472177021205425, 0.0416704579256475, 0.04183660098351538, 0.04167726798914373, 0.04136243695393205, 0.04163339803926647, 0.04162447899580002, 0.04144031717441976, 0.04157721484079957, 0.04194313008338213, 0.04140338487923145, 0.041503812186419964, 0.041434484999626875, 0.04178843693807721, 0.04168868693523109, 0.0415377460885793, 0.04142530704848468, 0.04172102082520723, 0.04156889417208731, 0.04162700194865465, 0.04145087394863367, 0.04163526790216565, 0.04185066814534366, 0.04139503091573715, 0.0413815570063889, 0.04159104009158909, 0.04175585601478815, 0.04173498786985874, 0.04158492595888674, 0.041564783081412315, 0.04157014098018408, 0.041667329147458076, 0.0416353449691087, 0.04146881983615458, 0.04162172297947109, 0.041599352145567536, 0.041593483882024884, 0.04173279297538102, 0.04171610902994871, 0.0416746879927814, 0.04155275411903858, 0.041575837880373, 0.0416127210482955, 0.04153098305687308, 0.0415535110514611, 0.041652523912489414, 0.04161446099169552, 0.04146622703410685, 0.04154982906766236, 0.041451636934652925, 0.04161669104360044, 0.04170698905363679, 0.04126280196942389, 0.04174879193305969, 0.04139942699111998, 0.04158035200089216, 0.04145237407647073, 0.04172840085811913, 0.04158544400706887, 0.041492262156680226, 0.0415773899294436, 0.041488731978461146, 0.04153813305310905, 0.04145396105013788, 0.04146827384829521, 0.041532933013513684, 0.0415733999107033, 0.041459042113274336, 0.04156600893475115, 0.041421537986025214, 0.04162493301555514, 0.041671862127259374, 0.04148559900932014, 0.04250404704362154, 0.04165323288179934, 0.04135305993258953, 0.04145831102505326, 0.04194690496660769, 0.041665379190817475, 0.04162268899381161, 0.041586912935599685, 0.04141578893177211, 0.04156230599619448, 0.04163335799239576, 0.041480167070403695, 0.04154979600571096, 0.04152567498385906, 0.04155185096897185, 0.04169867001473904, 0.04154910589568317, 0.04183946503326297, 0.04229688597843051, 0.04133775900118053, 0.04150499007664621, 0.04196660011075437, 0.04147859988734126, 0.04152112896554172, 0.04166503809392452, 0.042196589056402445, 0.04130560997873545, 0.04148838296532631, 0.041911535896360874, 0.041277515003457665, 0.04139110213145614, 0.0414381108712405, 0.04161201510578394, 0.041785150999203324, 0.04147392394952476, 0.041409028926864266, 0.041706333169713616, 0.04167566983960569, 0.04133920697495341, 0.04137971205636859, 0.0413379049859941, 0.04161566309630871, 0.04153666505590081, 0.04142472590319812, 0.041791822062805295, 0.04167356598190963, 0.041610040003433824, 0.04227259894832969, 0.04319975897669792, 0.04484758200123906, 0.04698536009527743, 0.04731241287663579, 0.047741072019562125, 0.04769897507503629, 0.04793723695911467, 0.04751567798666656, 0.047865842934697866, 0.04829094815067947, 0.046954056015238166, 0.04748795600607991, 0.04754678579047322, 0.047646749997511506, 0.04785085306502879, 0.04787740297615528, 0.04808641504496336, 0.04769911500625312, 0.04767722403630614, 0.047643540892750025, 0.04809032008051872, 0.047274156007915735, 0.047573605086654425, 0.0474786339327693, 0.04746407805941999, 0.04771368484944105, 0.047823291970416903, 0.04763504909351468, 0.047534401062875986, 0.04930524597875774, 0.04828264191746712, 0.04790201294235885, 0.047653155168518424, 0.047603129874914885, 0.04753308813087642, 0.04764138185419142, 0.047583399107679725, 0.047547356924042106, 0.048063975060358644, 0.04734683409333229, 0.047834134893491864, 0.04733297508209944, 0.047521492931991816, 0.047508653020486236, 0.0475247020367533, 0.047860585851594806, 0.047575260046869516, 0.04821266210637987, 0.06660168198868632, 0.028256631921976805, 0.04989982000552118, 0.043771468102931976, 0.04197304486297071, 0.04162303195334971, 0.041821248130872846, 0.04172730795107782, 0.04173753992654383, 0.04153212299570441, 0.0418685101903975, 0.041597783798351884, 0.04165417212061584, 0.04144239192828536, 0.041467680130153894, 0.04158693691715598, 0.041304728016257286, 0.04165256302803755, 0.04142238385975361, 0.04169026110321283, 0.04160398500971496, 0.04191374103538692, 0.041574724949896336, 0.041547179920598865, 0.04155358416028321, 0.041406249860301614, 0.041378775145858526, 0.04179872293025255, 0.041421901900321245, 0.041501239174976945, 0.04262944892980158, 0.04148891707882285, 0.041698820888996124, 0.04161470988765359, 0.041680759051814675, 0.04145950195379555, 0.04148637317121029, 0.04166426695883274, 0.04153542104177177, 0.04158749687485397, 0.04154293006286025, 0.04160238988697529, 0.04169340315274894, 0.04158449894748628, 0.04155726591125131, 0.04132741107605398, 0.04180125310085714, 0.04191531194373965, 0.0415538779925555, 0.041503672022372484, 0.04157637688331306, 0.041706212097778916, 0.04269737796857953, 0.041531174909323454, 0.041327370097860694, 0.04201377811841667, 0.041564241982996464, 0.041824404848739505, 0.04158356715925038, 0.04160774080082774, 0.04162282310426235, 0.04199563595466316, 0.04494985402561724, 0.04731775796972215, 0.047442781971767545, 0.04753666603937745, 0.04746379912830889, 0.04763489589095116, 0.04761522402986884, 0.047568632988259196, 0.047573955031111836, 0.04753561690449715, 0.04791651712730527, 0.048271378967911005, 0.047243496868759394, 0.04743869602680206, 0.04741409397684038, 0.04762101103551686, 0.04761793208308518, 0.04761700099334121, 0.04783209296874702, 0.047551011899486184, 0.04763166117481887, 0.047564951004460454, 0.04747976781800389, 0.04762266110628843, 0.047695196932181716, 0.047516669146716595, 0.047529242001473904, 0.04742429102770984, 0.048259220784530044, 0.04744021501392126, 0.04806049005128443, 0.049415003042668104, 0.04918121895752847, 0.06367780501022935, 0.06529680010862648, 0.06371955387294292, 0.06436694506555796, 0.06473148800432682, 0.06368275708518922, 0.06458148290403187, 0.06439960701391101, 0.06408547586761415, 0.06433619419112802, 0.06601430498994887, 0.06463081086985767, 0.06447580317035317, 0.0644512427970767, 0.06465705507434905, 0.06448985612951219, 0.06458518491126597, 0.0641036438755691, 0.06470063701272011, 0.06399954319931567, 0.06426762300543487, 0.06482898397371173, 0.0643351508770138, 0.06497584795579314, 0.06500905007123947, 0.06422186689451337, 0.06418558210134506, 0.06443005590699613, 0.06438811006955802, 0.06418473809026182, 0.06461531203240156, 0.06465518986806273, 0.06463974295184016, 0.06530796410515904, 0.0655108829960227, 0.06524176988750696, 0.06604063510894775, 0.06488761794753373, 0.06552106607705355, 0.06559273204766214, 0.06524426187388599, 0.06541566015221179, 0.06556726479902864, 0.0650275971274823, 0.06539501692168415, 0.06496750703081489, 0.06516435695812106, 0.06477475399151444, 0.06495179515331984, 0.06508123292587698, 0.06484501902014017, 0.065169848036021, 0.06555958092212677, 0.06491076806560159, 0.06538604083471, 0.06494532315991819, 0.06521446886472404, 0.06491659302264452, 0.06468073511496186, 0.06503255805000663, 0.06526010390371084, 0.06469151400960982, 0.06506291893310845, 0.06479156296700239, 0.06509365211240947, 0.06498437887057662, 0.06515617319382727, 0.06482865381985903, 0.06509562418796122, 0.0649784249253571, 0.06504601496271789, 0.06518928403966129, 0.06522562098689377, 0.0648186260368675, 0.06511613191105425, 0.06506414501927793, 0.06518389098346233, 0.06484504998661578, 0.0648965120781213, 0.06611266895197332, 0.06502057402394712, 0.06485088402405381, 0.06613070494495332, 0.06524317804723978, 0.0654716519638896, 0.06542319408617914, 0.06550804595462978, 0.06508115003816783, 0.06524012004956603, 0.065997120924294, 0.06533872592262924, 0.06478303694166243, 0.06514563201926649, 0.06486358703114092, 0.06526091112755239, 0.06512132100760937, 0.06585564091801643, 0.06428054207935929, 0.06502082594670355, 0.06477129901759326, 0.0650198319926858, 0.06474044593051076, 0.06522866105660796, 0.06475720601156354, 0.06532356492243707, 0.06530777411535382, 0.06541356700472534, 0.064731374848634, 0.06561874016188085, 0.06513297487981617, 0.06504391506314278, 0.06504617584869266, 0.06561473803594708, 0.06443622219376266, 0.06446576793678105, 0.06425617495551705, 0.06449159095063806, 0.06410053302533925, 0.06494598905555904, 0.06416140589863062, 0.06431330810301006, 0.06413203198462725, 0.06441671308130026, 0.06431000679731369, 0.06441041803918779, 0.0645746230147779, 0.06427054107189178, 0.06431429693475366, 0.06434355396777391, 0.06400269595906138, 0.06449651019647717, 0.06422653887420893, 0.06439273012802005, 0.06518180086277425], [0.04730500606819987, 0.04774409788660705, 0.04769290308468044, 0.04793318687006831, 0.04750948306173086, 0.0478644969407469, 0.04835986904799938, 0.04688860010355711, 0.04748006584122777, 0.047554579097777605, 0.04764711204916239, 0.047852779971435666, 0.04788049892522395, 0.04812487098388374, 0.04760564211755991, 0.047762949019670486, 0.04761688690632582, 0.048238158924505115, 0.04715126100927591, 0.047562260180711746, 0.047479535918682814, 0.04745422606356442, 0.0477153358515352, 0.04782913299277425, 0.04763601208105683, 0.0475345510058105, 0.049297953955829144, 0.04827116406522691, 0.047920150915160775, 0.0476450661662966, 0.0476176468655467, 0.04752483498305082, 0.04763917508535087, 0.04758369107730687, 0.047551682917401195, 0.048082063905894756, 0.04789441009052098, 0.04737266688607633, 0.0472339300904423, 0.04790938692167401, 0.04716913518495858, 0.047507656970992684, 0.04785233293659985, 0.04755762405693531, 0.04844955890439451, 0.06621244107373059, 0.028204069007188082, 0.049888783833011985, 0.04377135704271495, 0.0419621670152992, 0.04163327510468662, 0.041817757999524474, 0.041725817834958434, 0.04173931712284684, 0.041533712996169925, 0.0418979509267956, 0.041607914958149195, 0.04159722104668617, 0.04144044010899961, 0.04148610006086528, 0.04158968385308981, 0.04130124906077981, 0.04166306392289698, 0.04141574609093368, 0.04168573091737926, 0.04160191910341382, 0.041918505914509296, 0.041579267010092735, 0.04154223692603409, 0.04155441699549556, 0.041402032133191824, 0.04138898686505854, 0.04179766005836427, 0.041413570987060666, 0.041515826946124434, 0.04261483205482364, 0.04149803100153804, 0.04169546812772751, 0.041612501023337245, 0.04168015089817345, 0.041460905922576785, 0.04153263918124139, 0.04161237878724933, 0.04154092515818775, 0.04159044101834297, 0.04154159897007048, 0.04161705891601741, 0.041680783964693546, 0.041584626073017716, 0.0415438620839268, 0.041341864969581366, 0.04179871897213161, 0.04191927402280271, 0.041588586987927556, 0.04147218191064894, 0.04154321597889066, 0.041723722126334906, 0.042710555950179696, 0.041533455019816756, 0.041328086983412504, 0.04200897295959294, 0.041564131155610085, 0.041832536924630404, 0.04157929006032646, 0.04160657781176269, 0.04165239701978862, 0.0419677069876343, 0.04496390814892948, 0.04726745095103979, 0.047428074991330504, 0.0475382620934397, 0.04746583686210215, 0.047676319954916835, 0.04757847497239709, 0.04733965313062072, 0.04779852507635951, 0.04757978394627571, 0.04791311686858535, 0.048267528181895614, 0.047201413894072175, 0.04744747607037425, 0.047406965866684914, 0.047631327994167805, 0.04764722497202456, 0.04761617002077401, 0.047834960045292974, 0.0475476561114192, 0.04759550583548844, 0.047569056041538715, 0.047511110082268715, 0.04759265203028917, 0.047694871900603175, 0.047515186946839094, 0.04753051814623177, 0.047463061986491084, 0.048219169955700636, 0.04748445888981223, 0.04805921018123627, 0.04937441390939057, 0.049134036991745234, 0.06372530292719603, 0.06529867905192077, 0.06334503507241607, 0.06435651890933514, 0.06473125796765089, 0.06385619402863085, 0.06440246314741671, 0.06436502281576395, 0.06420404906384647, 0.06433350313454866, 0.06593798100948334, 0.06463305279612541, 0.06448326818645, 0.06444693193770945, 0.06473302491940558, 0.06447901204228401, 0.06451723794452846, 0.06414995598606765, 0.06467468407936394, 0.06391981290653348, 0.06440710206516087, 0.064771993085742, 0.06428785785101354, 0.0650739970151335, 0.06485942704603076, 0.06435553194023669, 0.06419350416399539, 0.06434329901821911, 0.06455322285182774, 0.064112754072994, 0.06459850003011525, 0.06466693594120443, 0.06474535609595478, 0.06520307692699134, 0.06558866892009974, 0.06515852408483624, 0.06604963005520403, 0.06488377787172794, 0.06551766395568848, 0.06560466019436717, 0.06523715401999652, 0.0654136489611119, 0.06564069702289999, 0.06495635979808867, 0.06539866817183793, 0.06495620287023485, 0.06516832602210343, 0.06477648112922907, 0.06494053895585239, 0.0650905289221555, 0.06482030404731631, 0.06518947309814394, 0.06555131101049483, 0.06491898489184678, 0.06528585194610059, 0.06503956695087254, 0.06530758203007281, 0.06481702416203916, 0.06468804785981774, 0.06503346096724272, 0.06525920704007149, 0.06468605506233871, 0.06506508099846542, 0.06479625194333494, 0.0650939519982785, 0.06498308014124632, 0.06516236695460975, 0.06482236995361745, 0.0650941499043256, 0.064978500129655, 0.06494452687911689, 0.06529351696372032, 0.06521939300000668, 0.06481106206774712, 0.06513223005458713, 0.06505227601155639, 0.06518213986419141, 0.06484929006546736, 0.0649068420752883, 0.06610401696525514, 0.06501853605732322, 0.06485003395937383, 0.0661309459246695, 0.06524292915128171, 0.06547235301695764, 0.06541889696381986, 0.06550265685655177, 0.0650759250856936, 0.06524853105656803, 0.06599401799030602, 0.06524044391699135, 0.06488007912412286, 0.06514286086894572, 0.06495909695513546, 0.06515806703828275, 0.0651276980061084, 0.06585853011347353, 0.06436989386565983, 0.06493496801704168, 0.06476876907981932, 0.06494585587643087, 0.06481672497466207, 0.0652264051605016, 0.06476352689787745, 0.06533214705996215, 0.0652882179711014, 0.06542073097079992, 0.06472702999599278, 0.06562336906790733, 0.06511758686974645, 0.06510983314365149, 0.065002336865291, 0.0655968391802162, 0.06445171684026718, 0.06445296597667038, 0.06426703603938222, 0.06448570918291807, 0.06410295586101711, 0.06494864099659026, 0.06414530496113002, 0.06433295295573771, 0.06412689806893468, 0.06433268799446523, 0.06439320603385568, 0.064408496953547, 0.06455981312319636, 0.06426432891748846, 0.06433617998845875, 0.06432792404666543, 0.06400227500125766, 0.06450080196373165, 0.06422652606852353, 0.06438943487592041, 0.06516903196461499, 0.06924497918225825, 0.05900385999120772, 0.06873055989854038, 0.06696473993360996, 0.06489654700271785, 0.06488767499104142, 0.06489412509836257, 0.06493150698952377, 0.06508672912605107, 0.06483796378597617, 0.06488525401800871, 0.06507385917939246, 0.06486715585924685, 0.0651068810839206, 0.06500060693360865, 0.06549090892076492, 0.06502028903923929, 0.06485251197591424, 0.06514414912089705, 0.06464574998244643, 0.06489556096494198, 0.0650498210452497, 0.06513416091911495, 0.06511444901116192, 0.06464869901537895, 0.06489779194816947, 0.06505601806566119, 0.06480466108769178, 0.06545458687469363, 0.065191640984267, 0.06496103201061487, 0.06497061904519796, 0.06488834391348064, 0.0652153491973877, 0.06460552895441651, 0.06508167390711606, 0.06470515299588442, 0.06490165996365249, 0.06506576901301742, 0.06477205199189484, 0.06466834805905819, 0.06518848100677133, 0.06497891596518457, 0.06500127213075757, 0.06476338487118483, 0.0651104690041393, 0.06494455202482641, 0.06497333594597876, 0.06527597107924521, 0.0650761160068214, 0.06512123695574701, 0.06510715815238655, 0.06502132280729711, 0.06490539317019284, 0.06488091894425452, 0.06508828792721033, 0.06492618308402598, 0.06522273900918663, 0.06494608405046165, 0.06502417381852865, 0.06513851904310286, 0.06497210799716413, 0.0649431049823761, 0.06489591114223003, 0.06508826394565403, 0.06448314106091857, 0.06500915298238397, 0.06507327081635594, 0.06499999109655619, 0.06532944296486676, 0.06489755306392908, 0.06501448387280107, 0.06481877714395523, 0.06516892206855118, 0.06613196199759841, 0.0647784338798374, 0.0651986189186573, 0.06501497700810432, 0.0650168030988425, 0.06501740799285471, 0.06473369803279638, 0.065803118981421, 0.06506172800436616, 0.0651440538931638, 0.06472085905261338, 0.06511362013407052, 0.06524110189639032, 0.06492345896549523, 0.06506324303336442, 0.06508402503095567, 0.06514154886826873, 0.06507217907346785, 0.06500704097561538, 0.06489870510995388, 0.0651095958892256, 0.06492347200401127, 0.06503801001235843, 0.06483390112407506, 0.06521222996525466, 0.06484437291510403, 0.06488156807608902, 0.0655478578992188, 0.06494446704164147, 0.06508261803537607, 0.06495887599885464, 0.06496674800291657, 0.06508283200673759, 0.0648841829970479, 0.06496275495737791, 0.06528518302366138, 0.06506714993156493, 0.06510579097084701, 0.06502562016248703, 0.06502643693238497, 0.06555578601546586, 0.06453344994224608, 0.0662193070165813, 0.06444822205230594, 0.06477608694694936, 0.06417655595578253, 0.06440887995995581, 0.06461286311969161, 0.06411389005370438, 0.06434980500489473, 0.06419973983429372, 0.06441552308388054, 0.06458967598155141, 0.06432037311606109, 0.06466244393959641, 0.06460306094959378, 0.06446153903380036, 0.0645669458899647, 0.06426521902903914, 0.06461077718995512, 0.0641737598925829, 0.06407299288548529, 0.0643390181940049, 0.06462722294963896, 0.06448897602967918, 0.0642328888643533, 0.06439487007446587, 0.06476593995466828, 0.06448131799697876, 0.06422268506139517, 0.06423795106820762, 0.06447552680037916, 0.06439756322652102, 0.06473658885806799, 0.06441886397078633, 0.06431710813194513, 0.06498171598650515, 0.06417382089421153, 0.06419664598070085, 0.06456680409610271, 0.06427931599318981, 0.06453325506299734, 0.06434872490353882, 0.06438953406177461, 0.06455964385531843, 0.064401401206851, 0.06436327588744462, 0.06461365800350904, 0.06437816191464663, 0.06438102014362812, 0.06401187693700194, 0.06438568304292858, 0.06436277204193175, 0.06505752680823207, 0.06402075802907348, 0.06422675913199782, 0.0647145479451865, 0.06426102202385664, 0.06423566094599664, 0.06451171194203198, 0.06453216704539955, 0.06410246598534286, 0.06420248094946146, 0.06450413004495203, 0.06459814612753689, 0.06439688499085605, 0.06383424997329712, 0.06531130895018578, 0.06413811794482172, 0.06441647908650339, 0.06423708610236645, 0.06444840598851442, 0.06468955194577575, 0.0643169789109379, 0.06439760606735945, 0.0645930110476911, 0.06475894595496356, 0.06408685888163745, 0.06476018810644746, 0.06467714090831578, 0.06439170800149441, 0.06427017808891833, 0.06448082393035293, 0.06464347196742892, 0.06458465312607586, 0.06410229904577136, 0.0643398358952254, 0.06463498412631452, 0.0648691519163549, 0.06397820496931672, 0.06398513889871538, 0.06425365014001727, 0.06466756807640195, 0.06433205399662256, 0.06481963186524808, 0.06528650294058025, 0.06375810108147562, 0.06441705790348351, 0.06449792115017772, 0.0643839470576495, 0.0641690818592906, 0.0641836291179061, 0.06462815380655229, 0.06404493399895728, 0.06468255701474845, 0.06424574600532651, 0.06481271306984127, 0.06460950593464077, 0.06439031404443085, 0.06411109701730311, 0.06449002795852721, 0.06431238795630634, 0.06417550519108772, 0.06454405398108065, 0.06529873795807362, 0.064326151041314, 0.06465604691766202, 0.06512325094081461, 0.06374274496920407, 0.06435210211202502, 0.06494094105437398, 0.06545029394328594], [0.05117382504977286, 0.04745493084192276, 0.04750750702805817, 0.04747960506938398, 0.04770178603939712, 0.047544753877446055, 0.047473710030317307, 0.04767288500443101, 0.04756070813164115, 0.047956340946257114, 0.04823099006898701, 0.04723204183392227, 0.04742486705072224, 0.04742780909873545, 0.04762070090509951, 0.04766633594408631, 0.04760588798671961, 0.04781658318825066, 0.04756256798282266, 0.0476067999843508, 0.047566396882757545, 0.04752775304950774, 0.04757160693407059, 0.04767057904973626, 0.04753771494142711, 0.047521986067295074, 0.04747961298562586, 0.04820419708266854, 0.0475007479544729, 0.047929784981533885, 0.04949283809401095, 0.04919489100575447, 0.06366210500709713, 0.06532189878635108, 0.0638076791074127, 0.06436211289837956, 0.06473618908785284, 0.0637687030248344, 0.06448845495469868, 0.06440211297012866, 0.06429464416578412, 0.06433725194074214, 0.06579983606934547, 0.06463426887057722, 0.06447592307813466, 0.0644545559771359, 0.06465765298344195, 0.06469833105802536, 0.0643707390408963, 0.06437421985901892, 0.06438095099292696, 0.0639970819465816, 0.06432566419243813, 0.06477472488768399, 0.0643813710194081, 0.06498125800862908, 0.06495157489553094, 0.06426633009687066, 0.06419564504176378, 0.06442852783948183, 0.06429250002838671, 0.0642803271766752, 0.06460434501059353, 0.06466694292612374, 0.0647095579188317, 0.0652392921037972, 0.06541005382314324, 0.06533647701144218, 0.06604654411785305, 0.06488528591580689, 0.06552005000412464, 0.06559852999635041, 0.06523976009339094, 0.06541059003211558, 0.0654683809261769, 0.06513170013204217, 0.0653982327785343, 0.064962302101776, 0.0651657679118216, 0.06477603618986905, 0.06494548683986068, 0.06508575705811381, 0.06482504005543888, 0.06518575106747448, 0.06555454898625612, 0.06491895881481469, 0.06538023403845727, 0.06494712899439037, 0.06527454103343189, 0.06486047897487879, 0.06467644800432026, 0.06503185792826116, 0.06526100006885827, 0.0646889409981668, 0.06506311311386526, 0.06479584285989404, 0.06509110610932112, 0.06498597096651793, 0.06516044097952545, 0.06482183490879834, 0.06509905913844705, 0.06497817602939904, 0.0650387539062649, 0.06519564404152334, 0.06522247707471251, 0.06480514397844672, 0.06513632787391543, 0.06505327601917088, 0.06518755108118057, 0.0648494828492403, 0.06489417422562838, 0.06611179397441447, 0.06502059800550342, 0.06485014501959085, 0.06612937687896192, 0.06524648610502481, 0.06547000980935991, 0.065420150058344, 0.06550324894487858, 0.06507916213013232, 0.06524587492458522, 0.06599658308550715, 0.06533391401171684, 0.06478787795640528, 0.06514264503493905, 0.06475558690726757, 0.0653654511552304, 0.0651247447822243, 0.06585842813365161, 0.06415971391834319, 0.06513733812607825, 0.0647760359570384, 0.06494672293774784, 0.06481380900368094, 0.06522790598683059, 0.06475826911628246, 0.06532791093923151, 0.0652981020975858, 0.06541948695667088, 0.06472826702520251, 0.06561998487450182, 0.06512558995746076, 0.0652999880257994, 0.0648103840649128, 0.06559784500859678, 0.06444354192353785, 0.06445674202404916, 0.06426474405452609, 0.06448307889513671, 0.06410473003052175, 0.06494740908965468, 0.06414700485765934, 0.06433024816215038, 0.06412889203056693, 0.06433188589289784, 0.06439275108277798, 0.06441243481822312, 0.06456657918170094, 0.06425735983066261, 0.06433638418093324, 0.06433269800618291, 0.06399984192103148, 0.06450302596203983, 0.0642278459854424, 0.06438970193266869, 0.0651696021668613, 0.06927405297756195, 0.05907920398749411, 0.06862533604726195, 0.0669626269955188, 0.06489660404622555, 0.06489467993378639, 0.06488636299036443, 0.06492991093546152, 0.06508862599730492, 0.06483940011821687, 0.06488259299658239, 0.06507323193363845, 0.06486843596212566, 0.06510625197552145, 0.06500509195029736, 0.06548839621245861, 0.06502537382766604, 0.06484818598255515, 0.06514565902762115, 0.06464447104372084, 0.06498676398769021, 0.06495997193269432, 0.06513645919039845, 0.06511357193812728, 0.0646533789113164, 0.06489135813899338, 0.06505667301826179, 0.0648014999460429, 0.06545881903730333, 0.06519390200264752, 0.06496226694434881, 0.06496886303648353, 0.06488644587807357, 0.06521465303376317, 0.06460749893449247, 0.06508023617789149, 0.06470703985542059, 0.06490075099281967, 0.06507597304880619, 0.06476266006939113, 0.06467299186624587, 0.06518817716278136, 0.06497506378218532, 0.0650048041716218, 0.06476014200598001, 0.06511877989396453, 0.06494133314117789, 0.06497435388155282, 0.0652733319438994, 0.06507756398059428, 0.0651233121752739, 0.06510287593118846, 0.06502324505709112, 0.06492056488059461, 0.06486884411424398, 0.06508896383456886, 0.06493118405342102, 0.06523227016441524, 0.06492978893220425, 0.06501962896436453, 0.06514270207844675, 0.0649710288271308, 0.06494774506427348, 0.06489988905377686, 0.06508410605601966, 0.06448693480342627, 0.06500282813794911, 0.06507263798266649, 0.06500210496596992, 0.06533238501287997, 0.06500257598236203, 0.06491060508415103, 0.06481584091670811, 0.06516771810129285, 0.06612686184234917, 0.06477955798618495, 0.06520481314510107, 0.06476543797180057, 0.06526796193793416, 0.06501693511381745, 0.06473553692921996, 0.06580131594091654, 0.06506352196447551, 0.0651416010223329, 0.06472934107296169, 0.0651051530148834, 0.06524773593991995, 0.06491767312400043, 0.06506547681055963, 0.06507886899635196, 0.0651441200170666, 0.06507325102575123, 0.06500847917050123, 0.06489770882762969, 0.06511415797285736, 0.06493013701401651, 0.06503377016633749, 0.06482770084403455, 0.0652088881470263, 0.06486029992811382, 0.06486648204736412, 0.06555492687039077, 0.06493775709532201, 0.0650888760574162, 0.06495746481232345, 0.0649617980234325, 0.06508496613241732, 0.06488812388852239, 0.06496344902552664, 0.06528022699058056, 0.06507599516771734, 0.06510102096945047, 0.06502549396827817, 0.06502777594141662, 0.06558530894108117, 0.06447685114108026, 0.06628683302551508, 0.06442217598669231, 0.06477393000386655, 0.06419167690910399, 0.06439230591058731, 0.064749764977023, 0.0639755860902369, 0.06434258911758661, 0.0642012048047036, 0.06441644602455199, 0.06459063710644841, 0.06432243785820901, 0.06466491403989494, 0.0645979130640626, 0.06445787590928376, 0.06456986698321998, 0.06426969519816339, 0.06461159884929657, 0.06417027115821838, 0.06407272885553539, 0.06447165412828326, 0.06449307990260422, 0.0644971551373601, 0.06423070491291583, 0.06439369102008641, 0.0647677918896079, 0.06447852402925491, 0.06421935302205384, 0.06423544907011092, 0.06447623297572136, 0.06440078490413725, 0.06473932997323573, 0.06441419315524399, 0.0643153260461986, 0.06498610391281545, 0.0641711400821805, 0.0641950259450823, 0.06456692493520677, 0.06428513093851507, 0.06453790701925755, 0.0643499051220715, 0.064378707902506, 0.06456359499134123, 0.0644023169297725, 0.06436014105565846, 0.0646112950053066, 0.06438476196490228, 0.06438206718303263, 0.06401520292274654, 0.06438529305160046, 0.06437125895172358, 0.06475957203656435, 0.06430547893978655, 0.0642363359220326, 0.06470625707879663, 0.06425980990752578, 0.06423464603722095, 0.06451623514294624, 0.06453383900225163, 0.06410005991347134, 0.0642049559392035, 0.06450089509598911, 0.06460226490162313, 0.06439296016469598, 0.06353519693948328, 0.06561516900546849, 0.06414129794575274, 0.06441284390166402, 0.06423252820968628, 0.06445387494750321, 0.06479047192260623, 0.0642155830282718, 0.0643959550652653, 0.06460151099599898, 0.06475052190944552, 0.0640925420448184, 0.06475528702139854, 0.06469437992200255, 0.06452081212773919, 0.06412754696793854, 0.06447749002836645, 0.06465262081474066, 0.06473537720739841, 0.06394924176856875, 0.06433524214662611, 0.0646425080485642, 0.06487915082834661, 0.06396661209873855, 0.06398603203706443, 0.0642517248634249, 0.06467020302079618, 0.06432517315261066, 0.06482731201685965, 0.06531254481524229, 0.06373053300194442, 0.06441727909259498, 0.06449344009160995, 0.06438516988418996, 0.06417331495322287, 0.06419564201496542, 0.0646200180053711, 0.06404145201668143, 0.06468188110738993, 0.06425075093284249, 0.06480839988216758, 0.06461073621176183, 0.06439102883450687, 0.06411263905465603, 0.06448491010814905, 0.06431729486212134, 0.06432989914901555, 0.06438203295692801, 0.06530574196949601, 0.06432121386751533, 0.06465839501470327, 0.06514915497973561, 0.06372695602476597, 0.06434582802467048, 0.06493165297433734, 0.06550331204198301, 0.06833906099200249, 0.05947884195484221, 0.0691557580139488, 0.06534051615744829, 0.06496882787905633, 0.06493660295382142, 0.06507441913709044, 0.06468118587508798, 0.06489481218159199, 0.0650571899022907, 0.06510427803732455, 0.06552748894318938, 0.06432884791865945, 0.06473899516277015, 0.06546604493632913, 0.06478742999024689, 0.06488624005578458, 0.06473303283564746, 0.0651368610560894, 0.0651332309935242, 0.06534220208413899, 0.06501811789348722, 0.06521607609465718, 0.0647707290481776, 0.0652082518208772, 0.06481665815226734, 0.06492129387333989, 0.06493213516660035, 0.06475533684715629, 0.06504077697172761, 0.06511253514327109, 0.06450245599262416, 0.06485341605730355, 0.06497398391366005, 0.06520868302322924, 0.06501483707688749, 0.06493399990722537, 0.06497594807296991, 0.06488496880047023, 0.06500973412767053, 0.06502831587567925, 0.06464309408329427, 0.06511871214024723, 0.0648027709685266, 0.06503407191485167, 0.06472232402302325, 0.06560043804347515, 0.0656870398670435, 0.06476929318159819, 0.06481460388749838, 0.0654290309175849, 0.06469078012742102, 0.06511827697977424, 0.06493795407004654, 0.06522669200785458, 0.06450312701053917, 0.06479240395128727, 0.06473736395128071, 0.06501330388709903, 0.06511666113510728, 0.06513840891420841, 0.0649586170911789, 0.06500903191044927, 0.06462908606044948, 0.06485589896328747, 0.06510611507110298, 0.06503470498137176, 0.0643184909131378, 0.06520343013107777, 0.06502131093293428, 0.06537920399568975, 0.06450383109040558, 0.06513720797374845, 0.0652279460337013, 0.0649599488824606, 0.0652074720710516, 0.06479386100545526, 0.06462847301736474, 0.0648145719897002, 0.06474400195293128, 0.06501482706516981, 0.06458095484413207, 0.06507317395880818, 0.06492055719718337, 0.06491528498008847, 0.06527635897509754, 0.06583121488802135, 0.06518129399046302, 0.06500685797072947, 0.06528069521300495, 0.06504864688031375, 0.06485624890774488, 0.06487756222486496, 0.0648130439221859, 0.06544660707004368, 0.06519996491260827, 0.06526064593344927, 0.06482825614511967, 0.06501346896402538, 0.06459897686727345, 0.06516262399964035, 0.06495510111562908, 0.06504239793866873, 0.0648982459679246, 0.06494454713538289, 0.06487997598014772, 0.06505359499715269, 0.0648284109774977, 0.06593354884535074], [0.06373188714496791, 0.06530104298144579, 0.06319810403510928, 0.06435550493188202, 0.06473382795229554, 0.06391447503119707, 0.06434057210572064, 0.06436629686504602, 0.06420415895991027, 0.06433422607369721, 0.06593495910055935, 0.06463553989306092, 0.06448249495588243, 0.06443391717039049, 0.06464573089033365, 0.06457512197084725, 0.06452234811149538, 0.06414873781614006, 0.06456123199313879, 0.06402941304259002, 0.06431895401328802, 0.06474902294576168, 0.06439984519965947, 0.06495669693686068, 0.0649714320898056, 0.06423880881629884, 0.06419113907031715, 0.06446842500008643, 0.06437157094478607, 0.06420026998966932, 0.06456489395350218, 0.06469412008300424, 0.0648246391210705, 0.06511177588254213, 0.06550029013305902, 0.06523828185163438, 0.06616671103984118, 0.06477817497216165, 0.06551230605691671, 0.06561324396170676, 0.06523149809800088, 0.06541561591438949, 0.06550606293603778, 0.06508415215648711, 0.06539500900544226, 0.06495592999272048, 0.06517460895702243, 0.06476376694627106, 0.06493276497349143, 0.06510431505739689, 0.06482363701798022, 0.06519112805835903, 0.06554542388767004, 0.0649211760610342, 0.06533435406163335, 0.06498983292840421, 0.06538578192703426, 0.06473611597903073, 0.06468853610567749, 0.06503077899105847, 0.06525211199186742, 0.06468162895180285, 0.06507757189683616, 0.06480210716836154, 0.06509309588000178, 0.06495323614217341, 0.06519216601736844, 0.06481781881302595, 0.06508450699038804, 0.06498458003625274, 0.06500009214505553, 0.06524578388780355, 0.06520950212143362, 0.06482208496890962, 0.06513329804874957, 0.06504697981290519, 0.0651770259719342, 0.06482705101370811, 0.06504944013431668, 0.06598200998269022, 0.06500411895103753, 0.0648663379251957, 0.06612704508006573, 0.06524182204157114, 0.06546354503370821, 0.06541890790686011, 0.06550844805315137, 0.06507550505921245, 0.06524710496887565, 0.06598334293812513, 0.06530492403544486, 0.06482516299001873, 0.06513760192319751, 0.06500374712049961, 0.06511410605162382, 0.0651307669468224, 0.06585947005078197, 0.06427639699541032, 0.06515381997451186, 0.06464418093673885, 0.06534852110780776, 0.06441218499094248, 0.06522845989093184, 0.06476593902334571, 0.06545264599844813, 0.06516593089327216, 0.06541601498611271, 0.06471945019438863, 0.0656288608442992, 0.06511675799265504, 0.06512118596583605, 0.06500513502396643, 0.06558542209677398, 0.0645768919494003, 0.06444049812853336, 0.06428180495277047, 0.0644849338568747, 0.06410093000158668, 0.06494954507797956, 0.06414955807849765, 0.0643284348770976, 0.06412418512627482, 0.06433878699317575, 0.06438710889779031, 0.06441141804680228, 0.06445798906497657, 0.0643660849891603, 0.06431389204226434, 0.06434107199311256, 0.06400329479947686, 0.0644978720229119, 0.06422328017652035, 0.06439022580161691, 0.06517646601423621, 0.06921507813967764, 0.058927007019519806, 0.06883149384520948, 0.06696797511540353, 0.0648892899043858, 0.06479339301586151, 0.06499281199648976, 0.06493826396763325, 0.06508463202044368, 0.0648399309720844, 0.06488656019791961, 0.06507531693205237, 0.06485772598534822, 0.06508807907812297, 0.0650173369795084, 0.06549248192459345, 0.0650243260897696, 0.06485213781706989, 0.06514289509505033, 0.06464781006798148, 0.06489622592926025, 0.06504715490154922, 0.06512938719242811, 0.06511519593186677, 0.06464374694041908, 0.0648672659881413, 0.06509432895109057, 0.06480622803792357, 0.0654486280400306, 0.06518870498985052, 0.0649637090973556, 0.06496688886545599, 0.06488986010663211, 0.06521942093968391, 0.06460129003971815, 0.06508429488167167, 0.0646842741407454, 0.0649231649003923, 0.0649750700686127, 0.06485750689171255, 0.06466686516068876, 0.06519106193445623, 0.06498139398172498, 0.06500018411315978, 0.06475848890841007, 0.06502308510243893, 0.06503694388084114, 0.06495981803163886, 0.06528802891261876, 0.06507110712118447, 0.0651216630358249, 0.06511036190204322, 0.06501450994983315, 0.06489594001322985, 0.06488896906375885, 0.06508323596790433, 0.06493366393260658, 0.06522542214952409, 0.06493803998455405, 0.06503472104668617, 0.06513165892101824, 0.0649656190071255, 0.06495249597355723, 0.06488728593103588, 0.06509765516966581, 0.0644818339496851, 0.06501002493314445, 0.06507717096246779, 0.06499025109224021, 0.06532320706173778, 0.06490414892323315, 0.06501206406392157, 0.0648226838093251, 0.06517177913337946, 0.066135621862486, 0.06478004902601242, 0.06518831313587725, 0.0649128828663379, 0.06512183416634798, 0.06501427083276212, 0.06473260000348091, 0.06580778700299561, 0.06506805610843003, 0.0651370829436928, 0.06471201102249324, 0.06511679105460644, 0.06515220692381263, 0.06501144403591752, 0.06506654690019786, 0.06509024719707668, 0.06514161382801831, 0.06504531996324658, 0.06502984603866935, 0.06489579798653722, 0.06510942918248475, 0.06492857681587338, 0.06486156303435564, 0.06500364700332284, 0.06521621393039823, 0.06483981315977871, 0.06488614692352712, 0.06554320408031344, 0.06493928795680404, 0.06508557102642953, 0.06496542389504611, 0.06496770516969264, 0.06507696583867073, 0.06488525215536356, 0.06496215099468827, 0.06528417393565178, 0.06506920489482582, 0.06509822909720242, 0.06503163604065776, 0.06501432601362467, 0.06552441883832216, 0.06461085099726915, 0.06595422606915236, 0.0644249590113759, 0.06484690099023283, 0.06431924691423774, 0.06433561211451888, 0.064672633074224, 0.0641253418289125, 0.0643871040083468, 0.0641991039738059, 0.06440939102321863, 0.0645942841656506, 0.06420190399512649, 0.06475688982754946, 0.06462336611002684, 0.06446210388094187, 0.0645677880384028, 0.06424763216637075, 0.06462892680428922, 0.06416396517306566, 0.06407775892876089, 0.06463760789483786, 0.06432652520015836, 0.0644868069794029, 0.06423403299413621, 0.06440125196240842, 0.06476715207099915, 0.06446388084441423, 0.06413382198661566, 0.06434745108708739, 0.06447299802675843, 0.06440035789273679, 0.06473878398537636, 0.06442029704339802, 0.06431793700903654, 0.06497499300166965, 0.06418224703520536, 0.06414446607232094, 0.06461402587592602, 0.0642723380587995, 0.06453377706930041, 0.06432849587872624, 0.06441005598753691, 0.06456848699599504, 0.06439733505249023, 0.06436353106983006, 0.06460911082103848, 0.06436539115384221, 0.06438887398689985, 0.06401254795491695, 0.06439141510054469, 0.06436095084063709, 0.06479780399240553, 0.06428562011569738, 0.06409201491624117, 0.06484460202045739, 0.06425323593430221, 0.06424103607423604, 0.06451507890596986, 0.06452509411610663, 0.0640998580493033, 0.06421027798205614, 0.06450120895169675, 0.0645964581053704, 0.06440035393461585, 0.06426965701393783, 0.06485727801918983, 0.06415617791935802, 0.06441637012176216, 0.06423248094506562, 0.0644488378893584, 0.06457941513508558, 0.06442358484491706, 0.06439733202569187, 0.06460035708732903, 0.0647523938678205, 0.06407948420383036, 0.06477075489237905, 0.06467402609996498, 0.0643932749517262, 0.06427287892438471, 0.06447271700017154, 0.06464440491981804, 0.06445262115448713, 0.06423352588899434, 0.06433293200097978, 0.06463092798367143, 0.06486811605282128, 0.06398077704943717, 0.06398501503281295, 0.06426675501279533, 0.06465659802779555, 0.06433887290768325, 0.0648054399061948, 0.06526683596894145, 0.06375888711772859, 0.0644468639511615, 0.06449923804029822, 0.06438002898357809, 0.0641747119370848, 0.06417929101735353, 0.06462112395092845, 0.06405196501873434, 0.06468121102079749, 0.06423407210968435, 0.06481040804646909, 0.06460089888423681, 0.06440328806638718, 0.06412341305986047, 0.06448462279513478, 0.06429268815554678, 0.06419463991187513, 0.0645463599357754, 0.06530013307929039, 0.06432694313116372, 0.06463209493085742, 0.06495157885365188, 0.06392940902151167, 0.06435709213837981, 0.0649502829182893, 0.06542181200347841, 0.0685807999689132, 0.05992172099649906, 0.06872082105837762, 0.0653353719972074, 0.06496914499439299, 0.0649307700805366, 0.06508599291555583, 0.06468038400635123, 0.06488540093414485, 0.06505445810034871, 0.06511148205026984, 0.06544459890574217, 0.06443177908658981, 0.06471377983689308, 0.0654165621381253, 0.06482072500512004, 0.06489799590781331, 0.06473395205102861, 0.06514030904509127, 0.0651311338879168, 0.06534607615321875, 0.06501860287971795, 0.06519144214689732, 0.0647510769777, 0.06524161482229829, 0.06482366612181067, 0.0649235409218818, 0.06492666411213577, 0.06475884304381907, 0.06503660697489977, 0.06509666098281741, 0.06441434193402529, 0.06495292810723186, 0.06498245988041162, 0.06520240800455213, 0.0650137891061604, 0.06494449893943965, 0.06496856804005802, 0.06489092903211713, 0.06500840093940496, 0.06500260788016021, 0.06466415314935148, 0.06511915381997824, 0.06480299611575902, 0.0650333680678159, 0.06482268893159926, 0.06554893590509892, 0.06563331210054457, 0.06477339286357164, 0.06482258113101125, 0.06542104599066079, 0.064690045081079, 0.06512489379383624, 0.06493652216158807, 0.06522845500148833, 0.06450273492373526, 0.06469569308683276, 0.0648283299524337, 0.06499421503394842, 0.06513678305782378, 0.0649942229501903, 0.06510091898962855, 0.0650120738428086, 0.06462857313454151, 0.06485270708799362, 0.06510243285447359, 0.06503607402555645, 0.06431802897714078, 0.06520743598230183, 0.0650222641415894, 0.06521565187722445, 0.06466589705087245, 0.06513610808178782, 0.06522368593141437, 0.06496256706304848, 0.06520224991254508, 0.06479912903159857, 0.0646251488942653, 0.06482252897694707, 0.06473226402886212, 0.06501413695514202, 0.06463126302696764, 0.06502747698687017, 0.0649242380168289, 0.06491902517154813, 0.06527007487602532, 0.06582926609553397, 0.06518674385733902, 0.0647268199827522, 0.06555489706806839, 0.06505058705806732, 0.064861147897318, 0.06487000011838973, 0.06482200906611979, 0.06543833180330694, 0.06519931298680604, 0.06526265107095242, 0.06482790992595255, 0.06500710500404239, 0.06460082298144698, 0.06516657304018736, 0.06495628296397626, 0.06502339104190469, 0.06480629416182637, 0.06505241990089417, 0.06488059391267598, 0.06504774419590831, 0.06483077886514366, 0.06591671495698392, 0.07544454396702349, 0.06839439808391035, 0.06528565706685185, 0.06468771095387638, 0.06497640791349113, 0.06542883114889264, 0.06371380388736725, 0.06533224694430828, 0.06510828016325831, 0.06414902093820274, 0.06504551297985017, 0.06539352401159704, 0.06528356997296214, 0.06512767798267305, 0.06516588712111115, 0.06516796397045255, 0.06512535805813968, 0.06521746190264821, 0.06507684593088925, 0.06592909502796829, 0.06430077506229281, 0.06526902387849987, 0.06507337116636336, 0.0650756808463484, 0.06555841909721494, 0.06528673809953034, 0.0651190048083663, 0.06522397813387215, 0.06494905101135373, 0.06504969298839569, 0.06640127999708056], [0.06760417716577649, 0.06531095295213163, 0.06351143284700811, 0.06434772606007755, 0.06470480700954795, 0.06394705711863935, 0.06432490283623338, 0.0644266630988568, 0.06422359007410705, 0.06432796199806035, 0.06587225780822337, 0.06463325605727732, 0.06448655901476741, 0.06443233694881201, 0.06466718506999314, 0.06454957602545619, 0.06452120887115598, 0.06417823419906199, 0.06459550093859434, 0.06397043890319765, 0.06436748406849802, 0.06471575796604156, 0.06438625091686845, 0.06497528217732906, 0.06494842283427715, 0.06427418114617467, 0.06419239286333323, 0.06441601901315153, 0.06441466696560383, 0.06421724520623684, 0.0645500069949776, 0.06470778887160122, 0.06475927005521953, 0.06504238490015268, 0.06551668513566256, 0.06520309089682996, 0.06607191707007587, 0.06488372106105089, 0.06551642087288201, 0.06561810011044145, 0.06526605901308358, 0.0653667168226093, 0.06558098108507693, 0.06505914707668126, 0.0653817499987781, 0.06494470196776092, 0.06516217486932874, 0.06474042613990605, 0.06496932986192405, 0.06508627720177174, 0.06481941300444305, 0.06520142778754234, 0.06554663111455739, 0.064908925909549, 0.06538047618232667, 0.06495806691236794, 0.06539288605563343, 0.06472051097080112, 0.06469873199239373, 0.06497127702459693, 0.06529967905953526, 0.06467861589044333, 0.06508034793660045, 0.06479322398081422, 0.06509235198609531, 0.06497231218963861, 0.06518349889665842, 0.06479532201774418, 0.0650958870537579, 0.06496694195084274, 0.06508919596672058, 0.06520685204304755, 0.06517275609076023, 0.06482995697297156, 0.06511244201101363, 0.06502948189154267, 0.06519557000137866, 0.0648241660092026, 0.06493560690432787, 0.06612948817200959, 0.06499345391057432, 0.06486244406551123, 0.06610086001455784, 0.06525594787672162, 0.06546375202015042, 0.06540743005461991, 0.06550790392793715, 0.06509010912850499, 0.06526466901414096, 0.06596501800231636, 0.0653313398361206, 0.06480628112331033, 0.06510334485210478, 0.06492770020850003, 0.06521717784926295, 0.0651215361431241, 0.06583622400648892, 0.06428074883297086, 0.06504097813740373, 0.06475220201537013, 0.06506338994950056, 0.06471355305984616, 0.06519695883616805, 0.06477728905156255, 0.06521405908279121, 0.06543504889123142, 0.06538754398934543, 0.06471797893755138, 0.06566220708191395, 0.0651252989191562, 0.06511218915693462, 0.06503227399662137, 0.06554729002527893, 0.0644233028870076, 0.06443433905951679, 0.06426848005503416, 0.06448870780877769, 0.06412258418276906, 0.06444170186296105, 0.06462938408367336, 0.06428836705163121, 0.06412387685850263, 0.06436270195990801, 0.06438012118451297, 0.06440253392793238, 0.06460303394123912, 0.06425474304705858, 0.06427142512984574, 0.06429825886152685, 0.06399645400233567, 0.06451856601051986, 0.06420117407105863, 0.06441622995771468, 0.06491570407524705, 0.06943997298367321, 0.05909158801659942, 0.06873922795057297, 0.06702340696938336, 0.0648684399202466, 0.06484366697259247, 0.06490917317569256, 0.06494625192135572, 0.06511582899838686, 0.06478198803961277, 0.06491968501359224, 0.06511081894859672, 0.0648500279057771, 0.06509426608681679, 0.0649840768892318, 0.06549441604875028, 0.06501489016227424, 0.0648531150072813, 0.06511422782205045, 0.0646963631734252, 0.0648792868014425, 0.06504794303327799, 0.0651678501162678, 0.0651203019078821, 0.06463148910552263, 0.06485721096396446, 0.06507774908095598, 0.06480004894547164, 0.06544821104034781, 0.06517448998056352, 0.06495493394322693, 0.0649750679731369, 0.06491254409775138, 0.06522401398979127, 0.06459704297594726, 0.06512487097643316, 0.06464600889012218, 0.06492180516943336, 0.06504377583041787, 0.06475894502364099, 0.06469576712697744, 0.06517978082410991, 0.06497048819437623, 0.06501649180427194, 0.06475012912414968, 0.06509365001693368, 0.06491634505800903, 0.06501342193223536, 0.06527097197249532, 0.06506295595318079, 0.06515149306505919, 0.06510088592767715, 0.06499763997271657, 0.06488344096578658, 0.06487521110102534, 0.0650836550630629, 0.0649531448725611, 0.06521281809546053, 0.06496375193819404, 0.06499939016066492, 0.06515710288658738, 0.0649569109082222, 0.06495471717789769, 0.06487559783272445, 0.06511872308328748, 0.06447111605666578, 0.06500598089769483, 0.06508016213774681, 0.06502201384864748, 0.0652876531239599, 0.06491462490521371, 0.06500159995630383, 0.06480832910165191, 0.06518753990530968, 0.06612992007285357, 0.06483983807265759, 0.0651110929902643, 0.0649127229116857, 0.06511721992865205, 0.06504978216253221, 0.06469858693890274, 0.06582162901759148, 0.06508549698628485, 0.06517706601880491, 0.06463580089621246, 0.06515297503210604, 0.0652478861156851, 0.06491318787448108, 0.06508065504021943, 0.06508802995085716, 0.06514138705097139, 0.06503974692896008, 0.06502769002690911, 0.06489005498588085, 0.06511600711382926, 0.06494344491511583, 0.06504632299765944, 0.06480014510452747, 0.0652059658896178, 0.06489662802778184, 0.06483481707982719, 0.06554352585226297, 0.06492739310488105, 0.06505517405457795, 0.06500573689118028, 0.06495729298330843, 0.06506344606168568, 0.06488150707446039, 0.06495859287679195, 0.06529593514278531, 0.0650756738614291, 0.06507945596240461, 0.06505588302388787, 0.06505091208964586, 0.06543588102795184, 0.06475582486018538, 0.06522763497196138, 0.06528527103364468, 0.06462171603925526, 0.06428342894650996, 0.06442670011892915, 0.06462260894477367, 0.06402298505418003, 0.06441183993592858, 0.06418111105449498, 0.06442890386097133, 0.06460708100348711, 0.06431738403625786, 0.06462410395033658, 0.06462411116808653, 0.06455592904239893, 0.06449071201495826, 0.06422325898893178, 0.06463430589064956, 0.06414331099949777, 0.06410668604075909, 0.06442890292964876, 0.06449980800971389, 0.0644949390552938, 0.06421975698322058, 0.06442445795983076, 0.06478947214782238, 0.06442796485498548, 0.06425298401154578, 0.06425365409813821, 0.06448996881954372, 0.06442482210695744, 0.0646781949326396, 0.06444225716404617, 0.06435312191024423, 0.0649144109338522, 0.06423088698647916, 0.06410343013703823, 0.06466337596066296, 0.0642671580426395, 0.06450871098786592, 0.06431021098978817, 0.06439737905748188, 0.06461992487311363, 0.06439307006075978, 0.06433587986975908, 0.06459669210016727, 0.06436513504013419, 0.06442778091877699, 0.06399764493107796, 0.06442007515579462, 0.06430558883585036, 0.0648041081149131, 0.06433593691326678, 0.06415919517166913, 0.06476502981968224, 0.06423983513377607, 0.06423849589191377, 0.06452693697065115, 0.06449100817553699, 0.06412938702851534, 0.06418391596525908, 0.06452839192934334, 0.06458999589085579, 0.06438894919119775, 0.06437812396325171, 0.06471757194958627, 0.06415121397003531, 0.0644471300765872, 0.06425273092463613, 0.0644336799159646, 0.06472728704102337, 0.06424958514980972, 0.0644139340147376, 0.06461831997148693, 0.06468766694888473, 0.06421810807660222, 0.06466937600634992, 0.06468764087185264, 0.0643460329156369, 0.06428901921026409, 0.06452281400561333, 0.06459312979131937, 0.06455529411323369, 0.06409318000078201, 0.06438385788351297, 0.06461793906055391, 0.06480592815205455, 0.06400825991295278, 0.06397628993727267, 0.06432281900197268, 0.0646164589561522, 0.06432332913391292, 0.06483488297089934, 0.06470188708044589, 0.06430320884101093, 0.0644748869817704, 0.0644655111245811, 0.06436358694918454, 0.06417185789905488, 0.06422632304020226, 0.06461315508931875, 0.06405223300680518, 0.06467397790402174, 0.06421978399157524, 0.06479802913963795, 0.06464099092409015, 0.06434374605305493, 0.06414759089238942, 0.06450115400366485, 0.06431224104017019, 0.06420328002423048, 0.06454620696604252, 0.06528484006412327, 0.06432732893154025, 0.06461168499663472, 0.06466114101931453, 0.06426119711250067, 0.0643393809441477, 0.06493524205870926, 0.06517123291268945, 0.06884680106304586, 0.05979766184464097, 0.06891499203629792, 0.06531197996810079, 0.06499292212538421, 0.06491774506866932, 0.06510350783355534, 0.06467714300379157, 0.06486692908219993, 0.06506959907710552, 0.06512383394874632, 0.06515176990069449, 0.06478073098696768, 0.06461161701008677, 0.065317926928401, 0.0649525502230972, 0.06489111296832561, 0.06473693298175931, 0.06510547082871199, 0.06516018719412386, 0.06538974493741989, 0.06496815197169781, 0.06533245393075049, 0.06482392409816384, 0.06501940009184182, 0.06480378285050392, 0.06498390203341842, 0.0648810479324311, 0.0647946810349822, 0.06501327594742179, 0.06509866309352219, 0.06451953900977969, 0.06481676106341183, 0.06500410987064242, 0.06519144307821989, 0.06502714287489653, 0.06494788406416774, 0.06498500611633062, 0.0648576789535582, 0.0650212129112333, 0.06498267804272473, 0.06465452094562352, 0.06515315896831453, 0.06482950318604708, 0.06500989198684692, 0.0650635960046202, 0.06540493993088603, 0.06551321898587048, 0.06479247892275453, 0.06483870605006814, 0.06541774398647249, 0.06467658700421453, 0.0651256290730089, 0.06489981687627733, 0.06523352116346359, 0.06454090191982687, 0.06475022598169744, 0.0647842581383884, 0.06498250877484679, 0.06514026806689799, 0.06515161297284067, 0.06495416909456253, 0.06496951496228576, 0.06461771694011986, 0.06487975316122174, 0.06509260786697268, 0.06506360112689435, 0.06428225501440465, 0.06524705700576305, 0.06501188292168081, 0.06519367499276996, 0.06464317603968084, 0.065176592906937, 0.06520844902843237, 0.06497664703056216, 0.06518285907804966, 0.06483188085258007, 0.06460020504891872, 0.06480456795543432, 0.06475006812252104, 0.06498901383019984, 0.06476763519458473, 0.06491090985946357, 0.06493476708419621, 0.06490194401703775, 0.06528566498309374, 0.06581657496280968, 0.06518433010205626, 0.06497987592592835, 0.0653134728781879, 0.0650553461164236, 0.06485007889568806, 0.06488871318288147, 0.06478252587839961, 0.06548487790860236, 0.0651788511313498, 0.06523607298731804, 0.0648375388700515, 0.06500696018338203, 0.06461944687180221, 0.06517482595518231, 0.06492875516414642, 0.06501242192462087, 0.06490036891773343, 0.06495373300276697, 0.0648579690605402, 0.06507557607255876, 0.06483940803445876, 0.06565568782389164, 0.07568531110882759, 0.06846636394038796, 0.06530004716478288, 0.06456442386843264, 0.06513893511146307, 0.06496584485284984, 0.06425481499172747, 0.06505018216557801, 0.06512187095358968, 0.06423319200985134, 0.06505725486204028, 0.06540201511234045, 0.06522773904725909, 0.06529450183734298, 0.06505006016232073, 0.06518381484784186, 0.06509733013808727, 0.0652180330362171, 0.06506498786620796, 0.06488988199271262, 0.06533649214543402, 0.0654174517840147, 0.06493012118153274, 0.0651140648405999, 0.06557953916490078, 0.06522760284133255, 0.06527712708339095, 0.06517483596689999, 0.06486356793902814, 0.06516982498578727, 0.06596528901718557, 0.06836452218703926], [0.0643661969806999, 0.06473058206029236, 0.0636646798811853, 0.06460091308690608, 0.06439687893725932, 0.06415186705999076, 0.06433531106449664, 0.06595378299243748, 0.06463084183633327, 0.06447527301497757, 0.06445366819389164, 0.06465601501986384, 0.06455759890377522, 0.06451619789004326, 0.06411340017803013, 0.06449536001309752, 0.06399975693784654, 0.06450521689839661, 0.0645945779979229, 0.0645270380191505, 0.06497879908420146, 0.06481019896455109, 0.0644233908969909, 0.06418431410565972, 0.0644290519412607, 0.06439051311463118, 0.06422848091460764, 0.06457155593670905, 0.06470166007056832, 0.06459533097222447, 0.06530526396818459, 0.06551494006998837, 0.06524008605629206, 0.06603883299976587, 0.06489057000726461, 0.06551981391385198, 0.0655924859456718, 0.06524860416539013, 0.06541188503615558, 0.06557068997062743, 0.0650234199129045, 0.06539560505189002, 0.06496952590532601, 0.06516035692766309, 0.0647776541300118, 0.06495740893296897, 0.06507314206101, 0.06484810495749116, 0.06516801798716187, 0.06556737795472145, 0.0649041491560638, 0.06538889883086085, 0.06495025521144271, 0.06518724886700511, 0.06494142604060471, 0.06468031206168234, 0.06503450591117144, 0.06525949900969863, 0.06469068094156682, 0.06506234011612833, 0.06479519791901112, 0.0650907380040735, 0.06498750508762896, 0.06514969887211919, 0.06483243312686682, 0.06509345700033009, 0.06498154392465949, 0.06504650507122278, 0.06518718204461038, 0.06522776396013796, 0.06481461995281279, 0.06511840899474919, 0.06506623607128859, 0.06518143601715565, 0.0648510039318353, 0.06489057606086135, 0.06611803686246276, 0.06502193701453507, 0.06484744301997125, 0.06613408611156046, 0.06523999897763133, 0.06547659984789789, 0.06541865202598274, 0.06550904805772007, 0.06508440198376775, 0.06523955892771482, 0.06599689205177128, 0.06534028705209494, 0.06478222692385316, 0.0651462790556252, 0.0648629548959434, 0.06526496610604227, 0.06512089702300727, 0.06585610006004572, 0.0642785020172596, 0.06502016098238528, 0.06477084080688655, 0.06502171908505261, 0.06474463711492717, 0.06522555486299098, 0.06475828913971782, 0.06532395980320871, 0.06530830799601972, 0.06541024008765817, 0.06473605101928115, 0.06561497203074396, 0.06513570505194366, 0.06513252295553684, 0.06495538004674017, 0.0656181019730866, 0.06448277202434838, 0.06446449900977314, 0.0642986549064517, 0.06444783904589713, 0.0641013509593904, 0.06489874189719558, 0.06421343609690666, 0.06431112089194357, 0.06413714913651347, 0.06437301100231707, 0.06434985296800733, 0.0644097060430795, 0.06457539298571646, 0.06427950505167246, 0.06431871699169278, 0.06433116993866861, 0.06395833590067923, 0.06449583801440895, 0.06422900618053973, 0.06438958691433072, 0.06520907208323479, 0.06907855486497283, 0.058882784098386765, 0.06881040590815246, 0.06696340208873153, 0.0648937770165503, 0.06489639193750918, 0.06488995300605893, 0.06492563593201339, 0.06513665709644556, 0.06478770193643868, 0.0648861899971962, 0.06507487897761166, 0.06486345012672246, 0.06510566384531558, 0.06501049012877047, 0.06548829399980605, 0.06502784695476294, 0.0648459349758923, 0.06514822295866907, 0.06464018300175667, 0.06481102807447314, 0.06513642286881804, 0.06514029111713171, 0.06511507206596434, 0.06465443898923695, 0.06489037699066103, 0.06505607790313661, 0.06479875696823001, 0.06545986700803041, 0.06519552902318537, 0.06496364111080766, 0.06497024884447455, 0.06488565215840936, 0.06521385884843767, 0.06460935901850462, 0.06507603311911225, 0.0647073988802731, 0.06495484616607428, 0.06502966489642859, 0.06475957995280623, 0.06467176810838282, 0.06519383587874472, 0.06497134198434651, 0.06500468798913062, 0.0647582330275327, 0.06512106815353036, 0.06494334596209228, 0.06498323893174529, 0.06526514701545238, 0.06507788388989866, 0.06512770103290677, 0.06509759812615812, 0.06502221385017037, 0.06494941702112556, 0.06484621902927756, 0.06509121600538492, 0.06492986204102635, 0.06523205991834402, 0.0649303940590471, 0.06501565198414028, 0.06514179497025907, 0.06497223908081651, 0.06495283008553088, 0.06489828578196466, 0.06508375122211874, 0.06448697578161955, 0.06500442908145487, 0.06506696599535644, 0.06500408193096519, 0.06534391618333757, 0.06478479690849781, 0.06512114498764277, 0.06481291307136416, 0.06516967993229628, 0.066124432021752, 0.06481745792552829, 0.06517782411538064, 0.06489995704032481, 0.06512492895126343, 0.06502277194522321, 0.06472959811799228, 0.06580263492651284, 0.06506318086758256, 0.06514181010425091, 0.06473179091699421, 0.06510667502880096, 0.06524633406661451, 0.06491675809957087, 0.06506695481948555, 0.0650766531471163, 0.06514228298328817, 0.06507326103746891, 0.0650086528621614, 0.0649007901083678, 0.06511658895760775, 0.06492875097319484, 0.06503673107363284, 0.06482654088176787, 0.0652068939525634, 0.06486430810764432, 0.06486658495850861, 0.0655553329270333, 0.06493769120424986, 0.06509223789907992, 0.06495324010029435, 0.06495995679870248, 0.06508598010987043, 0.0648894771002233, 0.06496356287971139, 0.06528347893618047, 0.06507631298154593, 0.06510032806545496, 0.06502252700738609, 0.06502815010026097, 0.06562779587693512, 0.06440811208449304, 0.06637948495335877, 0.0643978149164468, 0.06473829597234726, 0.06419258122332394, 0.06467162584885955, 0.06454653898254037, 0.06390241114422679, 0.06433736998587847, 0.06420045485720038, 0.06441846303641796, 0.06458915607072413, 0.064332515001297, 0.06465771398507059, 0.06460272800177336, 0.06455723708495498, 0.06446496490389109, 0.06427162699401379, 0.06461245706304908, 0.06416836101561785, 0.064073784975335, 0.0644789319485426, 0.06507978797890246, 0.06390438391827047, 0.06423701299354434, 0.06438445812091231, 0.06477320892736316, 0.06447559501975775, 0.06422166991978884, 0.06422990816645324, 0.06448044301941991, 0.06448140786960721, 0.06466248095966876, 0.0644094820600003, 0.06431635795161128, 0.0649904899764806, 0.06416854402050376, 0.06419409113004804, 0.06456523295491934, 0.06428698706440628, 0.06453953101299703, 0.06435083388350904, 0.06438048114068806, 0.0645602848380804, 0.06440594606101513, 0.0643610660918057, 0.0646076810080558, 0.06438788399100304, 0.0643921468872577, 0.06400465709157288, 0.06438643299043179, 0.06437829695641994, 0.06448994693346322, 0.06457351613789797, 0.064232945907861, 0.06470313994213939, 0.0642602511215955, 0.06423226301558316, 0.0645180658902973, 0.06453620712272823, 0.06410263385623693, 0.06421943102031946, 0.06448477413505316, 0.06460568797774613, 0.064394760876894, 0.06410564598627388, 0.0650409059599042, 0.06413988908752799, 0.06440933607518673, 0.06423705792985857, 0.06445764098316431, 0.06480140006169677, 0.06420324905775487, 0.06439613597467542, 0.06460265093483031, 0.06474747392348945, 0.0642591470386833, 0.06458793999627233, 0.06469469098374248, 0.06425007316283882, 0.06440235394984484, 0.06447700504213572, 0.0646591039840132, 0.06444871285930276, 0.064236642094329, 0.06432704394683242, 0.0646517479326576, 0.06491773808375001, 0.06392702111043036, 0.06398708396591246, 0.06424379791133106, 0.06468067597597837, 0.06431792094372213, 0.06483050202950835, 0.0653414879925549, 0.06370268110185862, 0.0644371029920876, 0.06447203108109534, 0.06438343692570925, 0.06416937196627259, 0.06420009792782366, 0.06463054497726262, 0.0640335320495069, 0.06467964313924313, 0.0642532620113343, 0.06480593280866742, 0.06461983011104167, 0.06438470492139459, 0.06411243695765734, 0.0644850970711559, 0.0643224329687655, 0.06403080513700843, 0.06467471783980727, 0.06531255599111319, 0.06431374419480562, 0.0646650658454746, 0.06517361802980304, 0.06370777101255953, 0.06434400309808552, 0.06493106693960726, 0.06550944689661264, 0.06833929009735584, 0.0597563530318439, 0.06887548300437629, 0.06534118694253266, 0.0649713019374758, 0.06492911302484572, 0.06507178908213973, 0.06468021892942488, 0.06489706295542419, 0.0650625501293689, 0.06510032690130174, 0.06561569892801344, 0.06422572419978678, 0.06476279790513217, 0.0655047451145947, 0.06475569284521043, 0.06487849494442344, 0.0647328810300678, 0.06513977004215121, 0.06512409797869623, 0.06534373806789517, 0.06501604686491191, 0.06548633519560099, 0.06451034196652472, 0.06520621501840651, 0.06481937994249165, 0.06491786101832986, 0.06493616802617908, 0.06474943901412189, 0.0650547849945724, 0.06510009802877903, 0.06451696786098182, 0.06484876410104334, 0.06496945093385875, 0.06520775705575943, 0.0650169369764626, 0.06493450794368982, 0.06497156014665961, 0.06488080183044076, 0.06501131202094257, 0.06503484910354018, 0.06463882583193481, 0.06511814403347671, 0.06480310717597604, 0.06503279181197286, 0.06461984408088028, 0.06565892905928195, 0.06573497992940247, 0.06477011810056865, 0.06480849487707019, 0.06543889502063394, 0.06468812492676079, 0.0651138520333916, 0.06493668095208704, 0.06523943319916725, 0.06449087779037654, 0.06490123202092946, 0.06463642115704715, 0.06501008686609566, 0.06511985906399786, 0.06520407507196069, 0.06489135487936437, 0.0650024909991771, 0.06463376013562083, 0.0648545608855784, 0.0651097591035068, 0.06505229393951595, 0.06430345308035612, 0.06520035397261381, 0.065025300020352, 0.06504787597805262, 0.0648331108968705, 0.06513674696907401, 0.06522949016653001, 0.06496741180308163, 0.06520089809782803, 0.06480292300693691, 0.06461892696097493, 0.06481329700909555, 0.0647634461056441, 0.06500070192851126, 0.06452925899066031, 0.06512602209113538, 0.06492068502120674, 0.06491545680910349, 0.06527022807858884, 0.06583587592467666, 0.06517981900833547, 0.06501066498458385, 0.06528049614280462, 0.06505853892304003, 0.06484484695829451, 0.06487769703380764, 0.06481142691336572, 0.06546005117706954, 0.06519026192836463, 0.0652578929439187, 0.06484242714941502, 0.06500656181015074, 0.06459578406065702, 0.06515718903392553, 0.06495654792524874, 0.06504498515278101, 0.06490718992426991, 0.06493360595777631, 0.064888701075688, 0.06504911207593977, 0.06483228784054518, 0.06595942913554609, 0.07524231588467956, 0.0683811551425606, 0.06529038585722446, 0.06469499692320824, 0.06496564205735922, 0.06553170015104115, 0.06359401880763471, 0.06537350616417825, 0.06510614603757858, 0.0642024539411068, 0.06502427393570542, 0.06540587102063, 0.06529413000680506, 0.06521683093160391, 0.06506153708323836, 0.0651649299543351, 0.06512986007146537, 0.06522578187286854, 0.06507957098074257, 0.06593659915961325, 0.06428645085543394, 0.06536086811684072, 0.0649834128562361, 0.06508175516501069, 0.06555021298117936, 0.0652961079031229, 0.06538671301677823, 0.06507517513819039, 0.06481408583931625, 0.06523127411492169, 0.06624681293033063, 0.06798716494813561, 0.07571293110959232, 0.0739991229493171], [0.06508314097300172, 0.0655044699087739, 0.06523598707281053, 0.0660279500298202, 0.06491935299709439, 0.06550712999887764, 0.0656158800702542, 0.06523017189465463, 0.06541207595728338, 0.06563860713504255, 0.0649560249876231, 0.06539557687938213, 0.06495279516093433, 0.0651593660004437, 0.06477474980056286, 0.0649326671846211, 0.06510834000073373, 0.0648247329518199, 0.06519106798805296, 0.06554546789266169, 0.06491947919130325, 0.06546750199049711, 0.06486039492301643, 0.06540939887054265, 0.06470966711640358, 0.06468689907342196, 0.06502926093526185, 0.0652531508821994, 0.0646786920260638, 0.06507650995627046, 0.06480594514869153, 0.06509006884880364, 0.0649561770260334, 0.0651900521479547, 0.0648181929718703, 0.06508407206274569, 0.0649862817954272, 0.06514024501666427, 0.06510291108861566, 0.06521030096337199, 0.06482229614630342, 0.0651276228018105, 0.06505197705700994, 0.06517971912398934, 0.06482070195488632, 0.06492088595405221, 0.06611209595575929, 0.06500249705277383, 0.06486815097741783, 0.06612139102071524, 0.06524853594601154, 0.06546068098396063, 0.06541712116450071, 0.06550554791465402, 0.06508070207200944, 0.06524462392553687, 0.06598399789072573, 0.0654544320423156, 0.06467706616967916, 0.06512996088713408, 0.0648595120292157, 0.06525719887576997, 0.06513612205162644, 0.06585850403644145, 0.06427517998963594, 0.0650165299884975, 0.06478450004942715, 0.06500500999391079, 0.0647515959572047, 0.06523314001969993, 0.06476037902757525, 0.06545233889482915, 0.06516501819714904, 0.06541538797318935, 0.06472081691026688, 0.06562914489768445, 0.06511533609591424, 0.06512093194760382, 0.065016824984923, 0.06557147204875946, 0.06438394193537533, 0.06443856819532812, 0.06428834982216358, 0.06445287517271936, 0.0641321639996022, 0.06503386585973203, 0.06405666307546198, 0.06433246494270861, 0.0641155019402504, 0.0643611392006278, 0.06437102798372507, 0.06441491493023932, 0.06459212186746299, 0.06422710209153593, 0.06430014106445014, 0.06430583097971976, 0.06404721597209573, 0.0644850549288094, 0.06422991515137255, 0.06439150893129408, 0.0651785689406097, 0.0691770960111171, 0.0591002800501883, 0.06869743694551289, 0.06698584696277976, 0.06487936316989362, 0.06499281991273165, 0.0647961141075939, 0.06491674087010324, 0.0650977420154959, 0.06480842898599803, 0.0649030499625951, 0.0650806580670178, 0.06487360293976963, 0.06510011409409344, 0.06500951084308326, 0.065497653093189, 0.06502383691258729, 0.06485915905795991, 0.06513002794235945, 0.06465881620533764, 0.06486172880977392, 0.06503550405614078, 0.06517089111730456, 0.06510848482139409, 0.06465329020284116, 0.0648624999448657, 0.06509513803757727, 0.06480446085333824, 0.0654387881513685, 0.06518842186778784, 0.06496525695547462, 0.06496891309507191, 0.0648971030022949, 0.06522279791533947, 0.06459167716093361, 0.06509215082041919, 0.06468643597327173, 0.06489630718715489, 0.06510904198512435, 0.06474770698696375, 0.06466849590651691, 0.0651909289881587, 0.06497542816214263, 0.06499376194551587, 0.06476016086526215, 0.06514015817083418, 0.06489910581149161, 0.06498001120053232, 0.06528182188048959, 0.06507544405758381, 0.06511805905029178, 0.0651132499333471, 0.06501092296093702, 0.06486499891616404, 0.06490485207177699, 0.06508780503645539, 0.06493600108660758, 0.06521907285787165, 0.06494513712823391, 0.0650278169196099, 0.06513726501725614, 0.06494507798925042, 0.06496761692687869, 0.06489397003315389, 0.06510741612873971, 0.06446944293566048, 0.06500549707561731, 0.06508037797175348, 0.06498720194213092, 0.0653228999581188, 0.06490443809889257, 0.06501249992288649, 0.06481210794299841, 0.06518505909480155, 0.06614390504546463, 0.06474988698028028, 0.06521267583593726, 0.06490897503681481, 0.06512552499771118, 0.06501517817378044, 0.06472625397145748, 0.06580434390343726, 0.06504644500091672, 0.06517442804761231, 0.06468805205076933, 0.0651296079158783, 0.0652704848907888, 0.06488314503803849, 0.06507249712012708, 0.06506492896005511, 0.0651564309373498, 0.06505007902160287, 0.06502464297227561, 0.06489589414559305, 0.06510591087862849, 0.06494290102273226, 0.06505653611384332, 0.06480003288015723, 0.0652098769787699, 0.06484480504877865, 0.064877572003752, 0.06555341696366668, 0.06493504904210567, 0.06506809499114752, 0.064985804958269, 0.06495785294100642, 0.06507559213787317, 0.06487117405049503, 0.06496923579834402, 0.06529427412897348, 0.06506496691145003, 0.0651022840756923, 0.06502626999281347, 0.06501024193130434, 0.0655540470033884, 0.06465337006375194, 0.06580587895587087, 0.06497790501452982, 0.06456522410735488, 0.06407139101065695, 0.06444322597235441, 0.06467868783511221, 0.06401996617205441, 0.06437745690345764, 0.06421416602097452, 0.06442259112372994, 0.06459183199331164, 0.06434863083995879, 0.06460847496055067, 0.06461545918136835, 0.0644852570258081, 0.06454531685449183, 0.06424263096414506, 0.0646262071095407, 0.06416080496273935, 0.06409650691784918, 0.06445907708257437, 0.06449421588331461, 0.06444676499813795, 0.06424865708686411, 0.06441389909014106, 0.06476985896006227, 0.06447375495918095, 0.06422720197588205, 0.0642530859913677, 0.0644771889783442, 0.06442451803013682, 0.06466588797047734, 0.06447280407883227, 0.06431251903995872, 0.06498126196675003, 0.06418091990053654, 0.06413447018712759, 0.06461135181598365, 0.06425965600647032, 0.06454474502243102, 0.06432930799201131, 0.06438546697609127, 0.06458242610096931, 0.06440992909483612, 0.06435827282257378, 0.06461200513876975, 0.06435532099567354, 0.06437914888374507, 0.06401151604950428, 0.06441111909225583, 0.06435781181789935, 0.06480472208932042, 0.06427768804132938, 0.0642470398452133, 0.06469167303293943, 0.06424909410998225, 0.06424182886257768, 0.06452242517843843, 0.06452101492322981, 0.06409644894301891, 0.06421205098740757, 0.06448696809820831, 0.06460815109312534, 0.06437478098087013, 0.0648169768974185, 0.06433670106343925, 0.06415149383246899, 0.06441187718883157, 0.06423704302869737, 0.06443529296666384, 0.0648256738204509, 0.06418035202659667, 0.06436390802264214, 0.06463203299790621, 0.06475816201418638, 0.06410226598381996, 0.06473984499461949, 0.06467762403190136, 0.06435772497206926, 0.06429879413917661, 0.06448790500871837, 0.06462795683182776, 0.06462355004623532, 0.06405721697956324, 0.06434454000554979, 0.06461857911199331, 0.06495700590312481, 0.06386859295889735, 0.06399582605808973, 0.06426876690238714, 0.0646405320148915, 0.06435990310274065, 0.06481574103236198, 0.06528578395955265, 0.0637064459733665, 0.06447471491992474, 0.06447967211715877, 0.06440078211016953, 0.06417267397046089, 0.06418889900669456, 0.06472570798359811, 0.06392266298644245, 0.06469611986540258, 0.06421527406200767, 0.06481669400818646, 0.06473050802014768, 0.06425129296258092, 0.06413102801889181, 0.06451106793247163, 0.06441284297034144, 0.06405357317999005, 0.06457546190358698, 0.06530564208514988, 0.06432045088149607, 0.06460726796649396, 0.06501498399302363, 0.06387756997719407, 0.06436356506310403, 0.06494949315674603, 0.06539773079566658, 0.06855778815224767, 0.0598150659352541, 0.0688886079005897, 0.06534840608946979, 0.06497488101013005, 0.0649221430066973, 0.06505701411515474, 0.06468595191836357, 0.0649126349017024, 0.06503720302134752, 0.0651059141382575, 0.06541467597708106, 0.06452376302331686, 0.06465402897447348, 0.0654888569843024, 0.06475026393309236, 0.06490295100957155, 0.0647455679718405, 0.0651282740291208, 0.06513796304352582, 0.06533971289172769, 0.0649921060539782, 0.06544380192644894, 0.06474562315270305, 0.06501989183016121, 0.06479880306869745, 0.06492504011839628, 0.06494408380240202, 0.06477444316260517, 0.06503319996409118, 0.06509049492888153, 0.0645386700052768, 0.06481517711654305, 0.06498758192174137, 0.0651927599683404, 0.06501751602627337, 0.0649684639647603, 0.06495573394931853, 0.06487799808382988, 0.06502429209649563, 0.06500296504236758, 0.06465213978663087, 0.06512410216964781, 0.06481933896429837, 0.06502377591095865, 0.06505482201464474, 0.06540842610411346, 0.06553546199575067, 0.06477952795103192, 0.06479654205031693, 0.06544887088239193, 0.06469711614772677, 0.06512135895900428, 0.06492983503267169, 0.06523407995700836, 0.0644795629195869, 0.06483865296468139, 0.06470990204252303, 0.06499282410368323, 0.06513549387454987, 0.06555645796470344, 0.06454341602511704, 0.06498835212551057, 0.06463914387859404, 0.06485351314768195, 0.06510003097355366, 0.06504511390812695, 0.06429789797402918, 0.0652325670234859, 0.06500994600355625, 0.06518867192789912, 0.06467352714389563, 0.0651524649001658, 0.06522480305284262, 0.06496709701605141, 0.06519356206990778, 0.06476813089102507, 0.06465426413342357, 0.06479295692406595, 0.06475585093721747, 0.06502436101436615, 0.06471809791401029, 0.06490723812021315, 0.06492226896807551, 0.06489971512928605, 0.06529537891037762, 0.0658560290466994, 0.06515988893806934, 0.06500747101381421, 0.06527462298981845, 0.06504955189302564, 0.06486000213772058, 0.06489815702661872, 0.06477949186228216, 0.06545699108392, 0.06517224293202162, 0.06527121411636472, 0.06485661002807319, 0.06499653682112694, 0.06461914698593318, 0.0651171961799264, 0.06496774987317622, 0.06504411902278662, 0.06493143201805651, 0.06489300495013595, 0.06490863603539765, 0.06502005597576499, 0.06486893096007407, 0.06591230514459312, 0.07540885591879487, 0.0684180511161685, 0.06530973198823631, 0.06473003583960235, 0.06492536002770066, 0.06533325207419693, 0.06387128005735576, 0.06531353993341327, 0.06512787705287337, 0.06417333404533565, 0.0649138989392668, 0.06539606093429029, 0.06540770991705358, 0.06513944105245173, 0.0650661401450634, 0.06513696094043553, 0.06516223098151386, 0.06520276702940464, 0.06507425592280924, 0.06487976992502809, 0.06535634212195873, 0.06538650998845696, 0.06493882788345218, 0.06510033505037427, 0.0656606589909643, 0.06511339009739459, 0.06529967905953526, 0.06521268980577588, 0.064817309146747, 0.06517139589414, 0.0662379190325737, 0.06816494814120233, 0.07565740682184696, 0.07417138409800828, 0.07279302505776286, 0.05854365788400173, 0.051260021049529314, 0.04895416600629687, 0.04795118188485503, 0.04736800002865493, 0.0474304910749197, 0.04725432093255222, 0.047451884020119905, 0.047614532988518476, 0.047736281994730234, 0.047567879082635045, 0.047298404853791, 0.04745392804034054, 0.04733289312571287, 0.04741866001859307, 0.047439804999157786, 0.04723313800059259, 0.0471438798122108, 0.04770379001274705, 0.04741949820891023, 0.04730348498560488, 0.047437229892238975, 0.04751340695656836, 0.047071645967662334, 0.047435353975743055, 0.04726907121948898, 0.047620818950235844, 0.04704652284272015, 0.04854959808290005], [0.06430678209289908, 0.06442008703015745, 0.06446611392311752, 0.06398252910003066, 0.06507909600622952, 0.06415509898215532, 0.06432907795533538, 0.06412115809507668, 0.06433981098234653, 0.06438804394565523, 0.06440938403829932, 0.0644631078466773, 0.06436266098171473, 0.06431051902472973, 0.06433786801062524, 0.0638686001766473, 0.06448520184494555, 0.06422923714853823, 0.06439233687706292, 0.06517441011965275, 0.06920311879366636, 0.059084674110636115, 0.06869754497893155, 0.06698179407976568, 0.0648786339443177, 0.0648355430457741, 0.06495941104367375, 0.0650566618423909, 0.06508747208863497, 0.0648232710082084, 0.0649033400695771, 0.06507439981214702, 0.0647325690370053, 0.06509981700219214, 0.0650081371422857, 0.06550019280984998, 0.06502381106838584, 0.0648561310954392, 0.06512847892008722, 0.06465881294570863, 0.06500895717181265, 0.06504725478589535, 0.06501350412145257, 0.06511581898666918, 0.06464460096322, 0.06486785202287138, 0.06509293406270444, 0.06480546086095273, 0.06544024706818163, 0.06518797995522618, 0.06496791495010257, 0.06496818806044757, 0.06489566015079618, 0.06522211479023099, 0.06459430302493274, 0.06509256409481168, 0.0646836010273546, 0.06502863788045943, 0.06498170807026327, 0.06474615400657058, 0.06466599903069437, 0.06519162491895258, 0.06497995811514556, 0.06499544298276305, 0.06476052384823561, 0.06513844802975655, 0.06491479114629328, 0.06496559898369014, 0.06528223096393049, 0.06507722288370132, 0.06511365016922355, 0.06511284690350294, 0.06501150503754616, 0.06487994291819632, 0.064901685109362, 0.06508111488074064, 0.06493250303901732, 0.06522596697323024, 0.06494692713022232, 0.06502540200017393, 0.06513566290959716, 0.06508170394226909, 0.06482941214926541, 0.06489490089006722, 0.0651043220423162, 0.06447317800484598, 0.06501202494837344, 0.06507789809256792, 0.06498698587529361, 0.06531985616311431, 0.06490851286798716, 0.06501262309029698, 0.06480996194295585, 0.06518418295308948, 0.06614430015906692, 0.06488120998255908, 0.06507759890519083, 0.06491095805540681, 0.06512738089077175, 0.06501537817530334, 0.06473104190081358, 0.06580171710811555, 0.0651820048224181, 0.06503541907295585, 0.0646997441072017, 0.0651201808359474, 0.06527100293897092, 0.06488236505538225, 0.06506954412907362, 0.06521081691607833, 0.06501519191078842, 0.06504601216875017, 0.06502614892087877, 0.06490214308723807, 0.06510579097084701, 0.0649374290369451, 0.0651654249522835, 0.06469617900438607, 0.06521029002033174, 0.06484408094547689, 0.06487881601788104, 0.0655490409117192, 0.06493730610236526, 0.06506743584759533, 0.06498739519156516, 0.06496174703352153, 0.06507260189391673, 0.0648766818922013, 0.06496579619124532, 0.06529186991974711, 0.06506968499161303, 0.06509847310371697, 0.06502852379344404, 0.06501234509050846, 0.06530351797118783, 0.06488273106515408, 0.06593804899603128, 0.06449469295330346, 0.06470704707317054, 0.06432170583866537, 0.06471319706179202, 0.06442394596524537, 0.06399157107807696, 0.0645414050668478, 0.06405100086703897, 0.06441733497194946, 0.06459528813138604, 0.06434989301487803, 0.06460609706118703, 0.06461928295902908, 0.06447999691590667, 0.06455354508943856, 0.06423848099075258, 0.06462807091884315, 0.06415866897441447, 0.06409297208301723, 0.0644594409968704, 0.06449746107682586, 0.06445192778483033, 0.06426210002973676, 0.06440020306035876, 0.06477258307859302, 0.0644666119478643, 0.0645183059386909, 0.06410922110080719, 0.06447451189160347, 0.06440323800779879, 0.0647057241294533, 0.06445453991182148, 0.06431498099118471, 0.06483164196833968, 0.06432066601701081, 0.06414692406542599, 0.06461237603798509, 0.06426790682598948, 0.06454034009948373, 0.06417847704142332, 0.06455668387934566, 0.06457065418362617, 0.06439658580347896, 0.0643639462068677, 0.06460845097899437, 0.0643597689922899, 0.06438998482190073, 0.06401594704948366, 0.06439069495536387, 0.06421086797490716, 0.06495088501833379, 0.06428421405144036, 0.06409611809067428, 0.06484055891633034, 0.06424924405291677, 0.06424273387528956, 0.06451648706570268, 0.06437570904381573, 0.064245929941535, 0.06421551597304642, 0.06447880901396275, 0.06446600100025535, 0.06454248307272792, 0.06478282297030091, 0.0642046129796654, 0.06415072595700622, 0.06456245016306639, 0.06423560786060989, 0.06443974515423179, 0.0648375409655273, 0.06402031797915697, 0.06454920908436179, 0.06460148887708783, 0.06460039899684489, 0.0642328280955553, 0.0647616300266236, 0.06467987690120935, 0.06438542087562382, 0.06427549105137587, 0.06447860016487539, 0.06448742095381021, 0.0646175688598305, 0.06406160211190581, 0.06434102705679834, 0.06463039899244905, 0.06498216697946191, 0.06383325601927936, 0.0639951208140701, 0.0644431000109762, 0.06448969920165837, 0.06433677300810814, 0.06481769378297031, 0.06537352199666202, 0.06362487515434623, 0.06447734101675451, 0.06446888693608344, 0.06440311903133988, 0.06417229794897139, 0.06418975512497127, 0.0647583908867091, 0.0638986430130899, 0.06469289003871381, 0.06422228086739779, 0.06481095612980425, 0.06475183903239667, 0.06423700088635087, 0.06413820199668407, 0.06448959303088486, 0.06444555195048451, 0.06403882405720651, 0.06455960706807673, 0.06530207302421331, 0.06432310701347888, 0.0646114437840879, 0.06509101414121687, 0.06380399991758168, 0.06436389801092446, 0.0651036121416837, 0.06541045196354389, 0.06857315101660788, 0.059943096013739705, 0.06871416186913848, 0.06521205906756222, 0.06497080391272902, 0.06491932808421552, 0.0652224849909544, 0.06467315903864801, 0.06476862798444927, 0.06518149701878428, 0.06510507897473872, 0.06535506900399923, 0.0645368779078126, 0.06456566601991653, 0.06553336512297392, 0.06470498582348228, 0.0648976732045412, 0.06474490999244153, 0.06513661192730069, 0.06512585002928972, 0.06547171482816339, 0.06500878813676536, 0.06567850988358259, 0.06451868498697877, 0.06487207300961018, 0.06480198306962848, 0.06506644794717431, 0.06480153207667172, 0.0647602491080761, 0.06504140398465097, 0.06509326188825071, 0.06453975196927786, 0.06481636012904346, 0.06499151699244976, 0.06520037003792822, 0.06500734179280698, 0.0649637000169605, 0.06495282496325672, 0.06487919017672539, 0.06502435589209199, 0.06500119809061289, 0.06466532684862614, 0.06511194701306522, 0.06481580412946641, 0.06502604996785522, 0.06497399508953094, 0.06547087384387851, 0.06556460703723133, 0.06477638310752809, 0.0649405438452959, 0.06530420109629631, 0.06469724886119366, 0.06511954311281443, 0.06493363692425191, 0.06523496098816395, 0.06461222609505057, 0.06470359093509614, 0.06471283407881856, 0.06498724594712257, 0.06513768713921309, 0.06510040280409157, 0.06499703810550272, 0.06499473587609828, 0.06464157020673156, 0.06484318501316011, 0.06511421478353441, 0.06503537017852068, 0.06430221092887223, 0.06522531295195222, 0.06501526897773147, 0.06533377896994352, 0.06452992605045438, 0.06515170098282397, 0.06522704311646521, 0.06496334006078541, 0.06519937794655561, 0.0649110700469464, 0.0645078329835087, 0.06494411989115179, 0.06460812711156905, 0.06502224295400083, 0.06479073595255613, 0.06497687590308487, 0.06492620520293713, 0.0649226438254118, 0.06526873400434852, 0.06571355718187988, 0.06530095590278506, 0.06516199791803956, 0.06512120412662625, 0.06504247384145856, 0.06486826320178807, 0.06475534988567233, 0.06493457709439099, 0.06544287200085819, 0.06518553895875812, 0.0652703728992492, 0.0647075129672885, 0.06499757803976536, 0.06461904803290963, 0.06528071803040802, 0.0649552559480071, 0.06489499914459884, 0.06492987577803433, 0.06505719712004066, 0.06474794703535736, 0.06517463992349803, 0.06471191113814712, 0.06604182696901262, 0.07530730403959751, 0.0685262018814683, 0.06517916009761393, 0.06475898087956011, 0.06500872713513672, 0.06532677891664207, 0.06383894407190382, 0.06528469687327743, 0.06511485809460282, 0.06416023802012205, 0.06504900380969048, 0.06539830705150962, 0.06527440203353763, 0.06513244193047285, 0.06505432608537376, 0.06526860408484936, 0.06502657290548086, 0.06520371395163238, 0.06507813395000994, 0.06584654212929308, 0.06438932893797755, 0.0653905279468745, 0.06493782112374902, 0.06509469705633819, 0.06566963600926101, 0.06528438488021493, 0.0655214770231396, 0.06482598790898919, 0.06494255503639579, 0.06540915695950389, 0.06602248619310558, 0.06817958084866405, 0.07563400617800653, 0.07414070493541658, 0.07277109893038869, 0.05853589391335845, 0.050823165103793144, 0.049298430094495416, 0.04802300198934972, 0.047363329911604524, 0.047360219061374664, 0.04732573195360601, 0.04744748794473708, 0.04761825711466372, 0.04773082188330591, 0.047509740106761456, 0.04736041487194598, 0.04745911015197635, 0.04732966190204024, 0.04734961409121752, 0.04743580287322402, 0.04723828798159957, 0.04720733012072742, 0.047637252835556865, 0.047421416034922004, 0.04730785801075399, 0.047501672990620136, 0.047445269068703055, 0.04709322890266776, 0.04748597205616534, 0.04727086494676769, 0.04754601721651852, 0.04704893380403519, 0.048652084078639746, 0.06123178103007376, 0.03894505091011524, 0.05459800804965198, 0.04972547106444836, 0.04791301884688437, 0.047407782170921564, 0.04750797082670033, 0.04769040900282562, 0.04764665802940726, 0.04761194810271263, 0.04773945198394358, 0.04744482599198818, 0.0474572388920933, 0.047472812002524734, 0.04746584012173116, 0.04760868498124182, 0.047433651983737946, 0.04734716610983014, 0.04743994679301977, 0.04767083213664591, 0.047321533085778356, 0.04740505781956017, 0.04770589410327375, 0.047454911982640624, 0.0472306739538908, 0.04759021708741784, 0.047462299931794405, 0.0476671380456537, 0.04763725702650845, 0.0475320010446012, 0.047512260964140296, 0.04746875283308327, 0.04750610399059951, 0.04726022621616721, 0.04771569580771029, 0.04758728318847716, 0.047559901839122176, 0.04738103412091732, 0.047534288838505745, 0.04753516614437103, 0.04759814101271331, 0.047417014837265015, 0.04774834308773279, 0.047316594049334526, 0.04746715701185167, 0.04780011693947017, 0.04776949994266033, 0.04771445016376674, 0.047309634974226356, 0.04754769499413669, 0.04750318103469908, 0.047581602819263935, 0.04741819319315255, 0.04757770197466016, 0.047402333933860064, 0.04743949603289366, 0.04746210505254567, 0.047458851942792535, 0.04743856005370617, 0.047506198985502124, 0.047479950822889805, 0.047977420035749674, 0.04773055715486407, 0.047451035818085074, 0.0473480150103569, 0.04723184811882675, 0.04766806401312351, 0.0474433999042958, 0.04737401404418051, 0.04747342108748853, 0.04772117594256997, 0.047717087902128696, 0.047414631117135286, 0.04780374001711607, 0.047649043845012784, 0.047706827986985445, 0.04770423611626029, 0.047483569011092186, 0.04747587093152106, 0.04750414611771703, 0.04849923588335514], [0.06419666297733784, 0.0649943589232862, 0.06403428711928427, 0.06409249105490744, 0.06437565688975155, 0.06415924103930593, 0.06441554590128362, 0.06459036818705499, 0.0643335678614676, 0.0646583701018244, 0.06460225395858288, 0.06433018809184432, 0.0646927289199084, 0.06427131406962872, 0.06461258488707244, 0.06417030491866171, 0.06407260498963296, 0.06448158412240446, 0.06525550899095833, 0.0637305579148233, 0.06423440598882735, 0.06438170513138175, 0.06477551790885627, 0.06447882112115622, 0.06426828587427735, 0.064227543072775, 0.06448554806411266, 0.06444176891818643, 0.06470504705794156, 0.06441283784806728, 0.06431726599112153, 0.06493487511761487, 0.06421914789825678, 0.06419413001276553, 0.0645602191798389, 0.0642938599921763, 0.06455496791750193, 0.06428494001738727, 0.0644303800072521, 0.06456091487780213, 0.06440510298125446, 0.06435929122380912, 0.06461112084798515, 0.06438498408533633, 0.06439721095375717, 0.06400129292160273, 0.06438399408943951, 0.06433128705248237, 0.06482689687982202, 0.06428689905442297, 0.06418620003387332, 0.0647528578992933, 0.06425360310822725, 0.06423472100868821, 0.0645175538957119, 0.0644912151619792, 0.06415432877838612, 0.06421835510991514, 0.06448522000573575, 0.06455726292915642, 0.06444376218132675, 0.06415694486349821, 0.06493678293190897, 0.064142607152462, 0.06445644004270434, 0.06424090592190623, 0.06464176089502871, 0.06457276199944317, 0.0642002560198307, 0.0643953331746161, 0.06465804181061685, 0.06469224509783089, 0.06403941288590431, 0.064859809121117, 0.06469110306352377, 0.06438609887845814, 0.06427106400951743, 0.064479238120839, 0.06460575293749571, 0.06459712889045477, 0.06408739415928721, 0.0643262870144099, 0.06465842598117888, 0.0649359948001802, 0.06390699208714068, 0.06399224209599197, 0.06428689695894718, 0.06463227700442076, 0.06432024505920708, 0.06482697394676507, 0.06538206292316318, 0.06366707710549235, 0.0644372368697077, 0.06446904805488884, 0.06438638409599662, 0.06416906300000846, 0.06419523898512125, 0.06464560283347964, 0.06403452204540372, 0.06467296206392348, 0.06424905802123249, 0.06480929185636342, 0.06462477706372738, 0.06437894399277866, 0.06410893006250262, 0.06448873109184206, 0.06432775291614234, 0.06416916102170944, 0.06452914094552398, 0.06531762611120939, 0.06432593381032348, 0.06465011509135365, 0.06518662511371076, 0.06369823589920998, 0.06434268387965858, 0.06498423800803721, 0.06551845814101398, 0.0683361638803035, 0.05975559703074396, 0.0688702929764986, 0.0652919439598918, 0.06496817711740732, 0.06492913910187781, 0.06507436884567142, 0.06467875698581338, 0.06489710696041584, 0.06505852611735463, 0.06510690087452531, 0.06566719012334943, 0.06419798196293414, 0.06611482496373355, 0.06414140714332461, 0.06476249499246478, 0.06486252183094621, 0.06473257602192461, 0.06514171697199345, 0.06512194406241179, 0.06539586698636413, 0.06496067298576236, 0.06522334110923111, 0.0647542579099536, 0.065231969114393, 0.06481864582747221, 0.06492196512408555, 0.06493140407837927, 0.06474902597256005, 0.06505890097469091, 0.06509680184535682, 0.06452432717196643, 0.06484421202912927, 0.06497283885255456, 0.06520631699822843, 0.06501803616993129, 0.06493268883787096, 0.06496944301761687, 0.06487881601788104, 0.06501001003198326, 0.06503686401993036, 0.06463854806497693, 0.06512237200513482, 0.06480013695545495, 0.06503548589535058, 0.06456048204563558, 0.0656910059042275, 0.06576177407987416, 0.06476970296353102, 0.06480918894521892, 0.06544210808351636, 0.06468787207268178, 0.06510933605022728, 0.06495019793510437, 0.06522734486497939, 0.06449133204296231, 0.06463489099405706, 0.064903125166893, 0.0650130279827863, 0.06511755497194827, 0.0652147310320288, 0.06488274387083948, 0.06500052311457694, 0.06463484885171056, 0.06485876813530922, 0.06510595488362014, 0.06505838595330715, 0.06430165399797261, 0.0651975111104548, 0.06502714194357395, 0.06519845104776323, 0.06467968807555735, 0.06513920798897743, 0.06523230788297951, 0.06496787304058671, 0.06520089809782803, 0.06480554095469415, 0.0646176019217819, 0.06480988301336765, 0.06476638396270573, 0.06499940506182611, 0.06453220103867352, 0.06512509589083493, 0.06492135114967823, 0.06491415482014418, 0.06527149514295161, 0.06583721493370831, 0.06517511396668851, 0.06501389504410326, 0.0652809978928417, 0.06510232505388558, 0.06480035604909062, 0.06488055712543428, 0.06480970093980432, 0.06550167594105005, 0.06515041110105813, 0.0652623069472611, 0.06483581406064332, 0.06501235184259713, 0.06459228112362325, 0.06515783001668751, 0.06495829788036644, 0.06504319491796196, 0.06490840716287494, 0.06493187602609396, 0.06489369296468794, 0.06505992286838591, 0.06482050404883921, 0.06601796601898968, 0.07519916002638638, 0.06837332807481289, 0.06528986082412302, 0.0647021580953151, 0.06501232297159731, 0.06554341013543308, 0.063548473874107, 0.06539396196603775, 0.0650759341660887, 0.06420161086134613, 0.06501193693839014, 0.065421047154814, 0.06528087891638279, 0.06522253691218793, 0.06505464599467814, 0.06521908100694418, 0.06507796607911587, 0.06522571598179638, 0.06508112396113575, 0.0659407910425216, 0.06427807593718171, 0.06536757317371666, 0.06497780699282885, 0.06508602784015238, 0.0656033311970532, 0.06524645700119436, 0.06506548286415637, 0.06540749315172434, 0.06485673878341913, 0.0651723020710051, 0.06624227692373097, 0.0680050051305443, 0.07536591589450836, 0.0743741241749376, 0.07252954295836389, 0.05853104800917208, 0.05099382693879306, 0.04917498491704464, 0.04797975602559745, 0.04734760615974665, 0.04739873297512531, 0.04728295002132654, 0.04745274386368692, 0.04761618212796748, 0.04772960883565247, 0.047555092023685575, 0.04731602012179792, 0.04746065987274051, 0.04732947307638824, 0.04738764092326164, 0.04743610508739948, 0.04723811196163297, 0.047168344957754016, 0.0476797129958868, 0.04741429607383907, 0.04731141007505357, 0.047467047814279795, 0.04747922206297517, 0.04710526904091239, 0.04743970790877938, 0.0472690740134567, 0.047587277134880424, 0.047051128931343555, 0.04860879294574261, 0.0612797720823437, 0.038950738962739706, 0.05457115592435002, 0.049713140120729804, 0.04791042907163501, 0.04744398989714682, 0.04747136891819537, 0.04768745298497379, 0.04768419708125293, 0.04757444094866514, 0.047738909954205155, 0.04745329008437693, 0.04745001811534166, 0.04747008881531656, 0.0474649490788579, 0.047619126038625836, 0.047434373991563916, 0.04734099702909589, 0.047448026947677135, 0.04767061909660697, 0.047321022022515535, 0.04739697487093508, 0.04771179612725973, 0.04744946793653071, 0.047235280042514205, 0.047590770991519094, 0.04745896300300956, 0.047665552934631705, 0.04763549193739891, 0.047533425968140364, 0.047552322037518024, 0.047440210124477744, 0.04750809399411082, 0.04725197283551097, 0.04772683302871883, 0.04757787100970745, 0.04756401409395039, 0.047380019910633564, 0.04753179498948157, 0.04753222712315619, 0.04759551794268191, 0.04742565704509616, 0.047753200866281986, 0.047307346016168594, 0.04747058101929724, 0.0478027630597353, 0.04776502586901188, 0.04771557403728366, 0.04730954114347696, 0.04721240792423487, 0.04783565108664334, 0.04758710484020412, 0.04742055502720177, 0.04757452802732587, 0.047405658988282084, 0.04743888811208308, 0.04745908081531525, 0.047456600004807115, 0.047439890215173364, 0.047525274800136685, 0.04746345803141594, 0.047977866139262915, 0.04773018602281809, 0.047446563839912415, 0.04736923612654209, 0.04721834696829319, 0.047669943887740374, 0.04743945598602295, 0.04737142403610051, 0.0474774029571563, 0.04771928209811449, 0.047718260902911425, 0.04741368303075433, 0.047804271103814244, 0.047652516048401594, 0.04770596302114427, 0.04770081699825823, 0.04748626099899411, 0.04748851782642305, 0.0474937770050019, 0.04853911395184696, 0.0637890191283077, 0.031827308936044574, 0.04912527906708419, 0.04349402291700244, 0.041542367078363895, 0.041295701870694757, 0.04155105212703347, 0.041608376894146204, 0.04131832509301603, 0.04100077995099127, 0.041582830948755145, 0.0412720141466707, 0.041527895955368876, 0.04141546110622585, 0.0414798348210752, 0.041403949027881026, 0.041462344117462635, 0.042183071840554476, 0.04140451108105481, 0.04216070310212672, 0.04098542779684067, 0.04134786198846996, 0.04199665319174528, 0.041515923803672194, 0.04149529803544283, 0.04133514012210071, 0.041663457872346044, 0.04143218602985144, 0.04140653205104172, 0.04157487605698407, 0.041624982841312885, 0.04120867419987917, 0.04146786383353174, 0.04144235816784203, 0.04159041587263346, 0.041575357085093856, 0.04153256700374186, 0.041440672939643264, 0.041675291024148464, 0.04122069105505943, 0.041532882023602724, 0.041496935999020934, 0.041520790895447135, 0.04136040294542909, 0.041407976066693664, 0.041555577889084816, 0.04138075211085379, 0.04159252904355526, 0.041603000834584236, 0.04133283509872854, 0.041471251053735614, 0.0414175889454782, 0.041695263935253024, 0.04136706213466823, 0.041679757880046964, 0.041317197028547525, 0.04162059212103486, 0.041321043856441975, 0.04153064917773008, 0.042281006928533316, 0.041444194968789816, 0.04165439493954182, 0.04130676295608282, 0.04132942110300064, 0.041296974988654256, 0.04163515800610185, 0.04172061709687114, 0.04129218100570142, 0.041598292998969555, 0.041393836960196495, 0.04154449584893882, 0.04158049006946385, 0.041504597989842296, 0.04163924907334149, 0.04151266789995134, 0.04142374801449478, 0.041521063074469566, 0.047338014002889395, 0.04431221797131002, 0.0472675240598619, 0.04744303994812071, 0.04761310899630189, 0.04755357699468732, 0.047417098889127374, 0.04729501414112747, 0.0476864748634398, 0.0474343600217253, 0.04737565713003278, 0.047652700915932655, 0.047550844959914684, 0.0473229989875108, 0.04757409100420773, 0.04758383915759623, 0.04718988295644522, 0.04770607501268387, 0.04759006900712848, 0.04739480186253786, 0.04705127817578614, 0.047675808891654015, 0.04725074511952698, 0.047591651789844036, 0.04750406299717724, 0.04755005519837141, 0.0475978038739413, 0.04738165414892137, 0.04760108795017004, 0.04761402099393308, 0.047845885856077075, 0.0472207791171968, 0.047278455924242735, 0.04758458794094622, 0.04777122917585075, 0.04686177591793239, 0.047547173919156194, 0.04960609809495509, 0.045835118973627687, 0.04733124002814293, 0.047926645958796144, 0.04756534309126437, 0.047597461845725775, 0.04789313906803727, 0.04774559591896832, 0.047619417076930404, 0.0472464959602803, 0.04758253204636276, 0.0475425710901618, 0.047479790868237615, 0.04806198296137154, 0.04753522016108036, 0.047314656898379326, 0.047600238118320704, 0.04758187383413315, 0.04742065304890275, 0.04732188698835671, 0.04742748406715691, 0.047457202104851604, 0.04746667086146772, 0.0474576260894537, 0.048321884823963046], [0.06504031387157738, 0.06541004311293364, 0.06527986284345388, 0.06521843303926289, 0.06506985612213612, 0.06516898004338145, 0.06512489286251366, 0.06521893804892898, 0.06508365296758711, 0.06593058397993445, 0.06429659714922309, 0.06535880779847503, 0.06498198700137436, 0.06508716219104826, 0.06554445694200695, 0.06529716588556767, 0.0651106049772352, 0.06533139501698315, 0.06483429600484669, 0.06523081916384399, 0.06625102600082755, 0.0679644369520247, 0.0748221140820533, 0.07398680690675974, 0.07296644593589008, 0.05865116696804762, 0.05103927617892623, 0.04914491390809417, 0.04801109991967678, 0.04735636617988348, 0.04738818993791938, 0.04730649385601282, 0.04743149201385677, 0.047617905074730515, 0.04774816706776619, 0.04754038597457111, 0.04732869705185294, 0.047407545847818255, 0.04737322800792754, 0.04736180906184018, 0.04746389901265502, 0.04722722806036472, 0.04718308080919087, 0.04768160521052778, 0.04741512378677726, 0.04730156110599637, 0.04747578199021518, 0.04745529405772686, 0.047118179965764284, 0.047449792036786675, 0.04725314793176949, 0.04759333492256701, 0.047052293084561825, 0.04833398712798953, 0.061699148965999484, 0.03904891503043473, 0.05450253491289914, 0.049781681038439274, 0.04791964893229306, 0.04744573403149843, 0.047463936964049935, 0.047703552059829235, 0.04768038587644696, 0.04757033311761916, 0.047742614056915045, 0.04740386386401951, 0.04750417615287006, 0.047478239983320236, 0.047464550007134676, 0.047586922999471426, 0.04744373494759202, 0.047350045992061496, 0.047419033013284206, 0.047680319054052234, 0.047298672841861844, 0.04743476095609367, 0.04769621114246547, 0.04747474193572998, 0.04720642208121717, 0.047592425951734185, 0.04747683089226484, 0.04767612717114389, 0.047640684992074966, 0.04752003983594477, 0.04751447611488402, 0.04743441799655557, 0.0475089258980006, 0.047278742073103786, 0.04773115389980376, 0.047585208201780915, 0.047519000014290214, 0.04739573481492698, 0.04755541402846575, 0.047544914996251464, 0.04758808412589133, 0.047406222904101014, 0.04772790498100221, 0.04734836798161268, 0.047470889054238796, 0.047788712894544005, 0.04777870001271367, 0.04771137214265764, 0.047303745057433844, 0.047390695894137025, 0.047676237067207694, 0.047570977825671434, 0.04741793405264616, 0.04757945309393108, 0.047388812992721796, 0.04745413502678275, 0.047459343913942575, 0.047461409121751785, 0.047427235869690776, 0.04751012101769447, 0.047489495016634464, 0.04797718790359795, 0.04771731304936111, 0.047466060146689415, 0.04730666498653591, 0.047266037901863456, 0.04765583691187203, 0.047449363162741065, 0.047381442971527576, 0.047464482951909304, 0.047740204026922584, 0.04770686593838036, 0.047411318169906735, 0.047811985947191715, 0.047650859924033284, 0.04768197704106569, 0.04772495897486806, 0.0474822788964957, 0.04746078816242516, 0.04750251304358244, 0.04824456898495555, 0.06423400598578155, 0.031894665909931064, 0.04908642591908574, 0.04340321011841297, 0.04165240004658699, 0.041306106839329004, 0.04154773405753076, 0.04161100694909692, 0.04132444504648447, 0.04099571704864502, 0.041596873896196485, 0.041260714177042246, 0.041544927982613444, 0.04139700182713568, 0.0414971720892936, 0.041393262101337314, 0.041462964843958616, 0.04214022308588028, 0.0413676870521158, 0.04206490097567439, 0.041138297878205776, 0.04137970297597349, 0.04198893904685974, 0.041517750127241015, 0.0415064268745482, 0.041330734035000205, 0.04165500006638467, 0.04143657488748431, 0.04141264408826828, 0.04158451105467975, 0.041570396861061454, 0.04122957307845354, 0.041479747043922544, 0.041439895052462816, 0.04158888082019985, 0.04154233308508992, 0.041566161904484034, 0.04144020099192858, 0.04166302504017949, 0.04123422200791538, 0.04153492511250079, 0.04149805288761854, 0.04152029613032937, 0.04135488695465028, 0.04141349485144019, 0.04155754507519305, 0.041383167961612344, 0.04158809198997915, 0.04160774918273091, 0.041327707935124636, 0.04146323585882783, 0.04142413008958101, 0.04167694109492004, 0.04137906595133245, 0.041661523981019855, 0.04132568300701678, 0.04162081400863826, 0.04131980799138546, 0.04154318291693926, 0.04227334004826844, 0.04145440296269953, 0.04164815507829189, 0.04131323704496026, 0.041330006904900074, 0.04129326110705733, 0.041635039960965514, 0.04170000902377069, 0.04127154592424631, 0.041633134009316564, 0.04138417192734778, 0.04155281116254628, 0.04157286696135998, 0.041524289874359965, 0.041627810103818774, 0.04149365099146962, 0.041433851001784205, 0.041514975018799305, 0.04735471704043448, 0.044207034865394235, 0.04728094604797661, 0.04743593209423125, 0.047623659018427134, 0.04755046288482845, 0.04742663004435599, 0.04728076094761491, 0.04769860696978867, 0.04742938606068492, 0.04734039097093046, 0.047676404006779194, 0.04754363396205008, 0.047338044038042426, 0.047562514897435904, 0.04759287601336837, 0.047173672122880816, 0.04771790001541376, 0.047564367996528745, 0.04740140889771283, 0.04707260010764003, 0.04767477698624134, 0.04723342997021973, 0.04760177596472204, 0.04748307401314378, 0.047570934984833, 0.04759113793261349, 0.047379006166011095, 0.04760981886647642, 0.04761504800990224, 0.04750509397126734, 0.047451021149754524, 0.04733260697685182, 0.04754761094227433, 0.04722715402022004, 0.04745657299645245, 0.047571629052981734, 0.04961233283393085, 0.04583728010766208, 0.04733219789341092, 0.047915334114804864, 0.047571429051458836, 0.047581827035173774, 0.04798696702346206, 0.04767781891860068, 0.04759422200731933, 0.04727345006540418, 0.04756649094633758, 0.04755694791674614, 0.04747775802388787, 0.04811778501607478, 0.04746475210413337, 0.04731215396896005, 0.04761029197834432, 0.0475685999263078, 0.047442248091101646, 0.04730542888864875, 0.0474397970829159, 0.047440971015021205, 0.04746896889992058, 0.047447239980101585, 0.04813689598813653, 0.07289650500752032, 0.02232743799686432, 0.04908689996227622, 0.045844484120607376, 0.04226049198769033, 0.041649091988801956, 0.0415497359354049, 0.041643031174317, 0.04122382192872465, 0.041759273037314415, 0.04146979399956763, 0.04152000602334738, 0.04151212493889034, 0.041419476037845016, 0.04123077495023608, 0.042609574040398, 0.04160482087172568, 0.04151158104650676, 0.04161754599772394, 0.041349817998707294, 0.0415025120601058, 0.04184660199098289, 0.04138635192066431, 0.04174761311151087, 0.04142198385670781, 0.04133880604058504, 0.0413122761528939, 0.04246685188263655, 0.04173404909670353, 0.041482824832201004, 0.04142767400480807, 0.041293648071587086, 0.04138886299915612, 0.04165943106636405, 0.04161350196227431, 0.04149251896888018, 0.04139116290025413, 0.04139302717521787, 0.04127651592716575, 0.041706417920067906, 0.041681969072669744, 0.04189716302789748, 0.04152950691059232, 0.04135327902622521, 0.04153159400448203, 0.04221067810431123, 0.041926370933651924, 0.04156807204708457, 0.04165200795978308, 0.04149545100517571, 0.041434396058321, 0.04153254092670977, 0.04154350096359849, 0.041581327095627785, 0.04133269586600363, 0.04127129097469151, 0.041384712094441056, 0.04152444191277027, 0.04168624500744045, 0.04167100111953914, 0.041399036068469286, 0.041486692847684026, 0.04128154506906867, 0.041632445994764566, 0.04161126702092588, 0.041648896876722574, 0.041484843008220196, 0.04157204506918788, 0.04129905300214887, 0.04169945511966944, 0.04183224984444678, 0.042048979084938765, 0.04140097089111805, 0.041498657083138824, 0.04141841409727931, 0.04148240899667144, 0.04173765587620437, 0.04151879297569394, 0.04133818601258099, 0.04160151700489223, 0.041148470947518945, 0.04166006715968251, 0.04166551283560693, 0.04143556905910373, 0.0415638720151037, 0.04143093409948051, 0.04134886781685054, 0.04171604220755398, 0.041569609893485904, 0.04236634587869048, 0.04144146107137203, 0.04140809108503163, 0.04150574700906873, 0.041710508055984974, 0.04157720389775932, 0.04145012889057398, 0.041474248049780726, 0.04133116197772324, 0.04127816506661475, 0.04158255294896662, 0.04166774405166507, 0.04168430995196104, 0.04146586498245597, 0.04139873804524541, 0.041552430018782616, 0.041419797111302614, 0.04171141888946295, 0.041509435046464205, 0.04155790992081165, 0.0415079640224576, 0.04130333592183888, 0.04138395609334111, 0.04137223097495735, 0.041882445104420185, 0.04146515694446862, 0.041413180995732546, 0.041550186928361654, 0.04182835901156068, 0.04284984013065696, 0.04232280794531107, 0.04174222587607801, 0.04148402903228998, 0.041815388947725296, 0.04248856520280242, 0.04175676195882261, 0.041618803050369024, 0.0417275729123503, 0.04152817907743156, 0.04162953584454954, 0.041685187024995685, 0.04173377109691501, 0.04147798800840974, 0.042529074009507895, 0.04162849183194339, 0.04152804799377918, 0.04182932013645768, 0.04170582187362015, 0.041633062064647675, 0.04152065305970609, 0.04152206005528569, 0.0415346110239625, 0.04180402378551662, 0.04166149511002004, 0.04169814009219408, 0.041419612942263484, 0.04146927804686129, 0.04155806382186711, 0.04162828903645277, 0.04163775104098022, 0.04172448394820094, 0.04162042704410851, 0.04159216093830764, 0.04162831511348486, 0.0418340559117496, 0.041645076125860214, 0.04305058182217181, 0.041592546040192246, 0.041537696961313486, 0.04170721210539341, 0.04168923804536462, 0.04146130499430001, 0.0416753978934139, 0.04151945002377033, 0.04155974695459008, 0.04143142211250961, 0.04175829398445785, 0.04181201406754553, 0.04164511198177934, 0.041660717921331525, 0.041675454936921597, 0.04136693314649165, 0.041710232850164175, 0.04173339204862714, 0.04162277514114976, 0.042436397867277265, 0.041701236041262746, 0.04156208108179271, 0.041943860007449985, 0.041756682796403766, 0.04159344919025898, 0.04157270793803036, 0.04170340392738581, 0.04143866500817239, 0.04160785814747214, 0.041711902944371104, 0.041655879002064466, 0.04181100893765688, 0.041707348078489304, 0.041260889964178205, 0.04162610392086208, 0.04171605594456196, 0.041599085088819265, 0.04189883195795119, 0.041598242009058595, 0.04141185409389436, 0.041639810893684626, 0.041627053171396255, 0.04253307497128844, 0.04531291895546019, 0.046961485873907804, 0.047384114004671574, 0.048055593157187104, 0.04758816189132631, 0.047997083980590105, 0.047475194092839956, 0.04824359994381666, 0.047876226948574185, 0.0477426100987941, 0.04779514786787331, 0.04797733901068568, 0.04783016419969499, 0.04873604397289455, 0.04922886798158288, 0.04786268388852477, 0.04796724300831556, 0.04790073004551232, 0.047971578082069755, 0.04782069893553853, 0.047880667028948665, 0.047937646973878145, 0.04807315906509757, 0.04785339394584298, 0.047921909019351006, 0.04794744402170181, 0.047979313880205154, 0.04786823294125497, 0.047993630170822144, 0.04782182886265218, 0.04776632715947926, 0.04761125287041068, 0.047670339001342654, 0.04792229714803398, 0.04804640798829496, 0.047728226985782385, 0.04788062395527959, 0.048633412923663855], [0.047272922936826944, 0.04744482412934303, 0.047610908979550004, 0.047554651042446494, 0.0474200239405036, 0.04728784202598035, 0.047693255823105574, 0.047429986065253615, 0.04736399604007602, 0.04766658693552017, 0.0475458730943501, 0.0473248278722167, 0.04757218505255878, 0.04758682195097208, 0.04718734906055033, 0.047706719022244215, 0.04758257302455604, 0.0473978640511632, 0.04705661186017096, 0.04767920495942235, 0.047240857034921646, 0.04759668395854533, 0.047491165110841393, 0.04755945410579443, 0.0475957109592855, 0.0473756929859519, 0.04761149501428008, 0.04761827504262328, 0.04751804890111089, 0.047503761015832424, 0.047303007915616035, 0.04757250007241964, 0.04718016809783876, 0.04746515885926783, 0.04756080312654376, 0.049607745837420225, 0.04583280603401363, 0.04733112803660333, 0.04791471199132502, 0.04756662808358669, 0.047607004875317216, 0.048163847997784615, 0.047474017133936286, 0.04761642683297396, 0.04725123313255608, 0.04757584887556732, 0.04754518298432231, 0.047483249101787806, 0.048334576888009906, 0.047256658086553216, 0.047312828944996, 0.04760664398781955, 0.047563713043928146, 0.04744119197130203, 0.04731342499144375, 0.04743208712898195, 0.04745093593373895, 0.047472056932747364, 0.04745128704234958, 0.04826620500534773, 0.0727812540717423, 0.02228200202807784, 0.04908241587691009, 0.04583792295306921, 0.04225076409056783, 0.04164150799624622, 0.04156046803109348, 0.04167860094457865, 0.041191704105585814, 0.041706004878506064, 0.04147680406458676, 0.04151344601996243, 0.04150823107920587, 0.04142198292538524, 0.041228205896914005, 0.04261061595752835, 0.04160687210969627, 0.04151112888939679, 0.04161476111039519, 0.041353785898536444, 0.041502078995108604, 0.04185890220105648, 0.04137334483675659, 0.041748333955183625, 0.04142157407477498, 0.041339454939588904, 0.04131014319136739, 0.042491875821724534, 0.04176799999549985, 0.04146740818396211, 0.041463675908744335, 0.04126683110371232, 0.041363416938111186, 0.04164924891665578, 0.04160490911453962, 0.041489258874207735, 0.0414091341663152, 0.0413780128583312, 0.04127210797742009, 0.041709955083206296, 0.04167754203081131, 0.04189866897650063, 0.041530967922881246, 0.04135273303836584, 0.041530398884788156, 0.042234441032633185, 0.04195688501931727, 0.041568877175450325, 0.04165555885992944, 0.0415062140673399, 0.04148516897112131, 0.04150092089548707, 0.041495385114103556, 0.04157466907054186, 0.041305133840069175, 0.04126425110734999, 0.041386289056390524, 0.04153476282954216, 0.041680859168991446, 0.04166621086187661, 0.04140276997350156, 0.04148329910822213, 0.04128236905671656, 0.04163218894973397, 0.041612270986661315, 0.0416488170158118, 0.041486346861347556, 0.041574212023988366, 0.041298049967736006, 0.04170202510431409, 0.04182555410079658, 0.04210485890507698, 0.04140732390806079, 0.04153215512633324, 0.041385138873010874, 0.04144886415451765, 0.04172696592286229, 0.04150635190308094, 0.04134115902706981, 0.04159687110222876, 0.04115028795786202, 0.041661332128569484, 0.041670202976092696, 0.04143114201724529, 0.04156453185714781, 0.04142950312234461, 0.04134823987260461, 0.04171777796000242, 0.041543891187757254, 0.04245367995463312, 0.041435776045545936, 0.04141547693870962, 0.041487439069896936, 0.04177990695461631, 0.04152637696824968, 0.04141916590742767, 0.041473074117675424, 0.04130159900523722, 0.04128192691132426, 0.04158329893834889, 0.041663025971502066, 0.04168459610082209, 0.04146676207892597, 0.04139620182104409, 0.04155450011603534, 0.0414203389082104, 0.0417133760638535, 0.04150535701774061, 0.04155632294714451, 0.041511462070047855, 0.04130256501957774, 0.041384716052562, 0.041387945879250765, 0.0419152791146189, 0.041477616876363754, 0.04141827113926411, 0.04158542491495609, 0.04177370201796293, 0.04283539787866175, 0.042299309046939015, 0.04173674411140382, 0.04148539691232145, 0.04181280708871782, 0.042490408988669515, 0.04175549396313727, 0.04161978489719331, 0.041730029974132776, 0.04153251717798412, 0.04161844891496003, 0.0416863530408591, 0.0417391168884933, 0.04144735401496291, 0.04263619100674987, 0.04162668716162443, 0.041516580851748586, 0.04190001403912902, 0.041667248122394085, 0.04158324678428471, 0.04150683502666652, 0.04148905212059617, 0.04153471696190536, 0.04180706199258566, 0.041664727963507175, 0.04169660108163953, 0.04141931585036218, 0.0414815170224756, 0.0415454781614244, 0.041629913030192256, 0.04163494985550642, 0.04172734008170664, 0.04161565192043781, 0.041594414971768856, 0.04163033212535083, 0.04183378582820296, 0.04164284304715693, 0.04314212314784527, 0.04155928501859307, 0.04154235194437206, 0.041784646920859814, 0.041615585098043084, 0.04142161482013762, 0.04167476715520024, 0.041494824923574924, 0.04155789501965046, 0.04143248894251883, 0.041759886080399156, 0.04181282292120159, 0.04164328100159764, 0.041660513961687684, 0.041675688000395894, 0.041376344161108136, 0.04169987584464252, 0.04173338203690946, 0.04162551206536591, 0.04250982403755188, 0.04170047398656607, 0.041544490959495306, 0.04202804691158235, 0.041706173215061426, 0.041556072887033224, 0.041558193042874336, 0.04167098784819245, 0.041436958126723766, 0.04160688491538167, 0.04171270504593849, 0.04165581101551652, 0.04180932999588549, 0.04172476893290877, 0.04124609613791108, 0.04162476398050785, 0.04171774792484939, 0.04159906902350485, 0.04190019308589399, 0.041595991933718324, 0.0414132559671998, 0.041652752086520195, 0.04160092305392027, 0.042608596850186586, 0.04532733606174588, 0.04654624592512846, 0.04741136194206774, 0.04808663600124419, 0.047637853072956204, 0.047975498950108886, 0.0474944980815053, 0.0481400650460273, 0.0479077878408134, 0.04771800502203405, 0.04776411619968712, 0.04804936982691288, 0.04783132206648588, 0.04871587594971061, 0.04917897912673652, 0.04785666102543473, 0.04796535987406969, 0.04790766607038677, 0.047976911067962646, 0.04782588477246463, 0.04787745513021946, 0.047935240902006626, 0.048076941166073084, 0.04790337989106774, 0.04824406001716852, 0.04762488207779825, 0.047973390901461244, 0.04779869597405195, 0.048075698083266616, 0.04781974502839148, 0.04769390495494008, 0.04760478506796062, 0.04765963600948453, 0.0479219458065927, 0.04804069106467068, 0.04775770101696253, 0.04784943908452988, 0.048635191982612014, 0.062047821935266256, 0.039707513991743326, 0.05422589392401278, 0.049527639988809824, 0.0481569420080632, 0.050156033132225275, 0.04673204687424004, 0.047924659214913845, 0.04824371496215463, 0.047565551940351725, 0.04781380691565573, 0.047786704963073134, 0.04761026310734451, 0.04758551903069019, 0.047851074021309614, 0.04776660283096135, 0.047739157220348716, 0.04758202796801925, 0.0479159050155431, 0.048006237950176, 0.0475879970472306, 0.047785858856514096, 0.047750598983839154, 0.04789386014454067, 0.047737587010487914, 0.05197909497655928, 0.06204825988970697, 0.06446877005510032, 0.06484212609939277, 0.06474540778435767, 0.06446824502199888, 0.06450822111219168, 0.06463001691736281, 0.06489645014517009, 0.06472369283437729, 0.06499000615440309, 0.06467952695675194, 0.06510256207548082, 0.06468697194941342, 0.065023043891415, 0.06493548606522381, 0.06469140201807022, 0.06453321594744921, 0.06471443106420338, 0.0646974800620228, 0.06464078580029309, 0.06481992709450424, 0.06449157209135592, 0.06501777190715075, 0.06465720711275935, 0.06500287586823106, 0.06444549607113004, 0.06486089900135994, 0.06439282395876944, 0.06576786004006863, 0.06479344586841762, 0.06489251903258264, 0.06467348709702492, 0.0647843440528959, 0.06534085189923644, 0.06500500300899148, 0.06566725997254252, 0.06529155792668462, 0.06537628406658769, 0.06534308195114136, 0.06514995195902884, 0.06553789903409779, 0.06488928315229714, 0.06548838783055544, 0.06517729698680341, 0.06518499320372939, 0.0649649128317833, 0.06530352914705873, 0.06525276997126639, 0.06539868097752333, 0.06499797105789185, 0.06511675496585667, 0.06493488606065512, 0.06532774888910353, 0.06506380904465914, 0.06505323597230017, 0.06486863410100341, 0.06539900996722281, 0.06530813686549664, 0.06492840102873743, 0.065150354988873, 0.06520305806770921, 0.0659418620634824, 0.0649974758271128, 0.06547033903189003, 0.06459167995490134, 0.06616600719280541, 0.06484970590099692, 0.06469825399108231, 0.0659308610484004, 0.0647138929925859, 0.06529307994060218, 0.06538071110844612, 0.06469251681119204, 0.06523881806060672, 0.06559474114328623, 0.06487447884865105, 0.06495257909409702, 0.06524584791623056, 0.06547201704233885, 0.06468080705963075, 0.06566007295623422, 0.06512251193635166, 0.06520011508837342, 0.0648935700301081, 0.06529925600625575, 0.06489339889958501, 0.06551469606347382, 0.06498821801505983, 0.06509952992200851, 0.06486155511811376, 0.06506594293750823, 0.06500956602394581, 0.06520560500212014, 0.0650411038659513, 0.06560146412812173, 0.06516099697910249, 0.06527414196170866, 0.06522858701646328, 0.0652548170182854, 0.06506412406452, 0.06542468490079045, 0.06517165410332382, 0.06502483482472599, 0.0663137121591717, 0.06492615886963904, 0.06491380301304162, 0.06551421503536403, 0.06523679010570049, 0.06530235800892115, 0.06538055278360844, 0.06534123001620173, 0.06533495709300041, 0.06535359495319426, 0.0650712379720062, 0.06529623712413013, 0.06533006601966918, 0.06525714183226228, 0.06542782299220562, 0.0654640041757375, 0.06558238598518074, 0.06550104194320738, 0.06529402593150735, 0.06545751704834402, 0.06525750691071153, 0.0652506051119417, 0.06506971688941121, 0.06538201519288123, 0.06515621696598828, 0.06530187302269042, 0.06551272794604301, 0.06540867686271667, 0.06601505214348435, 0.06495590298436582, 0.06560026691295207, 0.06522719096392393, 0.06576678017154336, 0.06525894301012158, 0.0653517919126898, 0.06524748099036515, 0.06531891692429781, 0.06518229120410979, 0.06543179298751056, 0.06533244485035539, 0.06517865904606879, 0.0652564310003072, 0.06472424091771245, 0.06639107619412243, 0.06538050179369748, 0.06550788599997759, 0.06517263315618038, 0.06509560788981616, 0.06518802512437105, 0.06526945787481964, 0.06554330210201442, 0.06512076104991138, 0.0650171518791467, 0.06522753508761525, 0.0650518978945911, 0.06506626610644162, 0.06514824205078185, 0.06543024582788348, 0.06505951308645308, 0.06512584700249135, 0.06516049499623477, 0.0651411609724164, 0.06539194309152663, 0.06557235983200371, 0.0651802031788975, 0.0648763629142195, 0.0661814350169152, 0.06551103992387652, 0.06520219310186803, 0.06589829898439348, 0.06555223185569048, 0.06504818610846996, 0.06572749908082187, 0.06557970494031906, 0.06433480186387897, 0.06543035199865699, 0.06445117713883519, 0.06469488493166864, 0.06502505112439394, 0.06485635787248611, 0.06465713796205819, 0.06490218010731041, 0.06498465989716351, 0.06482710409909487, 0.06549078389070928], [0.05100224609486759, 0.04749282100237906, 0.048095920821651816, 0.047674708999693394, 0.047950875014066696, 0.047574416967108846, 0.04805799317546189, 0.047866606852039695, 0.04775717714801431, 0.04775435384362936, 0.04802873497828841, 0.04780860501341522, 0.04868663917295635, 0.04921356285922229, 0.047808130038902164, 0.04801338096149266, 0.04788386705331504, 0.047988510923460126, 0.047844580141827464, 0.04785701585933566, 0.047953044064342976, 0.04807416000403464, 0.04786499310284853, 0.048017095774412155, 0.04785132803954184, 0.04796028113923967, 0.04782823892310262, 0.048074689926579595, 0.04782203701324761, 0.04767606616951525, 0.0476773448754102, 0.04759796103462577, 0.04793203296139836, 0.04802808305248618, 0.04776943288743496, 0.04781571705825627, 0.048409911105409265, 0.06227729399688542, 0.03974728984758258, 0.05421058903448284, 0.04955044109374285, 0.048095575999468565, 0.05021935305558145, 0.046717053977772593, 0.047929149819538, 0.04823776404373348, 0.0475707920268178, 0.047811904922127724, 0.047799346037209034, 0.047605738043785095, 0.047576420940458775, 0.047850721050053835, 0.04776656907051802, 0.047745825024321675, 0.04758570482954383, 0.0479063221719116, 0.047978322952985764, 0.04766111099161208, 0.0477042100392282, 0.0477502578869462, 0.04788597603328526, 0.047767353942617774, 0.05195792415179312, 0.06200807588174939, 0.06447613006457686, 0.06483206804841757, 0.06474009994417429, 0.06444561900570989, 0.06455990904942155, 0.06459865299984813, 0.06487397383898497, 0.06479283422231674, 0.06496859388425946, 0.0646340420935303, 0.06512604001909494, 0.06468336191028357, 0.06503161392174661, 0.06492304895073175, 0.06468146620318294, 0.06452368479222059, 0.06470045121386647, 0.06474829581566155, 0.06463500810787082, 0.06484699994325638, 0.06447008694522083, 0.06501454603858292, 0.06465412396937609, 0.06497794413007796, 0.06447986303828657, 0.06484077288769186, 0.0644312771037221, 0.06575991585850716, 0.06478778808377683, 0.06490703602321446, 0.06469801999628544, 0.0647578330244869, 0.06515977694652975, 0.06516777910292149, 0.06569818989373744, 0.06523645995184779, 0.06556355208158493, 0.06536272191442549, 0.06514715496450663, 0.06554061500355601, 0.06488178297877312, 0.06549187516793609, 0.06517307600006461, 0.06517124688252807, 0.06497674598358572, 0.06523861410096288, 0.06532082404009998, 0.06589995697140694, 0.0645127878524363, 0.06515479902736843, 0.06493514217436314, 0.06532802386209369, 0.06506584212183952, 0.06505036680027843, 0.06487088720314205, 0.06539743393659592, 0.06530777388252318, 0.06493025203235447, 0.06514904112555087, 0.06520170997828245, 0.06612177193164825, 0.0651704769115895, 0.06541760405525565, 0.06424513808451593, 0.06653425889089704, 0.06443886505439878, 0.06462572608143091, 0.06595565192401409, 0.06476632412523031, 0.06531453388743103, 0.06530807400122285, 0.06509194290265441, 0.06523894402198493, 0.06525737210176885, 0.06488844705745578, 0.06494920887053013, 0.06524577410891652, 0.06547207105904818, 0.06505250185728073, 0.06555272010155022, 0.06485729990527034, 0.06519013713113964, 0.06495359097607434, 0.06558833899907768, 0.06480792490765452, 0.0651930149178952, 0.06505973101593554, 0.06510123400948942, 0.06485755112953484, 0.06506982282735407, 0.06500456901267171, 0.06520879617892206, 0.06537552899681032, 0.0656051249243319, 0.06518473802134395, 0.0649137960281223, 0.06522555788978934, 0.0652532980311662, 0.06506733107380569, 0.06541809090413153, 0.06517930096015334, 0.06502221408300102, 0.06631821510381997, 0.06491630082018673, 0.06516789016313851, 0.065259211929515, 0.06551079591736197, 0.06502549513243139, 0.06537519488483667, 0.06534244702197611, 0.06533895991742611, 0.06536200316622853, 0.06505919783376157, 0.06530171516351402, 0.06532692583277822, 0.0652596231084317, 0.06541422312147915, 0.0654821558855474, 0.06556381494738162, 0.06542538409121335, 0.06538352789357305, 0.06545650213956833, 0.06525238184258342, 0.06525556719861925, 0.06506638182327151, 0.06528768199495971, 0.06525678304024041, 0.06528383702971041, 0.0655228269752115, 0.06540918606333435, 0.06602698401547968, 0.06521694594994187, 0.06533461599610746, 0.06560642900876701, 0.0652553339023143, 0.06538816704414785, 0.06534208706580102, 0.06515574897639453, 0.06541383406147361, 0.06518761301413178, 0.06543370289728045, 0.06533110607415438, 0.0651769619435072, 0.0651391320861876, 0.06512342696078122, 0.06639470695517957, 0.06509310798719525, 0.06539044599048793, 0.06527789705432951, 0.06510749994777143, 0.06518736598081887, 0.06555313500575721, 0.0652444651350379, 0.06512931198813021, 0.0650027038063854, 0.0652468812186271, 0.06505153980106115, 0.06506604910828173, 0.06543468893505633, 0.06514320196583867, 0.06505676009692252, 0.06513246102258563, 0.06515638087876141, 0.06513966200873256, 0.065388984978199, 0.06557048414833844, 0.06518148793838918, 0.06517492793500423, 0.0657876250334084, 0.06560561596415937, 0.06519333203323185, 0.06619661813601851, 0.0656424870248884, 0.06495534395799041, 0.0652305509429425, 0.0656177350319922, 0.0645798419136554, 0.06511125713586807, 0.06460152892395854, 0.06480167689733207, 0.06496149208396673, 0.06480896705761552, 0.06466590682975948, 0.06499975104816258, 0.06498860195279121, 0.06482646521180868, 0.06546880980022252, 0.06495966599322855, 0.06876222719438374, 0.06602086802013218, 0.06501143798232079, 0.06581587693654001, 0.06526968092657626, 0.06471720710396767, 0.06511071790009737, 0.06518869497813284, 0.06520548416301608, 0.06514152092859149, 0.06512177595868707, 0.06516863103024662, 0.06490091793239117, 0.06487141503021121, 0.06494307611137629, 0.06538445083424449, 0.06515561603009701, 0.065352767938748, 0.06589565705507994, 0.06473960308358073, 0.06539152306504548, 0.06520209391601384, 0.06550363195128739, 0.06583386892452836, 0.06423816620372236, 0.06567044695839286, 0.06462662201374769, 0.06594524998217821, 0.06513717491179705, 0.06537359114736319, 0.0649673449806869, 0.06509495386853814, 0.0659695090726018, 0.06464518490247428, 0.06510120397433639, 0.06542843300849199, 0.0653323030564934, 0.06560559594072402, 0.06578784808516502, 0.06563133094459772, 0.06588128698058426, 0.06539530213922262, 0.06554997689090669, 0.06583176599815488, 0.0654923899564892, 0.06513224611990154, 0.06628733105026186, 0.06482262001372874, 0.06455677701160312, 0.06515526585280895, 0.06648806505836546, 0.06427279091440141, 0.06509902095422149, 0.06486546411179006, 0.0652687840629369, 0.06507356790825725, 0.06504029803909361, 0.06517244689166546, 0.06533021805807948, 0.06513067800551653, 0.06556152296252549, 0.06500685913488269, 0.06485477997921407, 0.0653210908640176, 0.06523769302293658, 0.06494935601949692, 0.07267189910635352, 0.0666656328830868, 0.0653514489531517, 0.06551144714467227, 0.06521756504662335, 0.06541007198393345, 0.06533128186129034, 0.06529646809212863, 0.0651029609143734, 0.06530018709599972, 0.0652276340406388, 0.06502687186002731, 0.06528334505856037, 0.06520824297331274, 0.0651575680822134, 0.06529020704329014, 0.06568961683660746, 0.06523638311773539, 0.0657259018626064, 0.06519240513443947, 0.0651792420540005, 0.06516730901785195, 0.0654990819748491, 0.06575288181193173, 0.06493221200071275, 0.06507583521306515, 0.06502399896271527, 0.06546939886175096, 0.06519047915935516, 0.06532465294003487, 0.06555909989401698, 0.0652575590647757, 0.06517401500605047, 0.06519094808027148, 0.0653876499272883, 0.06505723507143557, 0.06553235184401274, 0.0655928251799196, 0.06545669096522033, 0.06543254503048956, 0.06536142295226455, 0.06511220894753933, 0.06522358790971339, 0.06572116515599191, 0.0664578159339726, 0.06578200310468674, 0.06614236789755523, 0.06483883690088987, 0.06485139019787312, 0.06662171497009695, 0.06436062091961503, 0.06504261307418346, 0.06521049584262073, 0.06488747801631689, 0.06476555718109012, 0.06511450093239546, 0.06465940503403544, 0.06495103100314736, 0.06514161895029247, 0.06488511594943702, 0.0647517500910908, 0.06458354601636529, 0.06502183293923736, 0.06459483597427607, 0.06516699702478945, 0.0648140509147197, 0.0648729600943625, 0.06447173492051661, 0.06477114302106202, 0.06454319600015879, 0.06491952296346426, 0.0648571839556098, 0.06491100019775331, 0.06463863188400865, 0.06501462403684855, 0.06493565300479531, 0.06477796006947756, 0.06462744902819395, 0.06528458883985877, 0.06443421309813857, 0.0652308149728924, 0.0644248928874731, 0.06503979419358075, 0.0648868358694017, 0.06508648907765746, 0.06460475199855864, 0.06466399389319122, 0.06450754217803478, 0.06493768491782248, 0.06464063189923763, 0.06490873405709863, 0.06486338400281966, 0.06467500003054738, 0.06434697704389691, 0.0648232779931277, 0.06476884498260915, 0.06499146902933717, 0.0645577940158546, 0.06485455483198166, 0.0648092629853636, 0.0650272211059928, 0.06440009805373847, 0.06434402684681118, 0.06511826417408884, 0.06495804782025516, 0.06471393606625497, 0.06484767701476812, 0.0646160040050745, 0.06515669007785618, 0.06477160495705903, 0.06455154903233051, 0.06530996388755739, 0.06505496194586158, 0.06476073106750846, 0.06480818800628185, 0.06487586908042431, 0.06454102392308414, 0.06475239200517535, 0.06471732491627336, 0.064588573994115, 0.06703531020320952, 0.06299188197590411, 0.06461946293711662, 0.06500440207310021, 0.06344313290901482, 0.0663165741134435, 0.0646743020042777, 0.06481284787878394, 0.06482530408538878, 0.0641869988758117, 0.06533580110408366, 0.06456914893351495, 0.06452986691147089, 0.06483690720051527, 0.06463993480429053, 0.06400934117846191, 0.06519101490266621, 0.0644051160197705, 0.0650779630523175, 0.06515814200975001, 0.06478022201918066, 0.06451203185133636, 0.06484662904404104, 0.06472059013321996, 0.06453794077970088, 0.0657751401886344, 0.06342356186360121, 0.06445409497246146, 0.06477340403944254, 0.06470307498238981, 0.0647187631111592, 0.06479158694855869, 0.06416580802761018, 0.06462215189822018, 0.06494593201205134, 0.06424858211539686, 0.06439898395910859, 0.06452785898000002, 0.06426247791387141, 0.06482007820159197, 0.06441914779134095, 0.06470909807831049, 0.06430764310061932, 0.06483331881463528, 0.06483952607959509, 0.06418567197397351, 0.06469872896559536, 0.06429347419179976, 0.06442381697706878, 0.06484196288511157, 0.0648347029928118, 0.06468285294249654, 0.06475897319614887, 0.06476315692998469, 0.0647430659737438, 0.06432860111817718, 0.06485953880473971, 0.0654733651317656, 0.06392642785795033, 0.06446581403724849, 0.0643210329581052, 0.06478855712339282, 0.0646993878763169, 0.06422669813036919, 0.06469178991392255, 0.06422117701731622, 0.06387367192655802, 0.06497094710357487, 0.06476564495824277, 0.06462459894828498, 0.06441948609426618, 0.0647676601074636, 0.06563109997659922], [0.047968940110877156, 0.047903090016916394, 0.0479665519669652, 0.04782064096070826, 0.047881359001621604, 0.0479343559127301, 0.048079753993079066, 0.0478815280366689, 0.04801789508201182, 0.04785670293495059, 0.04797554505057633, 0.047838849015533924, 0.04802013887092471, 0.04782457719556987, 0.047734706895425916, 0.04761702590622008, 0.047671554144471884, 0.04791970504447818, 0.048044692957773805, 0.04772856994532049, 0.047881965059787035, 0.04867348284460604, 0.06174971698783338, 0.0396927772089839, 0.05422825599089265, 0.04952272796072066, 0.04816860193386674, 0.050163863925263286, 0.046848803060129285, 0.04791440791450441, 0.04823717521503568, 0.04755434882827103, 0.04781656316481531, 0.04779935488477349, 0.04761350713670254, 0.0475769757758826, 0.047851360170170665, 0.047769981902092695, 0.047742858063429594, 0.047578651923686266, 0.047907943138852715, 0.04799457499757409, 0.04760093684308231, 0.04777462617494166, 0.04775039479136467, 0.047894787043333054, 0.04773752414621413, 0.05197337898425758, 0.06213623587973416, 0.06446828995831311, 0.06484307115897536, 0.06474130204878747, 0.06446191691793501, 0.06453987001441419, 0.06460177688859403, 0.0648809049744159, 0.06475943909026682, 0.06498098094016314, 0.06469172495417297, 0.06509148119948804, 0.06462668301537633, 0.06508470280095935, 0.0649432041682303, 0.06469745980575681, 0.06453199801035225, 0.06468543712981045, 0.06472521391697228, 0.06463926215656102, 0.06485101394355297, 0.06447922601364553, 0.06500133499503136, 0.06467612995766103, 0.06507850205525756, 0.06431148201227188, 0.06479572784155607, 0.06439028796739876, 0.06578259519301355, 0.06478243181481957, 0.06490274611860514, 0.06468469905667007, 0.0648212160449475, 0.0651911988388747, 0.06523289810866117, 0.06567700998857617, 0.06526912888512015, 0.06472680205479264, 0.06546811503358185, 0.06514718802645802, 0.06552722305059433, 0.06479638488963246, 0.06548864603973925, 0.06517877290025353, 0.06517031602561474, 0.06496870703995228, 0.06534524890594184, 0.06522879912517965, 0.0658465689048171, 0.06457553501240909, 0.06519778096117079, 0.06493986421264708, 0.06532642478123307, 0.06504670204594731, 0.06506662210449576, 0.0648743559140712, 0.065392279997468, 0.06531312502920628, 0.06493030092678964, 0.06514276005327702, 0.06520175095647573, 0.06601377204060555, 0.0652875080704689, 0.0651448629796505, 0.06456184294074774, 0.06617574114352465, 0.0647363408934325, 0.06486202590167522, 0.0656431692186743, 0.06475539389066398, 0.06541538494639099, 0.06516689714044333, 0.06505313701927662, 0.0652088588103652, 0.06537074712105095, 0.06499896105378866, 0.06495676399208605, 0.06523694097995758, 0.06547845504246652, 0.06478137080557644, 0.06547349807806313, 0.06511829909868538, 0.06520181987434626, 0.06499853706918657, 0.06529110693372786, 0.06499173701740801, 0.06525366194546223, 0.06514107109978795, 0.06510764197446406, 0.06485040602274239, 0.06507558701559901, 0.06500646099448204, 0.0652051588986069, 0.0650485239457339, 0.06560767302289605, 0.0652451179921627, 0.06518185301683843, 0.06522627314552665, 0.06525104702450335, 0.06506693898700178, 0.06577245891094208, 0.06482392805628479, 0.0650255880318582, 0.06631376990117133, 0.06527133006602526, 0.06466601486317813, 0.0654090151656419, 0.06533765885978937, 0.06508858897723258, 0.06538252998143435, 0.06533466605469584, 0.06533778202719986, 0.06546805892139673, 0.06494171801023185, 0.06531276111491024, 0.06531416205689311, 0.0652577499859035, 0.06541917496360838, 0.06559735909104347, 0.06556687597185373, 0.06542803882621229, 0.06526942900381982, 0.06543514202348888, 0.06525430898182094, 0.0653670250903815, 0.06508016097359359, 0.0655039269477129, 0.06503842514939606, 0.06516966503113508, 0.06551285996101797, 0.06541467783972621, 0.06615079520270228, 0.06504681496880949, 0.06550294696353376, 0.0658622698392719, 0.06523077213205397, 0.06513110897503793, 0.06535537610761821, 0.06516243587248027, 0.06541141192428768, 0.06518962699919939, 0.06543435412459075, 0.0653258659876883, 0.06517413095571101, 0.0653781269211322, 0.06526501616463065, 0.06572698894888163, 0.06538323406130075, 0.06562715186737478, 0.06492110318504274, 0.06512288679368794, 0.06529071112163365, 0.06538180797360837, 0.06529457704164088, 0.06524871895089746, 0.06488510896451771, 0.06536326906643808, 0.0650519288610667, 0.06496105110272765, 0.06538023008033633, 0.06530615198425949, 0.06505345203913748, 0.06510701100341976, 0.06507002399303019, 0.0652529438957572, 0.06528273411095142, 0.06556079699657857, 0.06519015482626855, 0.06551811913959682, 0.0655491549987346, 0.06550514698028564, 0.06518662092275918, 0.06614531902596354, 0.06531687290407717, 0.06519418116658926, 0.06546882400289178, 0.06562414905056357, 0.06455987086519599, 0.06506415992043912, 0.06461923103779554, 0.06480195512995124, 0.06498877704143524, 0.06465018796734512, 0.06478862091898918, 0.06499860901385546, 0.06499218591488898, 0.06483039911836386, 0.06532430299557745, 0.06504559004679322, 0.06869543390348554, 0.06663063308224082, 0.06500777695327997, 0.06538858986459672, 0.06521541206166148, 0.06464901100844145, 0.06511535495519638, 0.06529513699933887, 0.06520492816343904, 0.06504181399941444, 0.06521342392079532, 0.06568285007961094, 0.0644005669746548, 0.06490538688376546, 0.06495515303686261, 0.06538528995588422, 0.06515562208369374, 0.06534783402457833, 0.06555283186025918, 0.06487935199402273, 0.06590773118659854, 0.06511457404121757, 0.06505228392779827, 0.06581890094093978, 0.06444207695312798, 0.0657032469753176, 0.06463187304325402, 0.06592744402587414, 0.06515479111112654, 0.06537205399945378, 0.06496947980485857, 0.06508991704322398, 0.06567662302404642, 0.06468158192001283, 0.06530381715856493, 0.06542519805952907, 0.06533463392406702, 0.06559913605451584, 0.06579121900722384, 0.06563846580684185, 0.06588226417079568, 0.06539330282248557, 0.0655544102191925, 0.06582737690769136, 0.06537901796400547, 0.06514207809232175, 0.06626884383149445, 0.06483342312276363, 0.06456511397846043, 0.06515493197366595, 0.06509985611774027, 0.0656069228425622, 0.0650987250264734, 0.06489382800646126, 0.06526723504066467, 0.06520493305288255, 0.06492049992084503, 0.06529761105775833, 0.06519053899683058, 0.06512762699276209, 0.06556040607392788, 0.0650154110044241, 0.0649719419889152, 0.06533134100027382, 0.06511678081005812, 0.06493853498250246, 0.0726713901385665, 0.06666984292678535, 0.0653299770783633, 0.0655335420742631, 0.06521549192257226, 0.06540697393938899, 0.06526104314252734, 0.06593422079458833, 0.06458928203210235, 0.06534515507519245, 0.0652146260254085, 0.06503649288788438, 0.06527997809462249, 0.06521291797980666, 0.065155966905877, 0.06561656319536269, 0.06539278500713408, 0.06511094188317657, 0.06572747905738652, 0.06519716395996511, 0.06530896690674126, 0.06515570799820125, 0.06551369512453675, 0.065615942934528, 0.06494092498905957, 0.06507630203850567, 0.06502538709901273, 0.06546729081310332, 0.06519714300520718, 0.06531935813836753, 0.06557113584131002, 0.06584403314627707, 0.06464490294456482, 0.06520748091861606, 0.06529468600638211, 0.06516507803462446, 0.0655307041015476, 0.06557667488232255, 0.06546038715168834, 0.06543627684004605, 0.06536721508018672, 0.06510527897626162, 0.06522583309561014, 0.06571642984636128, 0.06623316300101578, 0.0664426579605788, 0.06541644013486803, 0.06503106793388724, 0.06494509894400835, 0.06629173015244305, 0.064458112930879, 0.06506396294571459, 0.0652246861718595, 0.0648847259581089, 0.06488281092606485, 0.06500774505548179, 0.06465808884240687, 0.06506923423148692, 0.06494382093660533, 0.06492964387871325, 0.06490891496650875, 0.06458549504168332, 0.06490341108292341, 0.06450062897056341, 0.06536209909245372, 0.06468979083001614, 0.06487345602363348, 0.06461246614344418, 0.06473982497118413, 0.06456589396111667, 0.06492333090864122, 0.0647195940837264, 0.06490799505263567, 0.06476624682545662, 0.0650329771451652, 0.0650254669599235, 0.06467534508556128, 0.06463837996125221, 0.06512934295460582, 0.06508931703865528, 0.0646839588880539, 0.06432346696965396, 0.06502926908433437, 0.0648912200704217, 0.06508022896014154, 0.06461474299430847, 0.06465322710573673, 0.06461662892252207, 0.06483020796440542, 0.06477515795268118, 0.06491078902035952, 0.0645735829602927, 0.06483204406686127, 0.06449759309180081, 0.06481773196719587, 0.06476089102216065, 0.06485850783064961, 0.06469589401967824, 0.0647267100866884, 0.06492084404453635, 0.06489313999190927, 0.06453277985565364, 0.0644474751316011, 0.06501124706119299, 0.06482711480930448, 0.06468823319301009, 0.06501142587512732, 0.06447787396609783, 0.06526383012533188, 0.06480436190031469, 0.06452167103998363, 0.06533060106448829, 0.0649222768843174, 0.06476536416448653, 0.06475803488865495, 0.06491538905538619, 0.06468942505307496, 0.06461341795511544, 0.06471242988482118, 0.06471153697930276, 0.06681316113099456, 0.06316652684472501, 0.06454171799123287, 0.06514321616850793, 0.064355936832726, 0.0652459601406008, 0.06468707392923534, 0.06480605597607791, 0.06471886299550533, 0.06489937705919147, 0.06497645704075694, 0.06457160692662, 0.06488504493609071, 0.06484133610501885, 0.06463841791264713, 0.0640129370149225, 0.06518959999084473, 0.0644003739580512, 0.06471358612179756, 0.06517296191304922, 0.0647657960653305, 0.06451516808010638, 0.06485957885161042, 0.06472736410796642, 0.06488877790980041, 0.06538143404759467, 0.06345300795510411, 0.06442727614194155, 0.06479495181702077, 0.06471851305104792, 0.06507154600694776, 0.06443090713582933, 0.06452711299061775, 0.0647337248083204, 0.06447445508092642, 0.06425435096025467, 0.06438236101530492, 0.0645091759506613, 0.06414472102187574, 0.0649719291832298, 0.06480489997193217, 0.06429001991637051, 0.06470434693619609, 0.06445916602388024, 0.06468313606455922, 0.06471240404061973, 0.06432437989860773, 0.0642798850312829, 0.06484833406284451, 0.06482020788826048, 0.06435139407403767, 0.06477509601972997, 0.06461047008633614, 0.06476343399845064, 0.06458634091541171, 0.06461987597867846, 0.06457651709206402, 0.06545494683086872, 0.06412784196436405, 0.0644645991269499, 0.06431926600635052, 0.06463573686778545, 0.06452447618357837, 0.06450441386550665, 0.06460287095978856, 0.06418837420642376, 0.06398248788900673, 0.06516591995023191, 0.06433944101445377, 0.06479666498489678, 0.0645122081041336, 0.06443119398318231, 0.06578906299546361, 0.06416773400269449, 0.06471509300172329, 0.0684754210524261, 0.06668933085165918, 0.06504645710811019, 0.06497322884388268, 0.0652450651396066, 0.06512229703366756, 0.06501891696825624, 0.06519913207739592, 0.06504025380127132, 0.06530003319494426, 0.06527878693304956, 0.06620305590331554], [0.05124099715612829, 0.04791563889011741, 0.04823374003171921, 0.04756562993861735, 0.047812031116336584, 0.047794579062610865, 0.047612361842766404, 0.04758022399619222, 0.047854873118922114, 0.04776571085676551, 0.047741588205099106, 0.04757990688085556, 0.04791176994331181, 0.04795205709524453, 0.047643031924963, 0.047733192099258304, 0.047749958001077175, 0.04789129993878305, 0.04774002591148019, 0.05197635316289961, 0.062055367045104504, 0.06446543079800904, 0.06484919018112123, 0.0647359318099916, 0.06446976610459387, 0.06450717011466622, 0.06462861178442836, 0.06492222100496292, 0.06470294599421322, 0.0649950411170721, 0.06467255693860352, 0.06510655209422112, 0.06468577194027603, 0.06502441898919642, 0.06493991892784834, 0.0646895170211792, 0.06452798214741051, 0.06472266698256135, 0.06469299201853573, 0.06464106496423483, 0.06482606288045645, 0.06449190014973283, 0.06500971596688032, 0.06465665902942419, 0.06502584996633232, 0.06441219407133758, 0.0648417598567903, 0.06439419207163155, 0.06576756690628827, 0.06479433295316994, 0.06490011909045279, 0.06473413109779358, 0.06473164982162416, 0.06504004402086139, 0.06531718699261546, 0.06568233319558203, 0.06528580281883478, 0.06536756316199899, 0.0653349650092423, 0.06519890180788934, 0.06549047213047743, 0.06489125802181661, 0.06548897991888225, 0.06518035498447716, 0.06518308911472559, 0.06496339989826083, 0.06531007494777441, 0.06524878810159862, 0.06555472989566624, 0.06483634118922055, 0.0650957899633795, 0.06493545486591756, 0.0653291221242398, 0.06506272684782743, 0.06505596404895186, 0.0648676510900259, 0.06539810285903513, 0.0653096151072532, 0.06492701196111739, 0.06515092798508704, 0.06520313303917646, 0.06605736911296844, 0.0649068879429251, 0.06549302395433187, 0.06454432290047407, 0.06622125208377838, 0.06483874190598726, 0.06468460918404162, 0.06596771301701665, 0.06469149584881961, 0.06527976202778518, 0.06541518308222294, 0.06464401190169156, 0.06525784498080611, 0.06559522403404117, 0.06488113710656762, 0.06494225794449449, 0.06524350587278605, 0.06547740800306201, 0.06468183919787407, 0.06541599193587899, 0.06536504090763628, 0.0651973020285368, 0.06486939312890172, 0.06533085787668824, 0.06498766504228115, 0.06541737401857972, 0.06498194485902786, 0.06509919604286551, 0.06486217398196459, 0.0650668281596154, 0.06501036393456161, 0.06520214187912643, 0.06505276612006128, 0.0656040059402585, 0.06515517900697887, 0.06526731792837381, 0.06522973510436714, 0.06525687291286886, 0.06506293616257608, 0.0655185878276825, 0.06509692803956568, 0.06500286608934402, 0.06631658808328211, 0.06501722894608974, 0.06491956789977849, 0.06541887205094099, 0.06536045693792403, 0.06518637109547853, 0.06538169085979462, 0.06536128302104771, 0.06531890598125756, 0.06534460210241377, 0.06507642194628716, 0.06529210810549557, 0.06533283391036093, 0.06526074511930346, 0.06542698293924332, 0.06545872194692492, 0.06558060902170837, 0.06551456009037793, 0.06528830993920565, 0.0654541039839387, 0.06526534096337855, 0.06524207792244852, 0.06506858207285404, 0.0653842429164797, 0.06515695923008025, 0.06530158896930516, 0.06551378802396357, 0.06540857581421733, 0.06601078598760068, 0.06506701814942062, 0.06549446191638708, 0.06523514213040471, 0.06576190399937332, 0.06525609199889004, 0.06535354978404939, 0.06499662902206182, 0.06556688505224884, 0.06518081808462739, 0.06543181696906686, 0.06533432891592383, 0.06518390006385744, 0.06525589106604457, 0.06484217382967472, 0.06639402220025659, 0.06525951786898077, 0.06550623010843992, 0.06517581292428076, 0.06516460189595819, 0.06511431513354182, 0.06538766389712691, 0.06543456809595227, 0.06511931004934013, 0.06502163899131119, 0.06521949102170765, 0.06505718687549233, 0.06506378715857863, 0.06525890785269439, 0.06531236600130796, 0.06506490893661976, 0.06518470705486834, 0.06509964796714485, 0.06514363014139235, 0.06541723595000803, 0.06561421393416822, 0.06511920201592147, 0.06531693506985903, 0.06574481492862105, 0.06550983199849725, 0.06519895512610674, 0.06606607581488788, 0.0656317020766437, 0.06489734607748687, 0.06524287792854011, 0.06563341710716486, 0.0645000918302685, 0.06517159403301775, 0.06460839509963989, 0.0649635549634695, 0.06502204295247793, 0.06458893697708845, 0.06466158805415034, 0.0651695008855313, 0.06498802709393203, 0.0648221920710057, 0.06552069797180593, 0.06475639389827847, 0.06851632404141128, 0.0662666130810976, 0.06507846899330616, 0.06531452503986657, 0.06579534080810845, 0.06469349120743573, 0.06519400794059038, 0.06519269593991339, 0.0651006200350821, 0.06514878594316542, 0.06505888607352972, 0.06523565389215946, 0.0649352811742574, 0.06483887089416385, 0.06493584299460053, 0.06536433706060052, 0.06517071090638638, 0.06536540412344038, 0.06558024603873491, 0.06516289385035634, 0.06544238608330488, 0.06517369905486703, 0.06506815692409873, 0.06613549287430942, 0.06427205400541425, 0.06564083904959261, 0.06461999309249222, 0.06593808787874877, 0.06513306801207364, 0.06537159602157772, 0.06497811200097203, 0.0650855649728328, 0.06555569195188582, 0.06474192417226732, 0.06544507597573102, 0.0654273850377649, 0.06543081696145236, 0.06558832386508584, 0.06570414220914245, 0.06569525878876448, 0.06588802020996809, 0.06532654585316777, 0.06554506812244654, 0.0658356579951942, 0.06550495303235948, 0.06512914481572807, 0.06637635407969356, 0.06473417091183364, 0.06455689016729593, 0.06514933286234736, 0.06668529310263693, 0.06410073186270893, 0.06509368820115924, 0.06488226284272969, 0.06524752196855843, 0.0651696091517806, 0.06493895291350782, 0.06526527809910476, 0.06524015287868679, 0.06513111502863467, 0.06556374300271273, 0.06500864308327436, 0.06485285586677492, 0.06531589501537383, 0.06525511294603348, 0.06493720202706754, 0.07267398503609002, 0.06665536505170166, 0.06535391695797443, 0.06553842802532017, 0.06520061590708792, 0.06540262512862682, 0.06526502105407417, 0.06560747395269573, 0.06482287892140448, 0.06525674904696643, 0.06522950902581215, 0.06503003486432135, 0.065281106159091, 0.06520686787553132, 0.0651594870723784, 0.06550876190885901, 0.06554625718854368, 0.06524054892361164, 0.06573220295831561, 0.06519851693883538, 0.06526536704041064, 0.06514200009405613, 0.06552039901725948, 0.06566704902797937, 0.06493638991378248, 0.06506849592551589, 0.06503055710345507, 0.06546993087977171, 0.06520651397295296, 0.06531160906888545, 0.06555911106988788, 0.06554824393242598, 0.0647718149702996, 0.06521273497492075, 0.06539690704084933, 0.06504907505586743, 0.06553202006034553, 0.06559964781627059, 0.06545559712685645, 0.06549148703925312, 0.0652985058259219, 0.06511110300198197, 0.06522674509324133, 0.06572170904837549, 0.06610460299998522, 0.06662466796115041, 0.06532262195833027, 0.06516494904644787, 0.06438256497494876, 0.06673513096757233, 0.0644095460884273, 0.06505817896686494, 0.06521202903240919, 0.06488424190320075, 0.06506118504330516, 0.06482623494230211, 0.06466604117304087, 0.06522877491079271, 0.06478049606084824, 0.06493467791005969, 0.06506604491733015, 0.06458388618193567, 0.06473846989683807, 0.06450362992472947, 0.0655360349919647, 0.06452892511151731, 0.06487065996043384, 0.06476447498425841, 0.06477528903633356, 0.06454254407435656, 0.06491885101422668, 0.06456224084831774, 0.06491018109954894, 0.06493724486790597, 0.06503387703560293, 0.06491150101646781, 0.06477762595750391, 0.06463599507696927, 0.06497417506761849, 0.06498278095386922, 0.06469049490988255, 0.06442847498692572, 0.06502671213820577, 0.06489163101650774, 0.0650828150101006, 0.0646130118984729, 0.0645596340764314, 0.06460525398142636, 0.06493778596632183, 0.06502971099689603, 0.0648309129755944, 0.06441223807632923, 0.06481129489839077, 0.0646609771065414, 0.06482578301802278, 0.06476196902804077, 0.06469132588244975, 0.0648565711453557, 0.06455763592384756, 0.06510885804891586, 0.06472383392974734, 0.0647043960634619, 0.06426010094583035, 0.06520515703596175, 0.06464286288246512, 0.06470982916653156, 0.06516321189701557, 0.06430850410833955, 0.06515871989540756, 0.06507821008563042, 0.06455499888397753, 0.0653094679582864, 0.06474469997920096, 0.06476033804938197, 0.0647740550339222, 0.06490275403484702, 0.0648576789535582, 0.06444195099174976, 0.06471785297617316, 0.06489879498258233, 0.0670972689986229, 0.06259934115223587, 0.0646310169249773, 0.06532620103098452, 0.06418025586754084, 0.06525584007613361, 0.06468436005525291, 0.06480932491831481, 0.06471820012666285, 0.06488620094023645, 0.065184815088287, 0.06455923290923238, 0.06469898298382759, 0.06483575701713562, 0.06463922699913383, 0.06401418405584991, 0.06518772896379232, 0.06439569289796054, 0.0649176670704037, 0.0651587110478431, 0.06478558387607336, 0.06452002516016364, 0.06483241799287498, 0.06471160287037492, 0.06473837699741125, 0.06561681115999818, 0.06341677578166127, 0.06444930913858116, 0.06484776805154979, 0.06462104502134025, 0.06488469382748008, 0.06463498808443546, 0.06421226495876908, 0.06473341607488692, 0.06478135194629431, 0.0642405899707228, 0.06442369893193245, 0.0645190910436213, 0.06427963706664741, 0.0647882700432092, 0.06469947705045342, 0.06446752580814064, 0.06448859116062522, 0.06461995793506503, 0.06472765910439193, 0.06448089797049761, 0.06452357582747936, 0.06432056007906795, 0.06461200304329395, 0.06481796386651695, 0.06466085906140506, 0.0646632551215589, 0.06464087287895381, 0.06479655695147812, 0.06435491214506328, 0.06468168902210891, 0.06451091798953712, 0.06544388085603714, 0.06430741399526596, 0.0644912461284548, 0.06431610183790326, 0.0644194302149117, 0.06452095881104469, 0.06438443413935602, 0.06471957685425878, 0.06419504014775157, 0.06425000890158117, 0.06496756593696773, 0.06438836921006441, 0.06502355495467782, 0.06441722786985338, 0.06438129814341664, 0.06602958193980157, 0.06388620706275105, 0.06450902996584773, 0.06853950791992247, 0.066809575073421, 0.06493484601378441, 0.06496141804382205, 0.06524994689971209, 0.06512130005285144, 0.06502016796730459, 0.06519862287677824, 0.06504441308788955, 0.06529492698609829, 0.06526954495348036, 0.0662318211980164, 0.06697773491032422, 0.06561174406670034, 0.0677843987941742, 0.06570637412369251, 0.06418100488372147, 0.06492210505530238, 0.06411956995725632, 0.06431068200618029, 0.06421513203531504, 0.06561475805938244, 0.06367943994700909, 0.06419449695385993, 0.06422999920323491, 0.0643573438283056, 0.06439717812463641, 0.06511541199870408, 0.06408076896332204, 0.06403389689512551, 0.06404299614951015, 0.06419397983700037, 0.06400399003177881, 0.06419981317594647, 0.06454712199047208, 0.06384238693863153, 0.0642968718893826, 0.06449212902225554, 0.06378752994351089, 0.06503691500984132, 0.06476365099661052], [0.06860535009764135, 0.06447116099298, 0.06484453985467553, 0.06474028318189085, 0.06446087500080466, 0.06454020901583135, 0.06459943600930274, 0.0648788339458406, 0.06475624488666654, 0.06498973700217903, 0.0646877670660615, 0.06509797088801861, 0.06475207721814513, 0.0649553257972002, 0.06494153710082173, 0.06469860509969294, 0.06452857283875346, 0.06464834697544575, 0.06476535019464791, 0.06463909684680402, 0.064847222995013, 0.06448178715072572, 0.06500418088398874, 0.06467375112697482, 0.06505081779323518, 0.06435093400068581, 0.06481341412290931, 0.06439132685773075, 0.06578196422196925, 0.0647809929214418, 0.0649025640450418, 0.06468801689334214, 0.06480637611821294, 0.06531566986814141, 0.06508334912359715, 0.06567114894278347, 0.06528056506067514, 0.06472340202890337, 0.065331244841218, 0.06518255500122905, 0.06550295813940465, 0.06491675996221602, 0.06548982090316713, 0.0651799039915204, 0.0651694389525801, 0.06496887910179794, 0.06533854501321912, 0.06521711102686822, 0.06545260595157743, 0.06499390001408756, 0.06522373505868018, 0.06494021089747548, 0.06532375793904066, 0.06504801311530173, 0.06506658508442342, 0.06487474078312516, 0.06539270211942494, 0.06531111989170313, 0.06492575420998037, 0.06514659896492958, 0.06520242686383426, 0.06585530913434923, 0.06512411381118, 0.06539103109389544, 0.06463467399589717, 0.06609056796878576, 0.06473606708459556, 0.06467610388062894, 0.0653332881629467, 0.06528432201594114, 0.06530406000092626, 0.0652502398006618, 0.06502137007191777, 0.06515337992459536, 0.06549344607628882, 0.06488754414021969, 0.06495941081084311, 0.06522475904785097, 0.06559074111282825, 0.06468820082955062, 0.06556926318444312, 0.0651239468716085, 0.06520161312073469, 0.06503938301466405, 0.06525881588459015, 0.06487331609241664, 0.06555677484720945, 0.06483795610256493, 0.06511246901936829, 0.06484733195975423, 0.0650721259880811, 0.06501711392775178, 0.06519944500178099, 0.06516136415302753, 0.0656072199344635, 0.06514633004553616, 0.06516249687410891, 0.065232302993536, 0.06524278712458909, 0.06506831897422671, 0.06551646697334945, 0.06508793798275292, 0.06502320407889783, 0.06631584791466594, 0.06499786814674735, 0.0649410979822278, 0.06541013298556209, 0.06532224081456661, 0.06520432815887034, 0.06538356584496796, 0.06533435313031077, 0.06533809588290751, 0.0653545840177685, 0.06505443598143756, 0.06531286099925637, 0.06531374203041196, 0.06525810109451413, 0.06541893584653735, 0.0654657541308552, 0.06556819309480488, 0.06555446283891797, 0.06527196010574698, 0.065436074975878, 0.06525234691798687, 0.06523054512217641, 0.0650783758610487, 0.06539232307113707, 0.06514358008280396, 0.06531122582964599, 0.06551229604519904, 0.0654187761247158, 0.0660062488168478, 0.06504996912553906, 0.06548418407328427, 0.06544547388330102, 0.06550690811127424, 0.06528354180045426, 0.0653580161742866, 0.06516625988297164, 0.06540720094926655, 0.06518298899754882, 0.06543284701183438, 0.06533584208227694, 0.06515437201596797, 0.06527053308673203, 0.06496033584699035, 0.06629913696087897, 0.06523553607985377, 0.0655071719083935, 0.06518689705990255, 0.06511952704750001, 0.06515833805315197, 0.06538344593718648, 0.06542406510561705, 0.06511205597780645, 0.06502464297227561, 0.06522017694078386, 0.06504776398651302, 0.06510907597839832, 0.06523399194702506, 0.06531260721385479, 0.06504190480336547, 0.0651192341465503, 0.06520877196453512, 0.06511024106293917, 0.0654276788700372, 0.06555221206508577, 0.06519720493815839, 0.06583493598736823, 0.06606376799754798, 0.06467387708835304, 0.06516902707517147, 0.06601765379309654, 0.06539986119605601, 0.06526060379110277, 0.06533309118822217, 0.06560796103440225, 0.06474755192175508, 0.0649127708747983, 0.06456851912662387, 0.06480390205979347, 0.06499904091469944, 0.06479268008843064, 0.06464344193227589, 0.0650140920188278, 0.06498735793866217, 0.06480878195725381, 0.06648608413524926, 0.06385732698254287, 0.06873761396855116, 0.0662353818770498, 0.06509155803360045, 0.06540344608947635, 0.06558262789621949, 0.06470842496491969, 0.06511921714991331, 0.06528909294866025, 0.06509531196206808, 0.0651590300258249, 0.06504294509068131, 0.06513766292482615, 0.06503599300049245, 0.06496502994559705, 0.06497430894523859, 0.06538169807754457, 0.06515721092000604, 0.06534197507426143, 0.06581905111670494, 0.0648932799231261, 0.06526498589664698, 0.06520755612291396, 0.06521297106519341, 0.06566132395528257, 0.06465600500814617, 0.06575353303924203, 0.06463536596857011, 0.06592973601073027, 0.06515623000450432, 0.06536326999776065, 0.06497654295526445, 0.06508933193981647, 0.06553936493583024, 0.06467518699355423, 0.06531215622089803, 0.06544367293827236, 0.06535232183523476, 0.06560198403894901, 0.06577127915807068, 0.06565624987706542, 0.0658842280972749, 0.06536316382698715, 0.06557188113220036, 0.06581049994565547, 0.06549552408978343, 0.06514380685985088, 0.06653981911949813, 0.06456828187219799, 0.06456347811035812, 0.06515761301852763, 0.06509675085544586, 0.06557647814042866, 0.06512143486179411, 0.0649041070137173, 0.0652634531725198, 0.06520347390323877, 0.06491676205769181, 0.06529921386390924, 0.06519183120690286, 0.06512467097491026, 0.06555077084340155, 0.0650272301863879, 0.0648418718483299, 0.06530699296854436, 0.06526931514963508, 0.06493696384131908, 0.07266684202477336, 0.0666700100991875, 0.06533950194716454, 0.06552662211470306, 0.06522142980247736, 0.06540685403160751, 0.06525862915441394, 0.0654149898327887, 0.06513734511099756, 0.06537229707464576, 0.06521504698321223, 0.0650338928680867, 0.06528067192994058, 0.06521326513029635, 0.06515497108921409, 0.06525633693672717, 0.06555610802024603, 0.06524501089006662, 0.06573759694583714, 0.06519155716523528, 0.06531288498081267, 0.06504587596282363, 0.0656217080540955, 0.06561442604288459, 0.0649440698325634, 0.06505415006540716, 0.0650385080371052, 0.06547056185081601, 0.06520259799435735, 0.06531277298927307, 0.06556989601813257, 0.06525481911376119, 0.06514059891924262, 0.06536114099435508, 0.06547904596664011, 0.06497956719249487, 0.06553221098147333, 0.06557850982062519, 0.06545728212222457, 0.06543633807450533, 0.0653684688732028, 0.06510519701987505, 0.0652245890814811, 0.06571511900983751, 0.0661842548288405, 0.06584813818335533, 0.06594279292039573, 0.06494351103901863, 0.06468911003321409, 0.06663280492648482, 0.0642762400675565, 0.0650628178846091, 0.06525926315225661, 0.06486296281218529, 0.06489099701866508, 0.0649913870729506, 0.06467624101787806, 0.06507412786595523, 0.06496871821582317, 0.06491836300119758, 0.0648958149831742, 0.06459804787300527, 0.06489068898372352, 0.06451881700195372, 0.06533419108018279, 0.0647147020790726, 0.06485953298397362, 0.06462555378675461, 0.0647240960970521, 0.06456824392080307, 0.06489227199926972, 0.06473797699436545, 0.06491066003218293, 0.06476430199109018, 0.06503784819506109, 0.0649155518040061, 0.06479119416326284, 0.06462703598663211, 0.06512532592751086, 0.06479950505308807, 0.0648719510063529, 0.0644321609288454, 0.06501760613173246, 0.06489286688156426, 0.06505686300806701, 0.06465064291842282, 0.06463270704261959, 0.0645372150465846, 0.06492337305098772, 0.06479391199536622, 0.06489268597215414, 0.06450662901625037, 0.06487461505457759, 0.06451466889120638, 0.06480388389900327, 0.06475158501416445, 0.065985850058496, 0.06360659399069846, 0.06473394995555282, 0.06488937302492559, 0.06496495008468628, 0.0644546439871192, 0.06469969707541168, 0.06477207480929792, 0.06482987920753658, 0.0646808568853885, 0.0650144130922854, 0.06561167887412012, 0.0640182460192591, 0.06491166912019253, 0.06451474991627038, 0.06533165508881211, 0.06492538680322468, 0.06477687414735556, 0.06475034891627729, 0.06491830106824636, 0.06469843396916986, 0.06460164999589324, 0.06471958407200873, 0.06469911988824606, 0.06486795214004815, 0.0650696367956698, 0.06457726121880114, 0.06509921094402671, 0.06456022802740335, 0.06507827993482351, 0.0647056249435991, 0.06480510905385017, 0.06471389788202941, 0.06471569114364684, 0.06515378504991531, 0.0645737478043884, 0.06472151819616556, 0.06482982193119824, 0.06464285287074745, 0.06520138308405876, 0.06399823212996125, 0.06439883285202086, 0.06489668413996696, 0.06516973790712655, 0.06474775495007634, 0.064456069143489, 0.06491654692217708, 0.0647324260789901, 0.064700837014243, 0.06558154476806521, 0.06340656918473542, 0.06437787786126137, 0.06488737603649497, 0.06468918710015714, 0.06490749190561473, 0.06457083206623793, 0.06436521094292402, 0.06462781992740929, 0.06476841517724097, 0.06422764691524208, 0.06439211289398372, 0.06450951308943331, 0.06562527013011277, 0.06349780596792698, 0.06458802102133632, 0.06445045000873506, 0.0645619339775294, 0.06461896584369242, 0.06470659608021379, 0.06449878797866404, 0.06455413810908794, 0.06428061379119754, 0.06461603310890496, 0.06481540901586413, 0.06466016895137727, 0.06463986611925066, 0.06459389184601605, 0.06481604813598096, 0.06459169485606253, 0.06456627696752548, 0.06460661115124822, 0.0654774708673358, 0.06414752709679306, 0.06445978092961013, 0.06426653498783708, 0.06466725608333945, 0.06455519492737949, 0.06438898597843945, 0.06471432512626052, 0.06417770287953317, 0.06393441511318088, 0.06508387485519052, 0.06454364210367203, 0.064720249036327, 0.06448688800446689, 0.0659665260463953, 0.06678943196311593, 0.06171133299358189, 0.0645701470784843, 0.0685393048916012, 0.06698565604165196, 0.06477378495037556, 0.06502327090129256, 0.06524649704806507, 0.06511845299974084, 0.06499509001150727, 0.06522268592379987, 0.06502588302828372, 0.06529340404085815, 0.0652534191031009, 0.06618883088231087, 0.06717585306614637, 0.06564448890276253, 0.0678050140850246, 0.0657224680762738, 0.06418304704129696, 0.06493986980058253, 0.06411199318245053, 0.06431130692362785, 0.06421288498677313, 0.06455166498199105, 0.06471549300476909, 0.06419496005401015, 0.06423846702091396, 0.06434710090979934, 0.06440291204489768, 0.06511517986655235, 0.06418326520361006, 0.06394006684422493, 0.06402823515236378, 0.06421354692429304, 0.06401439709588885, 0.06420305278152227, 0.06449584499932826, 0.06385032017715275, 0.06430548196658492, 0.06447560293599963, 0.0640429500490427, 0.06480841594748199, 0.06474900408647954, 0.07433389686048031, 0.06861481606028974, 0.06491707195527852, 0.06385912001132965, 0.06375463795848191, 0.06430105306208134, 0.06344718509353697, 0.0637747950386256, 0.06364146480336785, 0.06395300221629441, 0.06345958495512605, 0.0639720419421792, 0.06426423788070679, 0.06337690004147589, 0.06375861004926264, 0.06389945512637496, 0.06395903695374727, 0.06362159200944006, 0.06358161196112633, 0.06440460192970932], [0.06535785901360214, 0.06518892385065556, 0.06548749306239188, 0.06487327604554594, 0.06551008205860853, 0.06518534081988037, 0.06515614711679518, 0.06497320695780218, 0.06529999803751707, 0.06524783303029835, 0.0654917808715254, 0.06495934701524675, 0.06523949513211846, 0.06494867289438844, 0.06533840601332486, 0.06503072893247008, 0.06507533020339906, 0.06488644098863006, 0.06538200890645385, 0.06529737892560661, 0.06494642817415297, 0.06514878594316542, 0.06520521803759038, 0.06584400194697082, 0.06514948699623346, 0.06530451402068138, 0.06467714789323509, 0.06598789896816015, 0.06474771001376212, 0.06483793817460537, 0.06530853291042149, 0.06523283990100026, 0.06534783612005413, 0.06519794790074229, 0.06506956205703318, 0.06516001094132662, 0.06543239206075668, 0.06493488908745348, 0.06495687994174659, 0.06523592094890773, 0.06535836705006659, 0.06493009394034743, 0.06550602009519935, 0.06513691786676645, 0.06522262305952609, 0.06500831595622003, 0.06523486715741456, 0.06494139484129846, 0.06538115511648357, 0.06499980599619448, 0.06512925704009831, 0.06484613101929426, 0.06507091480307281, 0.06501052319072187, 0.06520908395759761, 0.06512082600966096, 0.06562086194753647, 0.06516123609617352, 0.06517782690934837, 0.06520178099162877, 0.06527422391809523, 0.06505498010665178, 0.0655117048881948, 0.06511139217764139, 0.06501939496956766, 0.06630640802904963, 0.0649970609229058, 0.06495128106325865, 0.065397041849792, 0.06527929217554629, 0.06517729884944856, 0.06539262016303837, 0.06533413589932024, 0.06534718396142125, 0.06537533598020673, 0.06501266011036932, 0.06532977498136461, 0.06530902301892638, 0.06525688990950584, 0.06542648212052882, 0.06547818495891988, 0.06555937393568456, 0.06549048493616283, 0.06531352712772787, 0.06543945404700935, 0.0652319910004735, 0.06531508779153228, 0.06504356418736279, 0.06537039787508547, 0.06514036608859897, 0.0653167818672955, 0.06550493114627898, 0.06540088704787195, 0.06602218979969621, 0.06507830903865397, 0.06545074004679918, 0.06545613706111908, 0.06554962787777185, 0.06525671319104731, 0.06535335187800229, 0.06516907108016312, 0.06535995984449983, 0.06520773004740477, 0.06544821592979133, 0.06532973400317132, 0.065177635056898, 0.0652593751437962, 0.06495502288453281, 0.06628417409956455, 0.06527260085567832, 0.06549728894606233, 0.06516642915084958, 0.06511583901010454, 0.06515726004727185, 0.06540062185376883, 0.06538635515607893, 0.06511981785297394, 0.06503753713332117, 0.06524063600227237, 0.0650704528670758, 0.06505644507706165, 0.0652670948766172, 0.06530631217174232, 0.06503042788244784, 0.06512641813606024, 0.06518650893121958, 0.06513349595479667, 0.06540891202166677, 0.06556403008289635, 0.06518179993145168, 0.06549029005691409, 0.06579815899021924, 0.06526916893199086, 0.06518844491802156, 0.06604231405071914, 0.06533407093957067, 0.0652921951841563, 0.06533364579081535, 0.0656253460329026, 0.06473934417590499, 0.06485662888735533, 0.0645756630692631, 0.06478190189227462, 0.06506676599383354, 0.06473351898603141, 0.06465185014531016, 0.06500895484350622, 0.06500593619421124, 0.06481596478261054, 0.06537006818689406, 0.06498659285716712, 0.06876198807731271, 0.06623443099670112, 0.06511322502046824, 0.06537713785655797, 0.06545019522309303, 0.06482715788297355, 0.06511013908311725, 0.06524483091197908, 0.06517326808534563, 0.06511796382255852, 0.06507729319855571, 0.06516230083070695, 0.06505171302706003, 0.06495975912548602, 0.06497141881845891, 0.06539474916644394, 0.06514801783487201, 0.06533802906051278, 0.06563079706393182, 0.06484968797303736, 0.06547847506590188, 0.06520245387218893, 0.06521838810294867, 0.06570772104896605, 0.06463952991180122, 0.0657483721151948, 0.06459941994398832, 0.06597619690001011, 0.06514234701171517, 0.06535730697214603, 0.06499197497032583, 0.06507503218017519, 0.06540342001244426, 0.06478461297228932, 0.0653053808491677, 0.06545582902617753, 0.06537716300226748, 0.06558640603907406, 0.06577077694237232, 0.06564810103736818, 0.06590284197591245, 0.06537181208841503, 0.06555634085088968, 0.06581554305739701, 0.06544724316336215, 0.06514528393745422, 0.06636023591272533, 0.06474887696094811, 0.06456485716626048, 0.06515214685350657, 0.06510232598520815, 0.06555775017477572, 0.06510955980047584, 0.06492311717011034, 0.0652657060418278, 0.06518422882072628, 0.06491760397329926, 0.06526461918838322, 0.06525135296396911, 0.06511401594616473, 0.06557165295816958, 0.06500878394581378, 0.06489005615003407, 0.06527729099616408, 0.06523206504061818, 0.06496032699942589, 0.07267790683545172, 0.06664888118393719, 0.06535956193692982, 0.06551360990852118, 0.0652259171474725, 0.06542071490548551, 0.06524663092568517, 0.06546306307427585, 0.06510558095760643, 0.06537754507735372, 0.0652086769696325, 0.06505495705641806, 0.06525809597223997, 0.06519597792066634, 0.06517838500440121, 0.06528050894849002, 0.0655248670373112, 0.06523394002579153, 0.06573231099173427, 0.06520989793352783, 0.06528021115809679, 0.06511177192442119, 0.06553422496654093, 0.0656628089491278, 0.06493061315268278, 0.06506668101064861, 0.06501571997068822, 0.0654847570694983, 0.06521071982569993, 0.06529299216344953, 0.06558064091950655, 0.06529880408197641, 0.0651230129878968, 0.06536249001510441, 0.06537336180917919, 0.06510543404147029, 0.06550134811550379, 0.06557772890664637, 0.06546439812518656, 0.06541988090611994, 0.06536544696427882, 0.06512962514534593, 0.06523101986385882, 0.06568752997554839, 0.06622847099788487, 0.06581432512030005, 0.06579833081923425, 0.06507334113121033, 0.06471085106022656, 0.06646766001358628, 0.06437612697482109, 0.06505405600182712, 0.06525205494835973, 0.06489873304963112, 0.06488408194854856, 0.0649977489374578, 0.06468068109825253, 0.06507357698865235, 0.064966059057042, 0.06489107897505164, 0.06476704799570143, 0.06472540087997913, 0.0649151869583875, 0.06454338505864143, 0.06529514212161303, 0.06467776582576334, 0.06487501412630081, 0.0646206708624959, 0.0647680601105094, 0.06454017502255738, 0.064865656895563, 0.06476580817252398, 0.06488534389063716, 0.0647577850613743, 0.06505141803063452, 0.06479524192400277, 0.06495780288241804, 0.06460825307294726, 0.06510896910913289, 0.06483584083616734, 0.06485707499086857, 0.06443435419350863, 0.0650123618543148, 0.06486596097238362, 0.06509762513451278, 0.0646587738301605, 0.06462217518128455, 0.06450513191521168, 0.06492828205227852, 0.0648469920270145, 0.06483842898160219, 0.06463668495416641, 0.06479085492901504, 0.0645250550005585, 0.06477753212675452, 0.06473511201329529, 0.0651392750442028, 0.06449730577878654, 0.06471707904711366, 0.06485969806089997, 0.06489733490161598, 0.06454125000163913, 0.06480869115330279, 0.06465569091960788, 0.0648502770345658, 0.06464816606603563, 0.06499781901948154, 0.06474389601498842, 0.06491154385730624, 0.06489744293503463, 0.06451636203564703, 0.06533013610169291, 0.0649222859647125, 0.0647977408953011, 0.06472477409988642, 0.06492077396251261, 0.06471048900857568, 0.06456972495652735, 0.06473843404091895, 0.06470008613541722, 0.06487587187439203, 0.06503767007961869, 0.06461146986111999, 0.06512545701116323, 0.06436944496817887, 0.06522173713892698, 0.06471751886419952, 0.0648071370087564, 0.06471212417818606, 0.06472623301669955, 0.0651443179231137, 0.06459545996040106, 0.0647085439413786, 0.06483338004909456, 0.06462588300928473, 0.06427136203274131, 0.06493520503863692, 0.06446388293989003, 0.06484130397439003, 0.0651715169660747, 0.06473583006300032, 0.06447071209549904, 0.0649235500022769, 0.0647549438290298, 0.06466050306335092, 0.06461991206742823, 0.06439630105160177, 0.06434085592627525, 0.06496305693872273, 0.06464531691744924, 0.06486994703300297, 0.06461505009792745, 0.06436752690933645, 0.06459474097937346, 0.06475706701166928, 0.06429705605842173, 0.06439221301116049, 0.06445761490613222, 0.0645636001136154, 0.06459273002110422, 0.06458523287437856, 0.06443617912009358, 0.06452920706942677, 0.06460321880877018, 0.06474768905900419, 0.0645288520026952, 0.06451412499882281, 0.06427390896715224, 0.06462975800968707, 0.06475578411482275, 0.0647433998528868, 0.06460365909151733, 0.06469428888522089, 0.06474883714690804, 0.06458629202097654, 0.06453743088059127, 0.06463566399179399, 0.06544668506830931, 0.0641481711063534, 0.06447088997811079, 0.06430587591603398, 0.064656080910936, 0.06455868203192949, 0.0644069870468229, 0.06468395702540874, 0.06417659600265324, 0.06396484887227416, 0.06509207212366164, 0.0644802178721875, 0.06478166300803423, 0.06451223720796406, 0.06475884793326259, 0.06548814196139574, 0.06420437199994922, 0.06457956600934267, 0.06850516307167709, 0.06683352100662887, 0.0649691210128367, 0.06497590593062341, 0.06525492505170405, 0.06510627688840032, 0.06501903012394905, 0.06520542991347611, 0.06505899201147258, 0.06524837086908519, 0.06532416422851384, 0.0659035739954561, 0.06746610882692039, 0.06564176711253822, 0.06777880596928298, 0.06576572288759053, 0.06418839911930263, 0.06490979087539017, 0.06411578715778887, 0.0642924509011209, 0.06424687406979501, 0.06453686999157071, 0.06476139300502837, 0.06420037290081382, 0.06421684799715877, 0.06436413107439876, 0.06441015703603625, 0.06510404287837446, 0.06405310914851725, 0.06405563792213798, 0.06404241896234453, 0.06410820316523314, 0.06403319491073489, 0.06424855603836477, 0.06453424389474094, 0.06384061509743333, 0.06430341606028378, 0.06448706099763513, 0.06405488983727992, 0.06478114612400532, 0.06449052901007235, 0.0746372570283711, 0.06855914392508566, 0.06496431585401297, 0.06382064521312714, 0.06377637898549438, 0.06427888781763613, 0.06349272117950022, 0.0637741438113153, 0.0636710140388459, 0.06389229511842132, 0.0634525059722364, 0.06398798106238246, 0.0642355049494654, 0.06345333997160196, 0.0637345789000392, 0.0638457341119647, 0.06397036998532712, 0.06368278805166483, 0.06359006185084581, 0.06413379008881748, 0.06887477193959057, 0.048443658044561744, 0.04700971907004714, 0.04681416600942612, 0.0471117589622736, 0.046966108959168196, 0.04708713595755398, 0.04705353616736829, 0.0469433618709445, 0.04721092991530895, 0.046709492104128, 0.04664256889373064, 0.047169888159260154, 0.04681811900809407, 0.04720756085589528, 0.04681868804618716, 0.04707989492453635, 0.04701418802142143, 0.04713427205570042, 0.04702103813178837, 0.04717479879036546, 0.04698020615614951, 0.04670478799380362, 0.046789902029559016, 0.04738732404075563, 0.04705387889407575, 0.046811244916170835, 0.046954665100201964, 0.04697235603816807, 0.0473106240388006, 0.04702616180293262, 0.04688362218439579, 0.04756521596573293, 0.04732889588922262, 0.04685194697231054, 0.046798028983175755, 0.04769546608440578], [0.0651878931093961, 0.06537881097756326, 0.06536349398083985, 0.06531666498631239, 0.06534256786108017, 0.06507695722393692, 0.06529366178438067, 0.06533393915742636, 0.06526280893012881, 0.0654233149252832, 0.06545819714665413, 0.0655814039055258, 0.06552047305740416, 0.06528343888930976, 0.06545512215234339, 0.06526910094544291, 0.06524124508723617, 0.06507606501691043, 0.06537530780769885, 0.06515722209587693, 0.06530298292636871, 0.06551737012341619, 0.0654036239720881, 0.06601005606353283, 0.0650685029104352, 0.0654914379119873, 0.06535877217538655, 0.06564813200384378, 0.06525322492234409, 0.06535820197314024, 0.0651555908843875, 0.06540350220166147, 0.0651808229740709, 0.0654328849632293, 0.06533118500374258, 0.06518539297394454, 0.06525933207012713, 0.06484096893109381, 0.06639753794297576, 0.06525505310855806, 0.06550642102956772, 0.06517942785285413, 0.06500886217691004, 0.06526783388108015, 0.06539285695180297, 0.06543371616862714, 0.06511670583859086, 0.06502280314452946, 0.06521466979756951, 0.06506174011155963, 0.06506385002285242, 0.06525895302183926, 0.06530763395130634, 0.06506807100959122, 0.06503220205195248, 0.06525308499112725, 0.06514259683899581, 0.06542067416012287, 0.06545234494842589, 0.0652815499342978, 0.0653166831471026, 0.06574531085789204, 0.06551216007210314, 0.06519679189659655, 0.06590540613979101, 0.06582096195779741, 0.06486356304958463, 0.0652482898440212, 0.06563703319989145, 0.06452338793314993, 0.06522886292077601, 0.0646919310092926, 0.0647999569773674, 0.06506837112829089, 0.06470995885320008, 0.0647120310459286, 0.06500404793769121, 0.0649837302044034, 0.06482645496726036, 0.06546855601482093, 0.06476665381342173, 0.0686448561027646, 0.06612999807111919, 0.06519682798534632, 0.0652021870482713, 0.06581027084030211, 0.0646793469786644, 0.06500695110298693, 0.06537731387652457, 0.06509409914724529, 0.06515454384498298, 0.06510495813563466, 0.065423090942204, 0.0647020370233804, 0.06481999391689897, 0.0649254759773612, 0.06536662112921476, 0.06517003499902785, 0.06536616990342736, 0.06573763396590948, 0.06503741815686226, 0.06548731890507042, 0.06516991998068988, 0.06519372295588255, 0.06601317808963358, 0.06421793904155493, 0.06562008801847696, 0.06461996585130692, 0.06593728717416525, 0.06513317301869392, 0.06537030800245702, 0.06497945496812463, 0.06508659501560032, 0.06559758097864687, 0.06488666404038668, 0.06533840298652649, 0.06543050101026893, 0.06538586085662246, 0.06558859697543085, 0.06574600213207304, 0.06565585103817284, 0.06589240487664938, 0.06536645093001425, 0.06553421006537974, 0.06583741004578769, 0.06546226097270846, 0.06512949010357261, 0.06638092990033329, 0.06473273201845586, 0.0645532920025289, 0.06514733796939254, 0.06673049204982817, 0.0640644549857825, 0.0650896499864757, 0.06488095410168171, 0.06524832197465003, 0.06517253583297133, 0.06493613915517926, 0.06526993284933269, 0.06523326318711042, 0.06513390480540693, 0.06556604616343975, 0.06501126801595092, 0.06489696097560227, 0.06527218082919717, 0.06525132618844509, 0.06494356994517148, 0.07267170888371766, 0.06665003020316362, 0.06535466900095344, 0.06554030696861446, 0.06520133302547038, 0.06540319579653442, 0.06526244501583278, 0.06563733099028468, 0.06477926298975945, 0.06523846602067351, 0.06522922497242689, 0.06503129517659545, 0.06528145098127425, 0.06520732492208481, 0.06515797413885593, 0.06554614286869764, 0.06558272312395275, 0.06520006596110761, 0.06573585304431617, 0.06519988877698779, 0.06526825809851289, 0.06514705112203956, 0.06551104178652167, 0.06566739617846906, 0.06493724696338177, 0.06506428588181734, 0.0650345531757921, 0.06546976300887764, 0.06520942598581314, 0.06531420187093318, 0.06555213104002178, 0.06559287593699992, 0.06475035916082561, 0.06515879486687481, 0.06539623392745852, 0.06507005007006228, 0.06553094601258636, 0.06557926395907998, 0.06545644602738321, 0.06534909713082016, 0.06544026802293956, 0.06511112488806248, 0.06522798491641879, 0.06572136399336159, 0.06635046098381281, 0.06647999514825642, 0.06547265802510083, 0.06495326687581837, 0.06477275700308383, 0.06659382814541459, 0.06434038886800408, 0.06503753107972443, 0.06521142995916307, 0.064933598972857, 0.06489818496629596, 0.06497213896363974, 0.06466183299198747, 0.06509380112402141, 0.06496730307117105, 0.06488020694814622, 0.06488278787583113, 0.06458750599995255, 0.06492388807237148, 0.0645655810367316, 0.06532544502988458, 0.06464487593621016, 0.06487405998632312, 0.0646084260661155, 0.06477440590970218, 0.06454136199317873, 0.06492375815287232, 0.0647138839121908, 0.06491255294531584, 0.06477870303206146, 0.06503444793634117, 0.06491358717903495, 0.06477761385031044, 0.06464219698682427, 0.06513491296209395, 0.0648641271982342, 0.064796709921211, 0.06442717788740993, 0.06504551903344691, 0.06488611595705152, 0.0650902260094881, 0.06460174499079585, 0.06466070609167218, 0.06451257690787315, 0.06492854817770422, 0.06488400581292808, 0.06481578899547458, 0.06442289217375219, 0.06497276294976473, 0.06454381509684026, 0.06477945600636303, 0.06475635780952871, 0.06484940904192626, 0.06475303112529218, 0.06471391394734383, 0.0649013000074774, 0.06488541490398347, 0.06454005907289684, 0.0641958019696176, 0.0652781780809164, 0.06485766801051795, 0.06466340995393693, 0.06498696492053568, 0.06447581807151437, 0.06515965703874826, 0.0649135468993336, 0.06455030106008053, 0.06531418790109456, 0.06491188216023147, 0.06475272099487484, 0.06481186300516129, 0.06487734895199537, 0.06468250788748264, 0.06460545817390084, 0.06471812096424401, 0.06473804102279246, 0.06712994980625808, 0.06286165700294077, 0.06451536901295185, 0.06514944601804018, 0.06417900417000055, 0.06543492479249835, 0.06466566701419652, 0.06481931312009692, 0.06454557296819985, 0.06490506511181593, 0.06517074396833777, 0.06456022104248405, 0.06470056879334152, 0.06483317911624908, 0.06463834806345403, 0.06401660782285035, 0.06518359598703682, 0.06445167306810617, 0.06486252206377685, 0.06515710800886154, 0.0647840709425509, 0.06453277007676661, 0.06482638092711568, 0.0647633180487901, 0.06468544597737491, 0.06563007598742843, 0.06346173700876534, 0.06439974205568433, 0.0648504130076617, 0.06466935202479362, 0.06482982193119824, 0.06463465397246182, 0.0643822280690074, 0.06456202990375459, 0.06478033913299441, 0.06424348498694599, 0.06448686891235411, 0.06445658393204212, 0.06428402406163514, 0.06484369491226971, 0.06465337099507451, 0.06446090620011091, 0.06451071100309491, 0.06459200894460082, 0.064758240012452, 0.06444993894547224, 0.06452437094412744, 0.06436641397885978, 0.06456570001319051, 0.06481870193965733, 0.06468657008372247, 0.06463882210664451, 0.06470089196227491, 0.06481071189045906, 0.06447389908134937, 0.06461338303051889, 0.06457356899045408, 0.06548605486750603, 0.06409715721383691, 0.06451628590002656, 0.06427816208451986, 0.06459308485500515, 0.06434249109588563, 0.06464844499714673, 0.06468364200554788, 0.06417107093147933, 0.06404813006520271, 0.06502364599145949, 0.06453945604152977, 0.06485064700245857, 0.06439117481932044, 0.0645685070194304, 0.06585897319018841, 0.06387187680229545, 0.0645215311087668, 0.06851818901486695, 0.06681565288454294, 0.0649342481046915, 0.06495645502582192, 0.0652484770398587, 0.06522941892035306, 0.06491736695170403, 0.06519172294065356, 0.0650520280469209, 0.06528878910467029, 0.06526861689053476, 0.06625338899903, 0.06697284895926714, 0.06560269906185567, 0.0677907990757376, 0.06569695798680186, 0.06418276391923428, 0.06493953801691532, 0.06410670699551702, 0.06431887298822403, 0.06420168303884566, 0.06563192093744874, 0.06367081613279879, 0.0641847150400281, 0.06423461390659213, 0.06435717293061316, 0.06439319811761379, 0.06511836498975754, 0.06408744095824659, 0.06403277488425374, 0.0640336622018367, 0.06420110096223652, 0.06400495185516775, 0.06426784116774797, 0.06447977689094841, 0.06383475102484226, 0.06430107192136347, 0.06449042004533112, 0.06357709807343781, 0.06525083188898861, 0.06479527405463159, 0.07395038101822138, 0.06857293611392379, 0.06500340183265507, 0.06376959895715117, 0.06383627420291305, 0.06421926780603826, 0.06350885215215385, 0.06378435995429754, 0.06366681889630854, 0.06384220416657627, 0.06354032084345818, 0.06397602008655667, 0.06417980301193893, 0.06350689986720681, 0.06370497401803732, 0.06382869300432503, 0.06396017619408667, 0.06372831901535392, 0.06357550295069814, 0.06444025388918817, 0.06768401013687253, 0.04844443779438734, 0.04704800620675087, 0.04678204283118248, 0.04715758003294468, 0.0469217540230602, 0.04708816111087799, 0.04704935196787119, 0.046845298958942294, 0.04730532201938331, 0.04656643606722355, 0.046790977008640766, 0.04721286799758673, 0.0468191048130393, 0.04719873401336372, 0.04683092702180147, 0.04707840201444924, 0.04700991394929588, 0.047138171968981624, 0.046976851066574454, 0.04721329594030976, 0.04694000701420009, 0.04670134815387428, 0.04683978599496186, 0.047337829833850265, 0.04709880705922842, 0.04676720895804465, 0.04699891107156873, 0.04706312599591911, 0.04718110105022788, 0.0470635499805212, 0.046882265945896506, 0.047526201931759715, 0.04737720103003085, 0.046802743105217814, 0.04683895409107208, 0.04806576878763735, 0.05135422619059682, 0.049147674813866615, 0.0475477110594511, 0.04757090611383319, 0.047645767917856574, 0.047376122092828155, 0.04716765182092786, 0.04732069815509021, 0.047912859823554754, 0.04743241216056049, 0.04711606400087476, 0.047562767984345555, 0.04750105086714029, 0.04740790603682399, 0.04746545199304819, 0.04758798400871456, 0.04735953197814524, 0.0473180681001395, 0.047228333074599504, 0.04746661987155676, 0.047466322081163526, 0.04722586087882519, 0.04721575416624546, 0.0475131229031831, 0.04741568793542683, 0.04739762912504375, 0.04754226794466376, 0.04749636701308191, 0.047365593956783414, 0.047294710064306855, 0.04723416990600526, 0.047751254169270396, 0.04750080988742411, 0.04805630096234381, 0.0472688649315387, 0.04752721101976931, 0.04793882602825761, 0.04710288206115365, 0.04737009690143168, 0.04758322215639055, 0.04739306401461363, 0.047323203878477216, 0.047327438136562705, 0.04750695591792464, 0.04746900196187198, 0.04748934111557901, 0.04731206689029932, 0.047285868087783456, 0.047576274955645204, 0.04740309086628258, 0.04748894600197673, 0.0474878940731287, 0.04724841704592109, 0.047463746974244714, 0.04724302398972213, 0.04766329494304955, 0.04754996206611395, 0.047187844989821315, 0.04744511307217181, 0.047664016019552946, 0.04740227805450559, 0.04718722589313984, 0.04744538106024265, 0.04766363883391023, 0.04721793299540877, 0.04761865804903209, 0.04703181516379118, 0.0475051139947027, 0.047633597860112786, 0.04828955791890621], [0.06503418507054448, 0.06461140792816877, 0.06480570696294308, 0.0651320752222091, 0.06463616993278265, 0.06465970585122705, 0.06500005605630577, 0.06499170907773077, 0.06482402887195349, 0.06546510802581906, 0.06492935516871512, 0.06868047988973558, 0.06611551903188229, 0.06522651389241219, 0.06527531309984624, 0.06582195009104908, 0.06470531783998013, 0.06511549814604223, 0.06526281498372555, 0.0651140648405999, 0.06514843204058707, 0.06478722300380468, 0.06523146107792854, 0.06515985890291631, 0.06491784402169287, 0.06494624214246869, 0.0653660788666457, 0.06516565894708037, 0.06535715702921152, 0.06556530995294452, 0.06513583404012024, 0.06530457804910839, 0.06532699102535844, 0.0650621319655329, 0.06585416500456631, 0.06451180996373296, 0.06571859796531498, 0.0646286141127348, 0.06593544408679008, 0.06513545685447752, 0.0653697089292109, 0.06497540907002985, 0.06508907512761652, 0.06533499481156468, 0.06491472502239048, 0.06544801313430071, 0.06542772799730301, 0.0653486370574683, 0.06559310178272426, 0.06577926012687385, 0.06562989298254251, 0.06588265695609152, 0.06539775803685188, 0.06554810190573335, 0.06583580118604004, 0.06549673690460622, 0.06513205589726567, 0.06636966206133366, 0.06474087107926607, 0.06455456488765776, 0.06515296199359, 0.06661926815286279, 0.06415580585598946, 0.06509586796164513, 0.06486722407862544, 0.06526500405743718, 0.06516586593352258, 0.06494789314456284, 0.06525859795510769, 0.06524227699264884, 0.06513015390373766, 0.06556405499577522, 0.06501029315404594, 0.06484591797925532, 0.06532188388518989, 0.06525186798535287, 0.06494086002930999, 0.07267212704755366, 0.06666081002913415, 0.06535204197280109, 0.06553326803259552, 0.06520175887271762, 0.06540536298416555, 0.06510563311167061, 0.06555276503786445, 0.06505849282257259, 0.0652817680966109, 0.06522921402938664, 0.06502883788198233, 0.0652792570181191, 0.06520890002138913, 0.06515875714831054, 0.06546361884102225, 0.06554560805670917, 0.0652417279779911, 0.06572703295387328, 0.06519791204482317, 0.0652623800560832, 0.06507971999235451, 0.0655833100900054, 0.06566979992203414, 0.06493725092150271, 0.06507091410458088, 0.06502565299160779, 0.06546882097609341, 0.06520429789088666, 0.06531507219187915, 0.06555923400446773, 0.06529078586027026, 0.06492445594631135, 0.06536731612868607, 0.06538889091461897, 0.06505626603029668, 0.06553134694695473, 0.06559656001627445, 0.0654544869903475, 0.06543263304047287, 0.06536114099435508, 0.06511055212467909, 0.06522698979824781, 0.06572167319245636, 0.06607719301246107, 0.06651465292088687, 0.06538469600491226, 0.06501361588016152, 0.06451573898084462, 0.06674550613388419, 0.06443175696767867, 0.06506131705828011, 0.0652195408474654, 0.06488374713808298, 0.06488198600709438, 0.0650044959038496, 0.06466765003278852, 0.06506825005635619, 0.06493915407918394, 0.0649338229559362, 0.06491051684133708, 0.0645822761580348, 0.06490169302560389, 0.06449938379228115, 0.06537730200216174, 0.06468210322782397, 0.06487294379621744, 0.06461140606552362, 0.06474790093488991, 0.06455572810955346, 0.06492536189034581, 0.06471870909444988, 0.06491215294227004, 0.06476333807222545, 0.06503099901601672, 0.06477591092698276, 0.06492729298770428, 0.06463971501216292, 0.06513154413551092, 0.06483947299420834, 0.06466989195905626, 0.06459021498449147, 0.06502573704347014, 0.06489224592223763, 0.06508158799260855, 0.06461487291380763, 0.06480070715770125, 0.06421151896938682, 0.06509037897922099, 0.06476698187179863, 0.06492079515010118, 0.06457248795777559, 0.06482275505550206, 0.0645005030091852, 0.06481689889915287, 0.06476776907220483, 0.06485730200074613, 0.06469407305121422, 0.06472090492025018, 0.06493042293004692, 0.06489684409461915, 0.06452227500267327, 0.06445803702808917, 0.06501117395237088, 0.06482011708430946, 0.06469366094097495, 0.06501337699592113, 0.06447426602244377, 0.0649868929758668, 0.06508039007894695, 0.06452765688300133, 0.06533248606137931, 0.06491601001471281, 0.06476298393681645, 0.06476026796735823, 0.06491667404770851, 0.06468853005208075, 0.06461130292154849, 0.06471651489846408, 0.06470992602407932, 0.0669289231300354, 0.06339390296489, 0.06419756193645298, 0.06514067598618567, 0.06435732496902347, 0.06525505613535643, 0.06468506692908704, 0.06480962294153869, 0.0647218490485102, 0.06488585914485157, 0.06518015079200268, 0.06455817818641663, 0.06470327684655786, 0.06483563501387835, 0.0646338420920074, 0.06401723506860435, 0.06518777599558234, 0.06439717789180577, 0.06491818511858582, 0.06515706982463598, 0.0647821081802249, 0.06451706378720701, 0.06483726715669036, 0.06471490184776485, 0.06472715898416936, 0.06560818618163466, 0.06341166701167822, 0.06446019583381712, 0.06477112113498151, 0.06470025493763387, 0.06489495607092977, 0.06461585382930934, 0.06448999722488225, 0.06430281279608607, 0.06494823819957674, 0.06423966586589813, 0.06440444011241198, 0.06453481991775334, 0.06426868890412152, 0.06480271206237376, 0.06466319505125284, 0.064474584069103, 0.06451078900136054, 0.06462385598570108, 0.06450752099044621, 0.0647032568231225, 0.06451682606711984, 0.0643186301458627, 0.06461803195998073, 0.06481887400150299, 0.06465566204860806, 0.06466628680936992, 0.06443401495926082, 0.06475370121188462, 0.06459450791589916, 0.06497301603667438, 0.06420422694645822, 0.06546748802065849, 0.06411580508574843, 0.06446819589473307, 0.06431891000829637, 0.06463154591619968, 0.06452466710470617, 0.06420633988454938, 0.06489863712340593, 0.06419544201344252, 0.06400207104161382, 0.06484331283718348, 0.06473583704791963, 0.06481921393424273, 0.06439901911653578, 0.06458600796759129, 0.0658035280648619, 0.06413108087144792, 0.06423189118504524, 0.06882817693985999, 0.06680355593562126, 0.06493580201640725, 0.06497197598218918, 0.06524571496993303, 0.06512232194654644, 0.06501999706961215, 0.06519811507314444, 0.06504042004235089, 0.06530035892501473, 0.0652714038733393, 0.06623297813348472, 0.06696021207608283, 0.06562279886566103, 0.06777469511143863, 0.06572077982127666, 0.06418155902065337, 0.06490919715724885, 0.06411697692237794, 0.0643146550282836, 0.06421436509117484, 0.06559606781229377, 0.0637060240842402, 0.06419205805286765, 0.06423201085999608, 0.06435543997213244, 0.06439915299415588, 0.06511286017484963, 0.06407853495329618, 0.06402732199057937, 0.06404855498112738, 0.06419099704362452, 0.06400486407801509, 0.06411559786647558, 0.06463126395829022, 0.06384235410951078, 0.064299640012905, 0.06447812193073332, 0.06380327208898962, 0.06503550196066499, 0.06474507693201303, 0.07419077912345529, 0.06854636687785387, 0.06500509497709572, 0.06377207394689322, 0.06375309615395963, 0.06430053105577826, 0.06344007281586528, 0.06377643998712301, 0.06374050909653306, 0.06385349109768867, 0.0634591628331691, 0.06397822196595371, 0.06425779219716787, 0.06349686090834439, 0.06363716395571828, 0.0639022330287844, 0.06396155711263418, 0.06372386100701988, 0.06357151386328042, 0.06444424111396074, 0.06766065396368504, 0.048454975010827184, 0.04697298794053495, 0.046863645082339644, 0.04708909592591226, 0.04698670492507517, 0.04707330698147416, 0.047058564145118, 0.047168158926069736, 0.04698608093895018, 0.04688352206721902, 0.0464771909173578, 0.04720597807317972, 0.046752942027524114, 0.04719833307899535, 0.04683263599872589, 0.04707484180107713, 0.04700768110342324, 0.047147298930212855, 0.04704216402024031, 0.04712942894548178, 0.047023910097777843, 0.046702010091394186, 0.04676758893765509, 0.04740747902542353, 0.04702551802620292, 0.04683638201095164, 0.046938268933445215, 0.04723413893952966, 0.04707912402227521, 0.046985807130113244, 0.04688470182009041, 0.047599771060049534, 0.04737620102241635, 0.04680151492357254, 0.04676756705157459, 0.0478455601260066, 0.05163479782640934, 0.049147179117426276, 0.04755428992211819, 0.0479029871057719, 0.04731607995927334, 0.0473752289544791, 0.047170104924589396, 0.04731948208063841, 0.04790970403701067, 0.04743684106506407, 0.04711513384245336, 0.047560669016093016, 0.04750234307721257, 0.04739627195522189, 0.047479084925726056, 0.04757896903902292, 0.047364842146635056, 0.04731893097050488, 0.04723254591226578, 0.04746294999495149, 0.04745599697344005, 0.04723518714308739, 0.0472110309638083, 0.04751116083934903, 0.04742058599367738, 0.047398122027516365, 0.0475301060359925, 0.04750777408480644, 0.04736102698370814, 0.047301959013566375, 0.047232344979420304, 0.04774830304086208, 0.04750028788112104, 0.04804006009362638, 0.047281961888074875, 0.04752468015067279, 0.04781513288617134, 0.04720520111732185, 0.04738400690257549, 0.04758938611485064, 0.04738568188622594, 0.0473353520501405, 0.04732524394057691, 0.04750377405434847, 0.04747436800971627, 0.04748924798332155, 0.047301836078986526, 0.04728788789361715, 0.0475808079354465, 0.04739649407565594, 0.04749825689941645, 0.047486338997259736, 0.047248485032469034, 0.047466766089200974, 0.047246909933164716, 0.047652228036895394, 0.04753646394237876, 0.04721030895598233, 0.047424665186554193, 0.04764166893437505, 0.047434208914637566, 0.04719011718407273, 0.047448732890188694, 0.047633976908400655, 0.0472432270180434, 0.04762452305294573, 0.04702993202954531, 0.04750211304053664, 0.04763387097045779, 0.04827647702768445, 0.051592350006103516, 0.04285446903668344, 0.041651755804196, 0.04133101017214358, 0.04132977896369994, 0.04151323204860091, 0.04168497794307768, 0.041306124068796635, 0.04150222102180123, 0.04143577883951366, 0.04155389894731343, 0.04125624313019216, 0.04141258401796222, 0.041712203063070774, 0.041104502975940704, 0.04144965880550444, 0.041459121042862535, 0.04140272596850991, 0.041451615979894996, 0.04133083508349955, 0.04150100192055106, 0.041557028191164136, 0.04153004102408886, 0.04160098684951663, 0.04111780598759651, 0.04167741094715893, 0.04157659015618265, 0.04140384495258331, 0.04143287497572601, 0.04156021401286125, 0.041425005067139864, 0.041437720879912376, 0.04128894908353686, 0.04150941199623048, 0.041404000017791986, 0.0414192290045321, 0.041857329895719886, 0.04155798000283539, 0.041574668139219284, 0.04150272300466895, 0.041770635871216655, 0.04154030093923211, 0.04151163809001446, 0.04147343710064888, 0.04146555392071605, 0.04156105197034776, 0.04654422611929476, 0.0444072459358722, 0.04753221292048693, 0.047501961002126336, 0.04764621495269239, 0.047356615075841546, 0.047512954100966454, 0.047552544856444, 0.04772167908959091, 0.04744867398403585, 0.047598245088011026, 0.04733288288116455, 0.047513277968391776, 0.04832300613634288, 0.047496907878667116, 0.04769515711814165, 0.04764945385977626, 0.04752096114680171, 0.04770331783220172, 0.04788356716744602, 0.04740988998673856, 0.0475317919626832, 0.04741912102326751, 0.048335749888792634], [0.06505885999649763, 0.06526158307678998, 0.06502356403507292, 0.06488421000540257, 0.06498800008557737, 0.0646733429748565, 0.0650785209145397, 0.06493958388455212, 0.06479299813508987, 0.0649010029155761, 0.06459273002110422, 0.06504307803697884, 0.0644953299779445, 0.06536194612272084, 0.06455781287513673, 0.06486322800628841, 0.06462184502743185, 0.06473318999633193, 0.06456324202008545, 0.06489264406263828, 0.0647386908531189, 0.06490854499861598, 0.06477030017413199, 0.06503117177635431, 0.06492027919739485, 0.06478711101226509, 0.0646320718806237, 0.0651272339746356, 0.06496036611497402, 0.06470895605161786, 0.06442739395424724, 0.06502100685611367, 0.06489588296972215, 0.06505805905908346, 0.06464765500277281, 0.06464101909659803, 0.0645321428310126, 0.06492458819411695, 0.06491566100157797, 0.06477062497287989, 0.06470183585770428, 0.06484448816627264, 0.06449901591986418, 0.06481639388948679, 0.06458988902159035, 0.06487582810223103, 0.06485187099315226, 0.06473122211173177, 0.06476331292651594, 0.06472070701420307, 0.06485531805083156, 0.06454533478245139, 0.06476633413694799, 0.06497194990515709, 0.06454280600883067, 0.06501308316364884, 0.0644695539958775, 0.06515782186761498, 0.06491297692991793, 0.06451418902724981, 0.06533251912333071, 0.06492570997215807, 0.0647770909126848, 0.06474848813377321, 0.06491932203061879, 0.06469802581705153, 0.06460602511651814, 0.06471868604421616, 0.06470152991823852, 0.06486368994228542, 0.0650886760558933, 0.06455716211348772, 0.065290437778458, 0.06437107315286994, 0.06507784989662468, 0.06470792298205197, 0.06479989807121456, 0.06472102995030582, 0.06471075699664652, 0.0651590321213007, 0.06457367283292115, 0.06471801619045436, 0.06483013089746237, 0.0646442431025207, 0.0640049409121275, 0.0651940640527755, 0.06455012690275908, 0.06474773096852005, 0.06517006200738251, 0.06475102412514389, 0.06445864005945623, 0.06491010682657361, 0.06488963216543198, 0.06454841187223792, 0.06565932906232774, 0.06349857291206717, 0.06420884910039604, 0.06502571888267994, 0.06471077096648514, 0.06474186200648546, 0.0647635841742158, 0.06433869781903923, 0.06446700100786984, 0.06476976699195802, 0.06441708304919302, 0.0643772091716528, 0.06434638984501362, 0.06429773499257863, 0.06498142797499895, 0.06442591897211969, 0.06445883819833398, 0.06455934187397361, 0.06461949599906802, 0.0647168371360749, 0.06468177679926157, 0.06435577012598515, 0.06428469391539693, 0.06461876700632274, 0.06481589912436903, 0.0650237868539989, 0.06446494814008474, 0.06462269695475698, 0.06475620088167489, 0.06442956905812025, 0.06477169413119555, 0.06460117688402534, 0.06528108892962337, 0.06414731917902827, 0.06446025590412319, 0.064465356990695, 0.06465146504342556, 0.06452618702314794, 0.06438442505896091, 0.06471892492845654, 0.06419152184389532, 0.06397570110857487, 0.06504547502845526, 0.06474920897744596, 0.06480843503959477, 0.06421022303402424, 0.0644096708856523, 0.06578481709584594, 0.06450504902750254, 0.06510358792729676, 0.06777009507641196, 0.06649362295866013, 0.0650814960245043, 0.0650177130009979, 0.06524472101591527, 0.0651203659363091, 0.06500687007792294, 0.06521176197566092, 0.06502645299769938, 0.06529524200595915, 0.06542136403732002, 0.06618551886640489, 0.0671756649389863, 0.06563389115035534, 0.06765662995167077, 0.06584484688937664, 0.06418270315043628, 0.06481186486780643, 0.06411368306726217, 0.06430925289168954, 0.0642141760326922, 0.06454813410528004, 0.06486015603877604, 0.0641836088616401, 0.06423741998150945, 0.06435321597382426, 0.0644006731454283, 0.06510982103645802, 0.06407307786867023, 0.06393612595275044, 0.06414374615997076, 0.06409240700304508, 0.06401261384598911, 0.06420880113728344, 0.06462860107421875, 0.0638336599804461, 0.06431348691694438, 0.06447282899171114, 0.06389007903635502, 0.06495313602499664, 0.06474491604603827, 0.07455162098631263, 0.06862498191185296, 0.0649555439595133, 0.06381872901692986, 0.06379097397439182, 0.0642613300587982, 0.06346782087348402, 0.06378440209664404, 0.0636736040469259, 0.06387156900018454, 0.0635041450150311, 0.06397439585998654, 0.0641705181915313, 0.06351359793916345, 0.0637243629898876, 0.0638044560328126, 0.06401109299622476, 0.06368161202408373, 0.0635750878136605, 0.06444156705401838, 0.06712759914807975, 0.04844681196846068, 0.04703550599515438, 0.046813166001811624, 0.04714995785616338, 0.046924219001084566, 0.04710288601927459, 0.04702556016854942, 0.046960902865976095, 0.047208643052726984, 0.046728882007300854, 0.04661936988122761, 0.047192807076498866, 0.04682614211924374, 0.04719537287019193, 0.0468349801376462, 0.04706269293092191, 0.04701703484170139, 0.04714751010760665, 0.046999370912089944, 0.04717283300124109, 0.046970120165497065, 0.04670931096188724, 0.046815939946100116, 0.047322058118879795, 0.047087881015613675, 0.04680936597287655, 0.04698815895244479, 0.047222119057551026, 0.047058493830263615, 0.047013610135763884, 0.0468873200006783, 0.04754202184267342, 0.047361240023747087, 0.04682216118089855, 0.04682097188197076, 0.04757228493690491, 0.05182639416307211, 0.04916503094136715, 0.04757680092006922, 0.04773334204219282, 0.04749154206365347, 0.0473765570204705, 0.04716470884159207, 0.04732197499834001, 0.04791625612415373, 0.047435970045626163, 0.0471136758569628, 0.047561213141307235, 0.047507860930636525, 0.04739215597510338, 0.047478958033025265, 0.04756757500581443, 0.04737759497947991, 0.04731543594971299, 0.04723061295226216, 0.04746384802274406, 0.047463873168453574, 0.047231877921149135, 0.04719292395748198, 0.04752268991433084, 0.047418152214959264, 0.047406858997419477, 0.04750743485055864, 0.04752964712679386, 0.047361213946715, 0.04729120200499892, 0.04722090996801853, 0.04776049195788801, 0.047494387021288276, 0.04798644292168319, 0.04730312013998628, 0.04755205498076975, 0.04765987489372492, 0.0473194511141628, 0.04741544905118644, 0.047600082820281386, 0.04738571308553219, 0.04734771396033466, 0.047309905057772994, 0.04751703189685941, 0.04747166018933058, 0.0474936249665916, 0.04727473692037165, 0.047293533105403185, 0.04759809491224587, 0.04738555592484772, 0.047491939971223474, 0.047495233127847314, 0.04725658195093274, 0.047466501127928495, 0.04725121590308845, 0.04763839300721884, 0.04755055787973106, 0.0472103210631758, 0.04741959902457893, 0.04769366001710296, 0.04738450003787875, 0.047185763949528337, 0.04743435210548341, 0.04761904990300536, 0.04727222304791212, 0.04762784601189196, 0.04701894288882613, 0.047508009942248464, 0.04763870616443455, 0.04809338483028114, 0.051798214204609394, 0.042839123867452145, 0.041666717967018485, 0.04133301298134029, 0.04133987915702164, 0.04150186898186803, 0.041693720035254955, 0.04130062088370323, 0.041508896043524146, 0.04143458395265043, 0.04154243599623442, 0.0412688369397074, 0.041415715124458075, 0.04161909595131874, 0.041152097983285785, 0.04147077398374677, 0.041438579093664885, 0.041421727975830436, 0.041442384012043476, 0.04136119596660137, 0.04149415693245828, 0.04155200417153537, 0.041537391021847725, 0.041584026999771595, 0.04113250784575939, 0.04168327711522579, 0.04152855300344527, 0.04144961084239185, 0.04142377316020429, 0.04154161782935262, 0.04143774718977511, 0.041447350988164544, 0.0412863299716264, 0.04151750402525067, 0.04139873292297125, 0.04142392007634044, 0.041856586933135986, 0.04154599690809846, 0.041583730140700936, 0.04149687383323908, 0.041769239120185375, 0.041547339875251055, 0.041508258087560534, 0.04144987813197076, 0.04148425278253853, 0.04157708608545363, 0.046521750977262855, 0.044382879976183176, 0.04752521705813706, 0.04752115299925208, 0.04764556512236595, 0.04736041184514761, 0.047504871152341366, 0.047556449891999364, 0.04771538800559938, 0.04743524105288088, 0.0476206480525434, 0.04738879879005253, 0.0474626780487597, 0.048311853082850575, 0.04750040196813643, 0.04767591110430658, 0.047648083884269, 0.04751425492577255, 0.04771945998072624, 0.04787912708707154, 0.047421676106750965, 0.04752206080593169, 0.04741049907170236, 0.04818478296510875, 0.05058734095655382, 0.04168284707702696, 0.04148537106812, 0.04158152686432004, 0.04150723619386554, 0.041214828845113516, 0.04140967410057783, 0.04148102295584977, 0.041607077000662684, 0.04149445006623864, 0.041443855967372656, 0.04143353085964918, 0.04189068800769746, 0.04153526807203889, 0.04124935413710773, 0.0417097348254174, 0.04159474512562156, 0.04133387701585889, 0.043388577876612544, 0.03952549514360726, 0.041493073804304004, 0.041278665186837316, 0.041383320931345224, 0.041778496000915766, 0.04135558707639575, 0.04136192495934665, 0.04145304090343416, 0.04145101713947952, 0.04156131180934608, 0.04163107112981379, 0.041358639020472765, 0.04145295103080571, 0.04161004489287734, 0.041581928031519055, 0.041805396089330316, 0.04093366279266775, 0.04175403411500156, 0.04136990290135145, 0.04129761504009366, 0.04164537298493087, 0.041573927039280534, 0.041276708943769336, 0.04142283112742007, 0.0414375108666718, 0.041606148006394506, 0.04137411108240485, 0.04152619210071862, 0.04192540980875492, 0.04128674208186567, 0.04119760589674115, 0.04143884219229221, 0.041495804907754064, 0.04156944598071277, 0.04160540411248803, 0.04137075389735401, 0.04245127411559224, 0.04175081988796592, 0.0415621770080179, 0.041457931976765394, 0.041467104107141495, 0.041603193851187825, 0.04137330106459558, 0.04151346790604293, 0.0413682630751282, 0.041499226121231914, 0.04143476695753634, 0.041470829863101244, 0.04143865406513214, 0.041566383093595505, 0.041630866937339306, 0.04140520002692938, 0.041628175880759954, 0.04159245314076543, 0.04151173192076385, 0.041781476931646466, 0.04153639613650739, 0.04137133783660829, 0.04140097415074706, 0.041400704998523, 0.04151283600367606, 0.04149565193802118, 0.04140192503109574, 0.04154662392102182, 0.04141334816813469, 0.04147246084176004, 0.04161963798105717, 0.04135050717741251, 0.041478150989860296, 0.04156939499080181, 0.04143347195349634, 0.041380046866834164, 0.041482027154415846, 0.04159568785689771, 0.04130833502858877, 0.041545080952346325, 0.04152040299959481, 0.04141738102771342, 0.04132373817265034, 0.04138711793348193, 0.04176501394249499, 0.041807496920228004, 0.04139380203559995, 0.041208649054169655, 0.04167220299132168, 0.04174062795937061, 0.041367664001882076, 0.04344712500460446, 0.04909179802052677, 0.047064275946468115, 0.047217490151524544, 0.04729631403461099, 0.04763357387855649, 0.04725087294355035, 0.04725833609700203, 0.046897135907784104, 0.04723209608346224, 0.04749504500068724, 0.04717601998709142, 0.047285485081374645, 0.04737167595885694, 0.04729868704453111, 0.04721455182880163, 0.04731162707321346, 0.04720166209153831, 0.04745823983103037, 0.047350223176181316, 0.04688965599052608, 0.04823447694070637], [0.04753054794855416, 0.0474977248813957, 0.04764445801265538, 0.04735663207247853, 0.047512819059193134, 0.047556638019159436, 0.04772111703641713, 0.047449531964957714, 0.0475953018758446, 0.04732720414176583, 0.04752374393865466, 0.0483205399941653, 0.047495142091065645, 0.047698256792500615, 0.047650273190811276, 0.0475357249379158, 0.04768700688146055, 0.0478843729943037, 0.047405508114025, 0.04753876687027514, 0.047417288180440664, 0.04840671503916383, 0.05010865582153201, 0.04172720108181238, 0.04146426008082926, 0.041590200969949365, 0.041509791975840926, 0.04122390900738537, 0.04139553685672581, 0.041497310157865286, 0.04159870883449912, 0.041491814190521836, 0.0414589059073478, 0.041415804997086525, 0.04190526995807886, 0.041518921963870525, 0.041268546134233475, 0.04169180290773511, 0.041576195042580366, 0.041345030069351196, 0.04339928086847067, 0.03951920312829316, 0.04150423198007047, 0.041270473040640354, 0.04138410580344498, 0.04178247903473675, 0.041343619115650654, 0.04136079899035394, 0.0414610740263015, 0.04144641803577542, 0.04155677300877869, 0.04164211777970195, 0.041352126048877835, 0.04145239805802703, 0.04161883611232042, 0.0415785969235003, 0.0417936029843986, 0.04095541103743017, 0.04175038100220263, 0.04137130989693105, 0.0412855080794543, 0.04164775391109288, 0.041571777081117034, 0.041286137886345387, 0.041431133169680834, 0.04142135102301836, 0.04161022184416652, 0.041378039168193936, 0.04151948494836688, 0.041929825907573104, 0.041286606108769774, 0.041200333973392844, 0.04144414095208049, 0.04148139292374253, 0.04157619597390294, 0.04161097505129874, 0.04135946906171739, 0.042455459013581276, 0.04176183487288654, 0.04155026306398213, 0.041458924068138, 0.041420345893129706, 0.041669916128739715, 0.04135534283705056, 0.04152330313809216, 0.041362290969118476, 0.04149033594876528, 0.04144635400734842, 0.04146062512882054, 0.04143667803145945, 0.041575839975848794, 0.041628035018220544, 0.04140455578453839, 0.04163124109618366, 0.04158964706584811, 0.041509198024868965, 0.04180015088059008, 0.04151905095204711, 0.041387587087228894, 0.04140113806352019, 0.04139192192815244, 0.04153188597410917, 0.0414872660767287, 0.04139410494826734, 0.04153924202546477, 0.04142967099323869, 0.04147368692792952, 0.04161110310815275, 0.04135623900219798, 0.041473014978691936, 0.04156707599759102, 0.041439207969233394, 0.041374227963387966, 0.04147925297729671, 0.04160301107913256, 0.041224522050470114, 0.0416241679340601, 0.04152973694726825, 0.04140736814588308, 0.04132247786037624, 0.041395128006115556, 0.0417994090821594, 0.04179002996534109, 0.04139718203805387, 0.041193469893187284, 0.041700853034853935, 0.041762204142287374, 0.04133843188174069, 0.043452664045616984, 0.049299870850518346, 0.04682615213096142, 0.04718708386644721, 0.04728021612390876, 0.04772235197015107, 0.04724795906804502, 0.047184586990624666, 0.04689077101647854, 0.04727490385994315, 0.04748102696612477, 0.047138314228504896, 0.04732447094283998, 0.04737429297529161, 0.04729301203042269, 0.047214750899001956, 0.04727524402551353, 0.04722667392343283, 0.04743512417189777, 0.04734667087905109, 0.04693292314186692, 0.048289907863363624, 0.04826020495966077, 0.045416716020554304, 0.047129641054198146, 0.04734614910557866, 0.04735149582847953, 0.047094885958358645, 0.04713698010891676, 0.04725322802551091, 0.04740843689069152, 0.04755906807258725, 0.04742064909078181, 0.04730155481956899, 0.04737794818356633, 0.04729057801887393, 0.04737251694314182, 0.047497401013970375, 0.04745505005121231, 0.04722035489976406, 0.04755216999910772, 0.047077874885872006, 0.047506048111245036, 0.04716833494603634, 0.04735684208571911, 0.04728917102329433, 0.04749911092221737, 0.04727168893441558, 0.04725811304524541, 0.04723146301694214, 0.04739690595306456, 0.04731034114956856, 0.04758051293902099, 0.04735751706175506, 0.04730042186565697, 0.047243412118405104, 0.04714268585667014, 0.04750815196894109, 0.04733832902275026, 0.047355194110423326, 0.04728230508044362, 0.047270249808207154, 0.04735740809701383, 0.04724785406142473, 0.04736443981528282, 0.047891002148389816, 0.046992562944069505, 0.04705466493032873, 0.047233783174306154, 0.047697750851511955, 0.04669654602184892, 0.04705740395002067, 0.047346879029646516, 0.04761835606768727, 0.0472834559623152, 0.04727684799581766, 0.047300757141783834, 0.04732091585174203, 0.047184414928779006, 0.04739072918891907, 0.04707299591973424, 0.04715673392638564, 0.04724344098940492, 0.0471649169921875, 0.04731309902854264, 0.04730326705612242, 0.04755954095162451, 0.04710695310495794, 0.04718599491752684, 0.04721546499058604, 0.04731360194273293, 0.04728316701948643, 0.04714613710530102, 0.04722786694765091, 0.047052423004060984, 0.04719337006099522, 0.04715127102099359, 0.04722112091258168, 0.047194830141961575, 0.04711763886734843, 0.047250896925106645, 0.04727910412475467, 0.047333726892247796, 0.04707836406305432, 0.04728130693547428, 0.04725823597982526, 0.04705246910452843, 0.04722044197842479, 0.04722901713103056, 0.04724891483783722, 0.04712490295059979, 0.04735562019050121, 0.04735793196596205, 0.047357640927657485, 0.047195144928991795, 0.04712289106100798, 0.04716581595130265, 0.047351168002933264, 0.04718660609796643, 0.04731259006075561, 0.04721647780388594, 0.04718906502239406, 0.047436005203053355, 0.047030746936798096, 0.04711885401047766, 0.047321982914581895, 0.047203551046550274, 0.047045080922544, 0.04692400502972305, 0.04729971499182284, 0.04792672395706177, 0.04690694413147867, 0.04713427601382136, 0.0473805139772594, 0.04697798890992999, 0.04710217914544046, 0.0472672488540411, 0.047287326073274016, 0.04721512086689472, 0.04712614300660789, 0.04706601402722299, 0.047117333160713315, 0.04751012590713799, 0.04733182699419558, 0.04719307296909392, 0.04722190205939114, 0.04727785289287567, 0.04729517106898129, 0.04743197001516819, 0.04718237998895347, 0.04752494394779205, 0.046994914999231696, 0.0471852021291852, 0.04729487793520093, 0.04702693107537925, 0.04744934686459601, 0.046852219151332974, 0.047167578944936395, 0.04734722594730556, 0.04716742504388094, 0.0470225028693676, 0.047503754030913115, 0.04694816400296986, 0.047352803172543645, 0.04703819681890309, 0.047241940163075924, 0.04710965394042432, 0.04742532502859831, 0.047221909975633025, 0.04696357692591846, 0.04744678293354809, 0.04727215017192066, 0.04707253002561629, 0.04736013384535909, 0.04745662910863757, 0.04727952601388097, 0.0471694590523839, 0.047400675946846604, 0.046993818832561374, 0.04715344309806824, 0.047206721967086196, 0.047021530102938414, 0.04733775486238301, 0.047141568968072534, 0.04745826008729637, 0.047411506064236164, 0.04718855884857476, 0.047196562169119716, 0.047470557037740946, 0.047358448849990964, 0.04710770910605788, 0.04724647291004658, 0.046590263955295086, 0.047982969088479877, 0.04726323997601867, 0.04741626791656017, 0.04734644619747996, 0.04733110195957124, 0.04723568796180189, 0.04711329797282815, 0.047300216974690557, 0.04738501808606088, 0.04747849493287504, 0.04724319907836616, 0.047904630890116096, 0.04667408112436533, 0.04742976790294051, 0.046999687096104026, 0.0472445038612932, 0.04731726413592696, 0.047400763956829906, 0.047121143899858, 0.04723675502464175, 0.04743088595569134, 0.04733925499022007, 0.04706779308617115, 0.04727081302553415, 0.048499759985134006, 0.04613691195845604, 0.04682947997935116, 0.04719968908466399, 0.04722471907734871, 0.04740848485380411, 0.047355791088193655, 0.04699723492376506, 0.047084887977689505, 0.04736973810940981, 0.04720882489345968, 0.04713040217757225, 0.047468203818425536, 0.047379108145833015, 0.047203032998368144, 0.04694389086216688, 0.04709283704869449, 0.04745503910817206, 0.047243993962183595, 0.04722425900399685, 0.04773987689986825, 0.04716471000574529, 0.04705815599299967, 0.0471772460732609, 0.048345784889534116, 0.04721235018223524, 0.047105352859944105, 0.04711339296773076, 0.04741028603166342, 0.04750797804445028, 0.047100130934268236, 0.04696281999349594, 0.04704035515896976, 0.04763582185842097, 0.04723483603447676, 0.047115113120526075, 0.047219313913956285, 0.047304391860961914, 0.04750280920416117, 0.04689875501208007, 0.0469183640088886, 0.0472663298714906, 0.04722353513352573, 0.04706075298599899, 0.0474200698081404, 0.0471340490039438, 0.04694618517532945, 0.04722817800939083, 0.04738335683941841, 0.04722246411256492, 0.04850280284881592, 0.04627411114051938, 0.04745655204169452, 0.04729548585601151, 0.047210540156811476, 0.04711219295859337, 0.04719464899972081, 0.047244159039109945, 0.04731036187149584, 0.04694869602099061, 0.047369020991027355, 0.04714223695918918, 0.04695604811422527, 0.04693717788904905, 0.04730373015627265, 0.04730104492045939, 0.04722392396070063, 0.047035550931468606, 0.04743753909133375, 0.04724110593087971, 0.047227174043655396, 0.04726204904727638, 0.04711308889091015, 0.04739591898396611, 0.04719818499870598, 0.04725990607403219, 0.04706332995556295, 0.047359707998111844, 0.04731459799222648, 0.047144509153440595, 0.04727510386146605, 0.047565824119374156, 0.04725007084198296, 0.04704662808217108, 0.047323263017460704, 0.047187383053824306, 0.04733751900494099, 0.04713931796140969, 0.047212497098371387, 0.04720674781128764, 0.04789160517975688, 0.047006733948364854, 0.047052758978679776, 0.047548078931868076, 0.04703707410953939, 0.047106164041906595, 0.04749025800265372, 0.04737067688256502, 0.047040688106790185, 0.04703228990547359, 0.04736354993656278, 0.04700097208842635, 0.047225679038092494, 0.04773350292816758, 0.047499984968453646, 0.04729294707067311, 0.04732757597230375, 0.04717192496173084, 0.04743661591783166, 0.04739693715237081, 0.04721924406476319, 0.047249282943084836, 0.047604785999283195, 0.04707869701087475, 0.04721839586272836, 0.047836288111284375, 0.04709877306595445, 0.04741985397413373, 0.04707042290829122, 0.04752529691904783, 0.047163133043795824, 0.04752760799601674, 0.047147843055427074, 0.047370319021865726, 0.04719156608916819, 0.0474861147813499, 0.04705461114645004, 0.047219709027558565, 0.047131708823144436, 0.048672733129933476, 0.047279272926971316, 0.047157366992905736, 0.04724658909253776, 0.047280011931434274, 0.047290723072364926, 0.0474408739246428, 0.04741742997430265, 0.04693074314855039, 0.04752065194770694, 0.047255693934857845, 0.04761280911043286, 0.047187224961817265, 0.04723733686842024, 0.04754671803675592, 0.04741731798276305, 0.04732867702841759, 0.047066007973626256, 0.0475210330914706, 0.04729197290726006, 0.04752239305526018, 0.04700848599895835, 0.047120077069848776, 0.04752011806704104, 0.04769550287164748, 0.04721849295310676, 0.047315509989857674, 0.04731028014793992, 0.047743425937369466, 0.04735723999328911, 0.04743574489839375, 0.04732259805314243, 0.047526118112728, 0.047291856026276946, 0.04712165193632245, 0.0484149178955704], [0.046535291941836476, 0.047239406034350395, 0.047342324862256646, 0.04747278802096844, 0.047296726144850254, 0.04734077397733927, 0.04692519991658628, 0.047252672957256436, 0.04750173701904714, 0.04710284806787968, 0.047356694005429745, 0.04736662586219609, 0.04730156413279474, 0.047215475933626294, 0.047245630994439125, 0.04720024508424103, 0.04745914391241968, 0.047346703009679914, 0.04695940110832453, 0.04822032083757222, 0.04853061819449067, 0.045423376839607954, 0.047136161010712385, 0.047349027125164866, 0.04731552489101887, 0.04709606501273811, 0.047136123990640044, 0.047291019931435585, 0.04736887104809284, 0.04757374292239547, 0.04741776315495372, 0.04721682402305305, 0.04740081890486181, 0.04731591301970184, 0.04733442794531584, 0.04749700101092458, 0.0475267490837723, 0.04723189398646355, 0.04750165902078152, 0.04712613788433373, 0.04748767800629139, 0.047152798157185316, 0.04737852583639324, 0.04727914300747216, 0.047516924096271396, 0.04726358805783093, 0.04726154496893287, 0.04725706786848605, 0.04740076698362827, 0.04730920307338238, 0.04758197604678571, 0.04735161690041423, 0.04729354497976601, 0.04724810505285859, 0.047139055095613, 0.04750494193285704, 0.04733186005614698, 0.0473264588508755, 0.04732194007374346, 0.0472597770858556, 0.04736327286809683, 0.047250768169760704, 0.04736740700900555, 0.04778934293426573, 0.046994078904390335, 0.04710900713689625, 0.04722946789115667, 0.04762192093767226, 0.04680704604834318, 0.04706417513079941, 0.047351480927318335, 0.047616139985620975, 0.04728621011599898, 0.04728245781734586, 0.0472660050727427, 0.04735113889910281, 0.04719209298491478, 0.04738742019981146, 0.047070409869775176, 0.04715579212643206, 0.047247543931007385, 0.04719749605283141, 0.047252733958885074, 0.0473267009947449, 0.047556486912071705, 0.04711050703190267, 0.04718430992215872, 0.047214090125635266, 0.04731444991193712, 0.04728820803575218, 0.0471419389359653, 0.047225154004991055, 0.04705354501493275, 0.047120776027441025, 0.047216649167239666, 0.047224521869793534, 0.04720078990794718, 0.04711422906257212, 0.04725678195245564, 0.0472710351459682, 0.04733773693442345, 0.04707633797079325, 0.047280725091695786, 0.04724287986755371, 0.04706722404807806, 0.04722086200490594, 0.04721810598857701, 0.047255323035642505, 0.047127772122621536, 0.047356595983728766, 0.0473571359179914, 0.047352683963254094, 0.047198757994920015, 0.04712636303156614, 0.04716312210075557, 0.047326865838840604, 0.047208320116624236, 0.047316354932263494, 0.04721188498660922, 0.047187109012156725, 0.047421013936400414, 0.04704945208504796, 0.047122566029429436, 0.047315201023593545, 0.047208623960614204, 0.047039211029186845, 0.046928685856983066, 0.04729660018347204, 0.047785192960873246, 0.046991518000140786, 0.04749546083621681, 0.04696631012484431, 0.04690942703746259, 0.047197778010740876, 0.047312495997175574, 0.04726153798401356, 0.0472459529992193, 0.04705410194583237, 0.0470484159886837, 0.04717313405126333, 0.04746888088993728, 0.04737291298806667, 0.047171532176434994, 0.047244729939848185, 0.04720168001949787, 0.04740349785424769, 0.04740430717356503, 0.04721861588768661, 0.047463369090110064, 0.047016110038384795, 0.04721893183887005, 0.04725388716906309, 0.04701648489572108, 0.04747835802845657, 0.046794022899121046, 0.04722814401611686, 0.047265814151614904, 0.04717145790345967, 0.047029390931129456, 0.04749835515394807, 0.047040923964232206, 0.047313940012827516, 0.047067096922546625, 0.04717230307869613, 0.0471497259568423, 0.047416751040145755, 0.04718086286447942, 0.047054789029061794, 0.047423550160601735, 0.047207766911014915, 0.047057891031727195, 0.04746419098228216, 0.047359772957861423, 0.04736592108383775, 0.047170701902359724, 0.04738659807480872, 0.04697194485925138, 0.04717125720344484, 0.04712134785950184, 0.04710830212570727, 0.047251927899196744, 0.04725420498289168, 0.0472912210971117, 0.04756734985858202, 0.047196866013109684, 0.04716907907277346, 0.04738605488091707, 0.047344743041321635, 0.04720758204348385, 0.047176463063806295, 0.04662197385914624, 0.04794846707955003, 0.04733530106022954, 0.04732709191739559, 0.04739032592624426, 0.04734461009502411, 0.04713544901460409, 0.04718662006780505, 0.04733604588545859, 0.04736371408216655, 0.047510732896625996, 0.04723434615880251, 0.04793689586222172, 0.04664006410166621, 0.047414171975106, 0.047059498028829694, 0.0472309710457921, 0.04763664095662534, 0.047035304829478264, 0.04717083415016532, 0.04723153985105455, 0.04741962905973196, 0.047268948052078485, 0.04705481603741646, 0.047353988979011774, 0.04838391300290823, 0.04612899501807988, 0.046991429990157485, 0.04720005998387933, 0.04722631792537868, 0.047404187032952905, 0.04736244399100542, 0.04699509800411761, 0.04708718997426331, 0.0473689129576087, 0.04721300513483584, 0.047124838922172785, 0.04747173492796719, 0.04737544618546963, 0.04717408888973296, 0.04694942710921168, 0.047382254851982, 0.0471745349932462, 0.047248111106455326, 0.04721831390634179, 0.047674712957814336, 0.04720519413240254, 0.04698000196367502, 0.04725749394856393, 0.04827085696160793, 0.047293839044868946, 0.047109329141676426, 0.047067012870684266, 0.04744424601085484, 0.047504483023658395, 0.04705715202726424, 0.04701231396757066, 0.047127884114161134, 0.04748354689218104, 0.04724505404010415, 0.047126956982538104, 0.04722554003819823, 0.047343866899609566, 0.04721039393916726, 0.0471751990262419, 0.04693743702955544, 0.04728564596734941, 0.04723036615177989, 0.04701208486221731, 0.04746452299878001, 0.04713240917772055, 0.0469524918589741, 0.04722854704596102, 0.047378797084093094, 0.047220294829458, 0.04832967207767069, 0.046411680057644844, 0.04745101183652878, 0.04728198400698602, 0.04722650209441781, 0.04703146405518055, 0.04718798492103815, 0.047270009061321616, 0.04732758202590048, 0.047004712047055364, 0.04730892786756158, 0.04704356496222317, 0.047114284010604024, 0.046944690169766545, 0.047319156816229224, 0.04729551915079355, 0.047230969881638885, 0.04704255098477006, 0.047432373045012355, 0.04723649891093373, 0.04723329609259963, 0.04725047992542386, 0.047125434037297964, 0.04738664999604225, 0.047203928930684924, 0.04725100821815431, 0.047070077853277326, 0.047358473064377904, 0.047319079050794244, 0.0471407410223037, 0.04727400792762637, 0.04754965798929334, 0.04725999501533806, 0.04705463000573218, 0.04729329189285636, 0.04721157415769994, 0.047340337885543704, 0.047148545971140265, 0.047204202972352505, 0.04720745305530727, 0.04781544394791126, 0.04705092706717551, 0.04701684904284775, 0.04751524003222585, 0.04708537389524281, 0.047056514071300626, 0.04746699100360274, 0.047324680956080556, 0.04718473507091403, 0.04700107593089342, 0.04739580792374909, 0.04699299018830061, 0.04716534586623311, 0.04772816598415375, 0.047461831010878086, 0.04739844403229654, 0.04732180992141366, 0.04704347299411893, 0.04749181098304689, 0.047458179062232375, 0.0472118491306901, 0.047270797891542315, 0.04753549303859472, 0.04707582900300622, 0.04724544892087579, 0.0477776990737766, 0.047202405985444784, 0.04737809603102505, 0.04708394990302622, 0.047519254963845015, 0.04764756700024009, 0.0470538551453501, 0.04711415385827422, 0.047402982134371996, 0.04711295408196747, 0.0475838128477335, 0.04700807202607393, 0.0472524210345, 0.04715234390459955, 0.048572010127827525, 0.04727680701762438, 0.047235620906576514, 0.047183119924739, 0.0473640242125839, 0.04727945080958307, 0.04735285299830139, 0.047457071021199226, 0.04697844898328185, 0.04743849905207753, 0.047252688091248274, 0.04763157293200493, 0.047204132890328765, 0.04730611899867654, 0.047357347095385194, 0.04758581495843828, 0.04723438201472163, 0.047169107012450695, 0.04743987205438316, 0.04733504308387637, 0.047470252960920334, 0.04702516691759229, 0.04720155009999871, 0.04743587481789291, 0.04773255414329469, 0.04725824994966388, 0.047171940095722675, 0.047402273863554, 0.047759203938767314, 0.047390023013576865, 0.047324445098638535, 0.04743660590611398, 0.047499031061306596, 0.04719587811268866, 0.04724452388472855, 0.04839222808368504, 0.04860565485432744, 0.03946240711957216, 0.041349146980792284, 0.041394941974431276, 0.041555653093382716, 0.041315812850371, 0.04124938300810754, 0.04154042503796518, 0.041174934012815356, 0.041307582054287195, 0.04144023102708161, 0.04137326986528933, 0.041467539966106415, 0.041334484005346894, 0.04152114619500935, 0.04138099099509418, 0.04166337801143527, 0.04123168089427054, 0.041483269073069096, 0.04142037685960531, 0.04161425307393074, 0.041238161036744714, 0.04130988987162709, 0.04151165904477239, 0.041328963125124574, 0.04173569194972515, 0.041253621922805905, 0.04135365900583565, 0.04153025895357132, 0.041414979146793485, 0.04128107498399913, 0.0412684918846935, 0.04139556596055627, 0.04163587698712945, 0.041585544124245644, 0.04139435989782214, 0.04151578014716506, 0.04137695604003966, 0.04132170486263931, 0.04140689712949097, 0.04160612681880593, 0.043002986116334796, 0.040044009918347, 0.04140646802261472, 0.04170624609105289, 0.04141151998192072, 0.04161645285785198, 0.041302829049527645, 0.04147984413430095, 0.041362107964232564, 0.041410336969420314, 0.04114800109528005, 0.042207375867292285, 0.041002518963068724, 0.04130924795754254, 0.04151183506473899, 0.04143301397562027, 0.04149000905454159, 0.04134576697833836, 0.04137312201783061, 0.04164322605356574, 0.04129466600716114, 0.041416445979848504, 0.04155980795621872, 0.04131515906192362, 0.041557413060218096, 0.041254254989326, 0.041359431808814406, 0.04136632801964879, 0.04159221099689603, 0.04139448213391006, 0.04133374895900488, 0.04146768408827484, 0.04158731200732291, 0.04182708100415766, 0.04153746389783919, 0.04145059688016772, 0.04161493107676506, 0.0411576540209353, 0.04130916902795434, 0.04147869301959872, 0.041360808070749044, 0.041429522912949324, 0.04141191905364394, 0.041506454814225435, 0.04143697419203818, 0.04156650183722377, 0.04143273318186402, 0.04143322282470763, 0.041470129042863846, 0.04142358712852001, 0.041295032016932964, 0.041459440952166915, 0.041426577838137746, 0.04150174115784466, 0.04137773299589753, 0.04149558302015066, 0.041404675925150514, 0.041569421999156475, 0.04140714299865067, 0.04136600811034441, 0.04132867185398936, 0.04138156399130821, 0.041388072073459625, 0.041479764971882105, 0.04145421809516847, 0.04204449290409684, 0.040971650974825025, 0.04136538109742105, 0.04139705887064338, 0.04143153317272663, 0.041384005919098854, 0.04140986688435078, 0.041463463101536036, 0.04135369905270636, 0.041366444900631905, 0.04142500110901892, 0.04144630581140518, 0.04139582114294171, 0.04118815204128623, 0.0416966259945184, 0.041480511892586946, 0.04138185805641115, 0.04141078586690128, 0.041391039034351707, 0.04139204206876457, 0.041965384036302567, 0.04100870992988348, 0.04149303212761879, 0.042115780990570784], [0.04719238309189677, 0.04741320898756385, 0.04737291485071182, 0.047398122027516365, 0.04726288700476289, 0.046996593941003084, 0.047303487081080675, 0.04748832690529525, 0.04711035522632301, 0.0473738107830286, 0.04735991405323148, 0.047288766130805016, 0.047229643911123276, 0.047239264007657766, 0.04722053091973066, 0.04742749803699553, 0.04737261403352022, 0.04691275698132813, 0.047987537924200296, 0.048818112118169665, 0.045439962996169925, 0.04713397100567818, 0.047357066068798304, 0.04733867687173188, 0.04709160001948476, 0.04714403301477432, 0.04726646398194134, 0.04739080206491053, 0.0474926820024848, 0.047340380027890205, 0.04732434800826013, 0.04727566381916404, 0.047437502071261406, 0.047288839938119054, 0.047477644169703126, 0.04749482101760805, 0.04731020983308554, 0.047387706115841866, 0.04725307389162481, 0.04754489497281611, 0.04716567508876324, 0.047351537039503455, 0.047302938997745514, 0.047506583854556084, 0.04726345813833177, 0.047244474990293384, 0.04727047495543957, 0.047397586051374674, 0.04730974603444338, 0.047574474941939116, 0.04735934897325933, 0.04726513708010316, 0.047259357990697026, 0.047096863854676485, 0.04748413502238691, 0.0473797891754657, 0.04731044895015657, 0.04729776387102902, 0.04728645319119096, 0.04736420582048595, 0.047242439119145274, 0.04735894803889096, 0.04766658088192344, 0.04702918394468725, 0.047142621129751205, 0.04720895295031369, 0.047314875992015004, 0.047185305040329695, 0.047095505986362696, 0.047356098890304565, 0.047613496193662286, 0.04728706879541278, 0.0472875430714339, 0.04728685901500285, 0.04733132501132786, 0.04719592002220452, 0.0473922120872885, 0.04706170386634767, 0.047159818932414055, 0.04724698816426098, 0.04723561485297978, 0.04722933913581073, 0.04731207084842026, 0.04754513199441135, 0.04711834108456969, 0.047187773976475, 0.04719149903394282, 0.047317075077444315, 0.04729320388287306, 0.047135632019490004, 0.04723149701021612, 0.04705807496793568, 0.04712791903875768, 0.047160885063931346, 0.047261670930311084, 0.04720180993899703, 0.04711516201496124, 0.04726534499786794, 0.04727335297502577, 0.0473321950994432, 0.047074747970327735, 0.04727463494054973, 0.04722308018244803, 0.04708354896865785, 0.04722634586505592, 0.047211962984874845, 0.04725977801717818, 0.0471341370139271, 0.04734128201380372, 0.04736061906442046, 0.04735791403800249, 0.047198079992085695, 0.04713201499544084, 0.047163309063762426, 0.04732318897731602, 0.047212988836690784, 0.04731664899736643, 0.04720891313627362, 0.0471869339235127, 0.04738856293261051, 0.04707404412329197, 0.04712695302441716, 0.047322270926088095, 0.047205748967826366, 0.047023453982546926, 0.046939382096752524, 0.04729957296513021, 0.047623963095247746, 0.04709387989714742, 0.04717362695373595, 0.04720377712510526, 0.04697134695015848, 0.047153804916888475, 0.04732696106657386, 0.04729386395774782, 0.04728201893158257, 0.0470305890776217, 0.047041461104527116, 0.04719015094451606, 0.047360562020912766, 0.047422162955626845, 0.047168551944196224, 0.047298804158344865, 0.047265961999073625, 0.04730039881542325, 0.04741264600306749, 0.04726205603219569, 0.047375635942444205, 0.04708965215831995, 0.0472363259177655, 0.04717487096786499, 0.04706699214875698, 0.0474741249345243, 0.046794265042990446, 0.047193896025419235, 0.047267779940739274, 0.0472610779106617, 0.04703803802840412, 0.047412856947630644, 0.04708660300821066, 0.04731056513264775, 0.047006150940433145, 0.04720098897814751, 0.047221943968907, 0.04736144398339093, 0.047095142072066665, 0.04715158394537866, 0.047377664130181074, 0.047188714845106006, 0.04715798213146627, 0.04742076899856329, 0.04729457292705774, 0.04738232702948153, 0.04718602495267987, 0.04738023807294667, 0.04699231591075659, 0.04711221600882709, 0.04713440197519958, 0.04713459499180317, 0.04732194100506604, 0.04723567911423743, 0.04727151687256992, 0.04753627302125096, 0.047249444061890244, 0.047184017021209, 0.047285781940445304, 0.04733360605314374, 0.04723760602064431, 0.04716508695855737, 0.04667472094297409, 0.047891882015392184, 0.04736673505976796, 0.04732521902769804, 0.04734459798783064, 0.047334411880001426, 0.04722032300196588, 0.04720639204606414, 0.04730780306272209, 0.04732575896196067, 0.04752613394521177, 0.04721552296541631, 0.047915980219841, 0.04670964297838509, 0.047390537802129984, 0.04710467113181949, 0.047224324895069, 0.0474154781550169, 0.04724676697514951, 0.047167833894491196, 0.04715579911135137, 0.0475111089181155, 0.047260696068406105, 0.04710260289721191, 0.04722582409158349, 0.04734594188630581, 0.047656489070504904, 0.04660281911492348, 0.047212314791977406, 0.04723544814623892, 0.04738616198301315, 0.04738108208402991, 0.04699481278657913, 0.04708293313160539, 0.047381324926391244, 0.04718690412119031, 0.04714868497103453, 0.04747080197557807, 0.04737547389231622, 0.047163638984784484, 0.04696141020394862, 0.047219199826940894, 0.04731595399789512, 0.04722372512333095, 0.04723823186941445, 0.04752014297991991, 0.047301934100687504, 0.046927208080887794, 0.04728272696956992, 0.04827026394195855, 0.04730103793554008, 0.04714190005324781, 0.04704386298544705, 0.047384548000991344, 0.047499798936769366, 0.04713047714903951, 0.04706875188276172, 0.04718270618468523, 0.047293861862272024, 0.04732796107418835, 0.04713445506058633, 0.047220950946211815, 0.04731986788101494, 0.047202775022014976, 0.04719824413768947, 0.04698290582746267, 0.047303046099841595, 0.04723726003430784, 0.04705537483096123, 0.04742039809934795, 0.04714363510720432, 0.046941587002947927, 0.04723887983709574, 0.04736667894758284, 0.047227042028680444, 0.0481089330278337, 0.046575168147683144, 0.0473942079115659, 0.04726850497536361, 0.04723127791658044, 0.04704342316836119, 0.04717818694189191, 0.047269460977986455, 0.04735701298341155, 0.047037521144375205, 0.04720043484121561, 0.04715517698787153, 0.04712406708858907, 0.04698279802687466, 0.04730177600868046, 0.047311357920989394, 0.0472373440861702, 0.04704178683459759, 0.04742801305837929, 0.04723743093200028, 0.04723735013976693, 0.04724597791209817, 0.047127678990364075, 0.047392564127221704, 0.04720124299637973, 0.047217520885169506, 0.047095743007957935, 0.04736616602167487, 0.047311567002907395, 0.047148199984803796, 0.04726259293965995, 0.04752623103559017, 0.04726909310556948, 0.04706848901696503, 0.04728475003503263, 0.047225645976141095, 0.04733699397183955, 0.047151496866717935, 0.04720268398523331, 0.047209621174260974, 0.04766537086106837, 0.04711200902238488, 0.0470722911413759, 0.04737380798906088, 0.047135514905676246, 0.047153038904070854, 0.04735596617683768, 0.04742328589782119, 0.04713636310771108, 0.047032313887029886, 0.04736090009100735, 0.04704281990416348, 0.04718027403578162, 0.0476314970292151, 0.04745056387037039, 0.04744953103363514, 0.0473047811537981, 0.04713304480537772, 0.047380320029333234, 0.04745919397100806, 0.04730206215754151, 0.047199460910633206, 0.047446625074371696, 0.04723753803409636, 0.04723410401493311, 0.04771123803220689, 0.04724205192178488, 0.04724471108056605, 0.04718663892708719, 0.04749837704002857, 0.047307927859947085, 0.04732760298065841, 0.04724979610182345, 0.047313730930909514, 0.04718539002351463, 0.04753669910132885, 0.046990899834781885, 0.04725764901377261, 0.04721918608993292, 0.0485188199672848, 0.04720176197588444, 0.047202146146446466, 0.04730939888395369, 0.0473578879609704, 0.04728341498412192, 0.04733848501928151, 0.04739019810222089, 0.04699122300371528, 0.04754132404923439, 0.047298781806603074, 0.047466452000662684, 0.04731980012729764, 0.04731637402437627, 0.047349862987175584, 0.047503497917205095, 0.047261066967621446, 0.04722982714883983, 0.04739523702301085, 0.04739072499796748, 0.047357946867123246, 0.04709527804516256, 0.04719881690107286, 0.04745146608911455, 0.0477286740206182, 0.04726177407428622, 0.04721307079307735, 0.04727820912376046, 0.04771483992226422, 0.047529414063319564, 0.04717402392998338, 0.04758434812538326, 0.047498853877186775, 0.04714828613214195, 0.04728821897879243, 0.04833310283720493, 0.04857306112535298, 0.03952303109690547, 0.04132064478471875, 0.04149527009576559, 0.041497339960187674, 0.04131569806486368, 0.04130444908514619, 0.04147186083719134, 0.04123942297883332, 0.04131521703675389, 0.04142359714023769, 0.04136221087537706, 0.04143789294175804, 0.041341192089021206, 0.041522273095324636, 0.041446377988904715, 0.04165395791642368, 0.04123321291990578, 0.04148542508482933, 0.04141743807122111, 0.04150157794356346, 0.041361986892297864, 0.04125349712558091, 0.04145411495119333, 0.041414191015064716, 0.04160954710096121, 0.04131278884597123, 0.041421439964324236, 0.04152065305970609, 0.04141608695499599, 0.04132860200479627, 0.04127540695481002, 0.041405148105695844, 0.0416399110108614, 0.04158688010647893, 0.04139440390281379, 0.04151993501000106, 0.04137303004972637, 0.04132749582640827, 0.04140364401973784, 0.04160131816752255, 0.042834467953071, 0.0401880859863013, 0.041409329045563936, 0.04171638889238238, 0.04141115304082632, 0.041593381902202964, 0.04131901916116476, 0.04148477083072066, 0.041363588999956846, 0.04141475213691592, 0.04115404188632965, 0.04207907314412296, 0.04110999498516321, 0.041321475990116596, 0.041501855943351984, 0.04143869993276894, 0.041483134031295776, 0.0413498820271343, 0.0413763620890677, 0.04162192181684077, 0.0413200871553272, 0.04140463192015886, 0.041541529120877385, 0.04133867099881172, 0.04150010389275849, 0.04128543892875314, 0.0413853891659528, 0.04135620081797242, 0.04157692822627723, 0.04137871088460088, 0.04136178898625076, 0.041479522129520774, 0.041564048966392875, 0.041852748952805996, 0.04152630898170173, 0.04144545388408005, 0.04158942401409149, 0.04119005799293518, 0.04131463309749961, 0.041481063002720475, 0.041362540097907186, 0.041430812794715166, 0.041409380035474896, 0.041518038138747215, 0.04141422803513706, 0.04158144583925605, 0.041432393016293645, 0.04143714904785156, 0.04146453202702105, 0.04142040200531483, 0.0413025610614568, 0.04146837699227035, 0.041422316804528236, 0.04149735905230045, 0.04137019696645439, 0.04148176801390946, 0.04142517107538879, 0.04157318896614015, 0.04140697605907917, 0.04136238503269851, 0.041329857893288136, 0.04138473700731993, 0.0413856019731611, 0.04148010700009763, 0.04145060502924025, 0.04178489511832595, 0.04120486602187157, 0.04136018082499504, 0.041405884083360434, 0.04141978290863335, 0.04139876295812428, 0.04141745320521295, 0.0414587389677763, 0.041337489848956466, 0.04138019704259932, 0.04142895108088851, 0.041438356041908264, 0.04139494081027806, 0.04133641021326184, 0.04155514086596668, 0.04149226099252701, 0.041375430999323726, 0.0414098440669477, 0.04139451589435339, 0.041392686078324914, 0.04187026806175709, 0.04106545099057257, 0.04149818606674671, 0.0420173869933933, 0.04565830482169986]], "generated_texts": ["\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-", "\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_REQUIREMENTS\ufffd_", "\ufffd_BREAK\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL\ufffd_CANCEL", "\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(\ufffd(", "\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.", "\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_", "\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_", "\u7684\u65b9\u9488\u653f\u7b56\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.", "\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_", "\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_", "\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_PATRICK\ufffd_", "\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.", "\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_PATHTYPE\ufffd_", "\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-", "\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-\ufffd-", "\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_id\ufffd_", "\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_", "(\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_BREAK\ufffd_B", "\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.", "\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_TYPEDEF\ufffd_", "\u7684\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd", "\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_\ufffd_", "ed\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd", "\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n", "\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_TICKET\ufffd_", "\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName", "\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n\ufffd\n", "\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_PARENTHESIS\ufffd_", "\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.", ".getElementById('output').innerHTML = 'Error: ' + e.message;\n    }\n  </script>\n</body>\n</html><|close|>response<|sep|><|close|>message<|sep|><|end_of_msg|> -1 & 0) {\n      document.getElementById('output').innerHTML = 'Error: ' + e.message;\n    }\n  </script>\n</body>\n</html><|close|>response<|sep|><|close|>message<|sep|><|end_of_msg|> -1 & 0) {\n      document.getElementById('output').innerHTML = 'Error: ' + e.message;\n    }\n  </script>\n</body>\n</html><|close|>response<|sep|><|close|>message<|sep|><|end_of_msg|> -1 & 0) {\n      document.getElementById('output').innerHTML = 'Error: ' + e.message;\n    }\n  </script>\n</body>\n</html><|close|>response<|sep|><|close|>message<|sep|><|end_of_msg|> -1 & 0) {\n      document.getElementById('output').innerHTML = 'Error: ' + e.message;\n    }\n  </script>\n</body>\n</html><|close|>response<|sep|><|close|>message<|sep|><|end_of_msg|> -1 & 0) {\n      document.getElementById('output').innerHTML = 'Error: ' + e.message;\n    }\n  </script>\n</body>\n</html><|close|>response<|sep|><|close|>message<|sep|><|end_of_msg|> -1 & 0) {\n      document.getElementById('output').innerHTML = 'Error: ' + e.message;\n    }\n  </script>\n</body>\n</html><|close|>response<|sep|><|close|>message<|sep|><|end_of_msg|> -1 & 0) {\n      document.getElementById('output').innerHTML = 'Error: ' + e.message;\n    }\n  </script>\n</body>\n</html><|close|>response<|sep|><|close|>message<|sep|><|end_of_msg|> -1 & 0) {\n      document.getElementById('output').innerHTML = 'Error: ' + e.message;\n    }\n  </script>\n</body>\n</html><|close|>response<|sep|><|close|>message<|sep|><|end_of_msg|> -1 & 0) {\n      document.getElementById('output').innerHTML = 'Error: ' + e.message;\n    }\n  </script>\n</body>\n</html><|close|>response<|sep|><|close|>message<|sep|><|end_of_msg|> -1 & 0) {\n      document.getElementById('output').innerHTML = 'Error: ' + e.message;\n    }\n  </script>\n</body>\n</html><|close|>response<|sep|><|close|>message<|sep|><|end_of_msg|> -1 &", "\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired", "\u7684\u58eb\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_PARENT\ufffd_", "\n\n[README.md]\n\n# 2025-03-24\n2025-03-24 \u5b66\u4e60\u8bb0\u5f55\n and 2025-03-24 \u5b66\u4e60\u8bb0\u5f55\n2025-03-24 \u5b66\u4e60\u8bb0\u5f55\n2025-03-24 \u5b66\u4e60\u8bb0\u5f55\n2025-03-24 \u5b66\u4e60\u8bb0\u5f55\n2025-03-24 \u5b66\u4e60\u8bb0\u5f55\n2025-03-24 \u5b66\u4e60\u8bb0\u5f55\n2025-03-24 \u5b66\u4e60\u8bb0\u5f55\n2025-03-24 \u5b66\u4e60\u8bb0\u5f55\n2025-03-24 \u5b66\u4e60\u8bb0\u5f55\n2025-03-24 \u5b66\u4e60\u8bb0\u5f55\n2025-03-24 \u5b66\u4e60\u8bb0\u5f55\n2025-03-24 \u5b66\u4e60\u8bb0\u5f55\n2025-03-24 \u5b66\u4e60\u8bb0\u5f55\n2025-03-24 \u5b66\u4e60\u8bb0\u5f55\n2025-03-24 \u5b66\u4e60\u8bb0\u5f55\n2025-03-24 \u5b66\u4e60\u8bb0\u5f55\n2025-03-24 \u5b66\u4e60\u8bb0\u5f55\n2025-03-24 \u5b66\u4e60\u8bb0\u5f55\n2025-03-24 \u5b66\u4e60\u8bb0\u5f55\n2025-03-24 \u5b66\u4e60\u8bb0\u5f55\n2025-03-24 \u5b66\u4e60\u8bb0\u5f55\n2025-03-24 \u5b66\u4e60\u8bb0\u5f55\n2025-03-24 \u5b66\u4e60\u8bb0\u5f55\n2025-03-24 \u5b66\u4e60\u8bb0\u5f55\n2025-03-24 \u5b66\u4e60\u8bb0\u5f55\n2025-03-24 \u5b66\u4e60\u8bb0\u5f55\n2025-03-24 \u5b66\u4e60\u8bb0\u5f55\n2025-03-24 \u5b66\u4e60\u8bb0\u5f55\n2025-03-24 \u5b66\u4e60\u8bb0\u5f55\n2025-03-24 \u5b66\u4e60\u8bb0\u5f55\n2025-03-24 \u5b66\u4e60\u8bb0\u5f55\n2025-03-24 \u5b66\u4e60\u8bb0\u5f55\n2025-03-24 \u5b66\u4e60\u8bb0\u5f55\n2025-03-24 \u5b66\u4e60\u8bb0\u5f55\n2025-03-24 \u5b66\u4e60\u8bb0\u5f55\n2025-03-24 \u5b66\u4e60\u8bb0\u5f55\n2025-03-24 \u5b66\u4e60\u8bb0\u5f55\n2025-03-24 \u5b66\u4e60\u8bb0\u5f55\n2025-03-24 \u5b66\u4e60\u8bb0\u5f55\n2025-03-24 \u5b66\u4e60\u8bb0\u5f55\n2025-03-24 \u5b66\u4e60\u8bb0\u5f55\n2025-03-24 \u5b66\u4e60\u8bb0\u5f55\n2025-03-24 \u5b66\u4e60\u8bb0\u5f55\n2025-03-24 \u5b66\u4e60\u8bb0\u5f55\n2025-03-24 \u5b66\u4e60\u8bb0\u5f55\n2025-03-24 \u5b66\u4e60\u8bb0\u5f55\n2025-03-24 \u5b66\u4e60\u8bb0\u5f55\n2025-03-24 \u5b66\u4e60\u8bb0\u5f55\n2025-03-", "\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN", "\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/usr\ufffd/", "\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN\ufffd_PATTERN", "\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/\ufffd/", "\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName\ufffd.getName", "\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.", "\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd.\ufffd."], "errors": ["", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", ""]}
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-c8-20260827-1340/bench/decode_16k_512_c8.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-c8-20260827-1340/bench/decode_16k_512_c8.log
new file mode 100644
index 0000000..8f21770
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-c8-20260827-1340/bench/decode_16k_512_c8.log
@@ -0,0 +1,110 @@
+/usr/local/lib/python3.12/dist-packages/torchao/quantization/quant_api.py:1731: SyntaxWarning: invalid escape sequence '\.'
+  """Configuration class for applying different quantization configs to modules or parameters based on their fully qualified names (FQNs).
+/sgl-workspace/sglang/python/sglang/kernels/ops/attention/fla/utils.py:223: UserWarning: Triton is not supported on current platform, roll back to CPU.
+  warnings.warn(
+/sgl-workspace/sglang/python/sglang/srt/layers/quantization/awq/awq.py:52: UserWarning: Only CUDA, HIP and XPU support AWQ currently.
+  warnings.warn(f"Only CUDA, HIP and XPU support AWQ currently.")
+/sgl-workspace/sglang/python/sglang/srt/layers/quantization/gguf.py:69: UserWarning: Only CUDA, MUSA and NPU support GGUF quantization currently.
+  warnings.warn(f"Only CUDA, MUSA and NPU support GGUF quantization currently.")
+benchmark_args=Namespace(backend='sglang', base_url=None, host='174.1.60.5', port=31000, ready_check_timeout_sec=60, dataset_name='random-ids', dataset_path='', dataset_offset=0, agentic_max_turns=None, speed_bench_category=None, speed_bench_output_len=512, model=None, served_model_name=None, tokenizer='/data/hf_models/Kimi-K3', num_prompts=40, sharegpt_output_len=None, sharegpt_context_len=None, random_input_len=16384, random_output_len=512, random_range_ratio=1.0, image_count=1, image_resolution='1080p', random_image_count=False, image_format='jpeg', image_content='random', request_rate=10000.0, use_trace_timestamps=False, max_concurrency=8, output_file='/data/hzy/sskj/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-c8-20260827-1340/bench/decode_16k_512_c8.jsonl', output_details=True, print_requests=False, disable_tqdm=True, disable_stream=False, return_logprob=False, top_logprobs_num=0, token_ids_logprob=None, logprob_start_len=-1, return_routed_experts=False, cache_report=False, seed=42, disable_ignore_eos=False, temperature=0.0, top_p=1.0, extra_request_body=None, apply_chat_template=False, profile=False, plot_throughput=False, profile_activities=['CPU', 'GPU'], profile_start_step=None, profile_steps=None, profile_num_steps=None, profile_by_stage=False, profile_stages=None, profile_output_dir=None, profile_prefix=None, lora_name=None, lora_request_distribution='uniform', lora_zipf_alpha=1.5, prompt_suffix='', pd_separated=False, profile_prefill_url=None, profile_decode_url=None, flush_cache=False, warmup_requests=0, tokenize_prompt=False, gsp_num_groups=64, gsp_prompts_per_group=16, gsp_system_prompt_len=2048, gsp_question_len=128, gsp_output_len=256, gsp_range_ratio=1.0, gsp_fast_prepare=False, gsp_send_routing_key=False, gsp_num_turns=1, gsp_ordered=False, gsp_group_distribution='uniform', gsp_zipf_alpha=None, mooncake_slowdown_factor=1.0, mooncake_num_rounds=1, mooncake_workload='conversation', fake_prefill=False, tag=None, header=None)
+Waiting up to 60s for http://174.1.60.5:31000/v1/models to become ready...
+Server ready in 0.0s.
+'[Errno 101] Network is unreachable' thrown while requesting HEAD https://huggingface.co/kimi-k3/resolve/main/config.json
+Retrying in 1s [Retry 1/5].
+'[Errno 101] Network is unreachable' thrown while requesting HEAD https://huggingface.co/kimi-k3/resolve/main/config.json
+Retrying in 1s [Retry 1/5].
+Fail to load tokenizer config with error=Can't load the configuration of 'kimi-k3'. If you were trying to load it from 'https://huggingface.co/models', make sure you don't have a local directory with the same name. Otherwise, make sure 'kimi-k3' is the correct path to a directory containing a config.json file
+
+WARNING It is recommended to use the `Chat` or `Instruct` model for benchmarking.
+Because when the tokenizer counts the output tokens, if there is gibberish, it might count incorrectly.
+
+Namespace(backend='sglang', base_url=None, host='174.1.60.5', port=31000, ready_check_timeout_sec=60, dataset_name='random-ids', dataset_path='', dataset_offset=0, agentic_max_turns=None, speed_bench_category=None, speed_bench_output_len=512, model='kimi-k3', served_model_name=None, tokenizer='/data/hf_models/Kimi-K3', num_prompts=40, sharegpt_output_len=None, sharegpt_context_len=None, random_input_len=16384, random_output_len=512, random_range_ratio=1.0, image_count=1, image_resolution='1080p', random_image_count=False, image_format='jpeg', image_content='random', request_rate=10000.0, use_trace_timestamps=False, max_concurrency=8, output_file='/data/hzy/sskj/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-c8-20260827-1340/bench/decode_16k_512_c8.jsonl', output_details=True, print_requests=False, disable_tqdm=True, disable_stream=False, return_logprob=False, top_logprobs_num=0, token_ids_logprob=None, logprob_start_len=-1, return_routed_experts=False, cache_report=False, seed=42, disable_ignore_eos=False, temperature=0.0, top_p=1.0, extra_request_body=None, apply_chat_template=False, profile=False, plot_throughput=False, profile_activities=['CPU', 'GPU'], profile_start_step=None, profile_steps=None, profile_num_steps=None, profile_by_stage=False, profile_stages=None, profile_output_dir=None, profile_prefix=None, lora_name=None, lora_request_distribution='uniform', lora_zipf_alpha=1.5, prompt_suffix='', pd_separated=False, profile_prefill_url=None, profile_decode_url=None, flush_cache=False, warmup_requests=0, tokenize_prompt=False, gsp_num_groups=64, gsp_prompts_per_group=16, gsp_system_prompt_len=2048, gsp_question_len=128, gsp_output_len=256, gsp_range_ratio=1.0, gsp_fast_prepare=False, gsp_send_routing_key=False, gsp_num_turns=1, gsp_ordered=False, gsp_group_distribution='uniform', gsp_zipf_alpha=None, mooncake_slowdown_factor=1.0, mooncake_num_rounds=1, mooncake_workload='conversation', fake_prefill=False, tag=None, header=None)
+
+#Input tokens: 655360
+#Output tokens: 20480
+Starting warmup with 0 sequences...
+Warmup completed with 0 sequences. Starting main benchmark run...
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+
+============ Serving Benchmark Result ============
+Backend:                                 sglang    
+Traffic request rate:                    10000.0   
+Max request concurrency:                 8         
+Successful requests:                     40        
+Benchmark duration (s):                  257.58    
+Total input tokens:                      655360    
+Total input text tokens:                 655360    
+Total generated tokens:                  20480     
+Total generated tokens (retokenized):    20467     
+Request throughput (req/s):              0.16      
+Input token throughput (tok/s):          2544.27   
+Output token throughput (tok/s):         79.51     
+Peak output token throughput (tok/s):    128.00    
+Peak concurrent requests:                13        
+Total token throughput (tok/s):          2623.78   
+Concurrency:                             7.39      
+----------------End-to-End Latency----------------
+Mean E2E Latency (ms):                   47570.72  
+Median E2E Latency (ms):                 51368.52  
+P90 E2E Latency (ms):                    54690.14  
+P95 E2E Latency (ms):                    56342.67  
+P99 E2E Latency (ms):                    56439.38  
+---------------Time to First Token----------------
+Mean TTFT (ms):                          17104.55  
+Median TTFT (ms):                        18754.68  
+P90 TTFT (ms):                           21811.56  
+P95 TTFT (ms):                           23181.66  
+P99 TTFT (ms):                           23331.12  
+-----Time per Output Token (excl. 1st token)------
+Mean TPOT (ms):                          59.62     
+Median TPOT (ms):                        63.20     
+P90 TPOT (ms):                           66.47     
+P95 TPOT (ms):                           66.55     
+P99 TPOT (ms):                           66.78     
+---------------Inter-Token Latency----------------
+Mean ITL (ms):                           59.72     
+Median ITL (ms):                         64.62     
+P90 ITL (ms):                            67.31     
+P95 ITL (ms):                            69.48     
+P99 ITL (ms):                            70.99     
+Max ITL (ms):                            207.90    
+==================================================
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-c8-20260827-1340/logs/manual_d_node0.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-c8-20260827-1340/logs/manual_d_node0.log
new file mode 100644
index 0000000..c1c8d8d
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-c8-20260827-1340/logs/manual_d_node0.log
@@ -0,0 +1,1799 @@
+/usr/local/lib/python3.12/dist-packages/torchao/quantization/quant_api.py:1731: SyntaxWarning: invalid escape sequence '\.'
+  """Configuration class for applying different quantization configs to modules or parameters based on their fully qualified names (FQNs).
+/sgl-workspace/sglang/python/sglang/launch_server.py:56: UserWarning: 'python -m sglang.launch_server' is still supported, but 'sglang serve' is the recommended entrypoint.
+  Example: sglang serve --model-path <model> [options]
+  warnings.warn(
+[2026-08-27 04:36:08] KV cache is forced as chunk cache for decode server
+[2026-08-27 04:36:11] Breakable CUDA graph is incompatible with MLA attention; disabling prefill CUDA graph.
+[2026-08-27 04:36:12] Attention backend not specified. Use triton backend by default.
+[2026-08-27 04:36:14] server_args=ServerArgs(model_path='/data/hf_models/Kimi-K3', tokenizer_path='/data/hf_models/Kimi-K3', tokenizer_mode='auto', tokenizer_backend='huggingface', tokenizer_worker_num=1, detokenizer_worker_num=1, skip_tokenizer_init=False, load_format='auto', model_loader_extra_config='{}', trust_remote_code=True, context_length=None, is_embedding=False, enable_multimodal=None, revision=None, model_impl='auto', model_config_parser='auto', json_model_override_args='{}', dtype='auto', quantization=None, quantization_param_path=None, kv_cache_dtype='auto', enable_fp32_lm_head=False, modelopt_quant=None, modelopt_checkpoint_restore_path=None, modelopt_checkpoint_save_path=None, modelopt_export_path=None, quantize_and_serve=False, rl_quant_profile=None, enable_tf32_matmul=False, mem_fraction_static=0.88, max_running_requests=None, max_queued_requests=None, max_total_tokens=None, chunked_prefill_size=8192, enable_dynamic_chunking=False, max_prefill_tokens=16384, prefill_max_requests=None, schedule_policy='fcfs', enable_priority_scheduling=False, disable_priority_preemption=False, default_priority_value=None, abort_on_priority_when_disabled=False, schedule_low_priority_values_first=False, priority_scheduling_preemption_threshold=10, retraction_policy='length', schedule_conservativeness=1.0, page_size=64, swa_full_tokens_ratio=0.8, disable_hybrid_swa_memory=False, radix_eviction_policy='lru', prefill_only_disable_kv_cache=False, disable_radix_cache=True, enable_page_major_kv_layout=False, enable_unified_memory=False, disable_chunked_prefix_cache=False, disable_overlap_schedule=False, num_continuous_decode_steps=1, scheduler_recv_interval=1, enable_mixed_chunk=False, nccl_port=None, dist_timeout=3600, dist_init_addr='174.1.60.5:20000', nnodes=4, node_rank=0, tp_size=32, dcp_size=1, pp_size=1, pp_max_micro_batch_size=None, pp_async_batch_depth=0, dp_size=1, load_balance_method='round_robin', attn_cp_size=1, moe_dp_size=1, dcp_comm_backend='ag_rs', dcp_replicate_q_proj=None, dwdp_size=1, enable_prefill_cp=False, cp_strategy=None, enable_dsa_cache_layer_split=False, enable_dsa_prefill_context_parallel=False, dsa_prefill_cp_mode='round-robin-split', enable_prefill_context_parallel=False, prefill_cp_mode='in-seq-split', enable_cp_decode_attn_tp=False, enable_dp_attention=False, enable_dp_attention_local_control_broadcast=False, enable_dp_lm_head=False, enable_attn_tp_input_scattered=False, disable_attn_tp_gather=False, enable_p2p_check=False, device='cuda', base_gpu_id=0, gpu_id_step=1, random_seed=369235288, watchdog_timeout=300, soft_watchdog_timeout=None, sleep_on_idle=False, use_ray=False, custom_sigquit_handler=None, numa_node=None, gc_threshold=None, host='0.0.0.0', port=30000, fastapi_root_path='', smg_grpc_mode=False, grpc_mode=False, grpc_port=None, sidecar=None, sidecar_args=None, skip_server_warmup=False, warmups=None, enable_http2=False, ssl_keyfile=None, ssl_certfile=None, ssl_ca_certs=None, ssl_keyfile_password=None, enable_ssl_refresh=False, api_key=None, admin_api_key=None, served_model_name='kimi-k3', weight_version='default', chat_template=None, hf_chat_template_name=None, completion_template=None, file_storage_path='sglang_storage', enable_cache_report=False, reasoning_parser=None, default_chat_template_kwargs=None, strip_thinking_cache=False, enable_strict_thinking=False, tool_call_parser=None, tool_server=None, sampling_defaults='model', asr_max_buffer_seconds=60, asr_max_concurrent_sessions=32, preferred_sampling_params=None, allow_auto_truncate=False, stream_interval=1, batch_notify_size=16, stream_response_default_include_usage=False, incremental_streaming_output=False, enable_streaming_session=False, enable_session_radix_cache=False, log_level='info', log_level_http=None, log_requests=False, log_requests_level=2, log_requests_format='text', log_requests_target=None, uvicorn_access_log_exclude_prefixes=[], crash_dump_folder=None, show_time_cost=False, enable_metrics=False, smg_http_sidecar_port=None, enable_mfu_metrics=False, enable_metrics_for_all_schedulers=False, load_snapshot_publish_interval=15, tokenizer_metrics_custom_labels_header='x-custom-labels', tokenizer_metrics_allowed_custom_labels=None, extra_metric_labels=None, bucket_time_to_first_token=None, bucket_inter_token_latency=None, bucket_e2e_request_latency=None, prompt_tokens_buckets=None, generation_tokens_buckets=None, gc_warning_threshold_secs=0.0, decode_log_interval=40, enable_request_time_stats_logging=False, kv_events_config=None, enable_forward_pass_metrics=False, forward_pass_metrics_worker_id='', forward_pass_metrics_ipc_name=None, enable_trace=False, trace_modules='request', otlp_traces_endpoint='localhost:4317', export_metrics_to_file=False, export_metrics_to_file_dir=None, stat_loggers=None, constrained_json_whitespace_pattern=None, constrained_json_disable_any_whitespace=False, attention_backend='triton', decode_attention_backend=None, prefill_attention_backend=None, sampling_backend='flashinfer', grammar_backend='xgrammar', radix_cache_backend=None, mm_attention_backend=None, fp8_gemm_runner_backend='auto', fp4_gemm_runner_backend='auto', bf16_gemm_backend='auto', dsa_prefill_backend=None, dsa_decode_backend=None, dsa_paged_mqa_logits_backend='auto', dsa_topk_backend='sgl-kernel', disable_flashinfer_autotune=False, flashinfer_autotune_skip_ops=None, mamba_backend='triton', cuda_graph_config=CudaGraphConfig(decode=PhaseConfig(backend='full', max_bs=16, bs=[1, 2, 4, 8, 12, 16], tc_compiler='eager', full_prefill_max_req=None), prefill=PhaseConfig(backend='disabled', max_bs=2048, bs=[4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], tc_compiler='eager', full_prefill_max_req=None)), cuda_graph_backend_decode=None, cuda_graph_backend_prefill=None, cuda_graph_max_bs_decode=16, cuda_graph_max_bs_prefill=None, cuda_graph_bs_decode=None, cuda_graph_bs_prefill=None, cuda_graph_tc_compiler=None, disable_prefill_cuda_graph=False, disable_decode_cuda_graph=False, disable_cuda_graph=False, disable_cuda_graph_padding=False, enable_profile_cuda_graph=False, enable_cudagraph_gc=False, debug_cuda_graph=False, enable_layerwise_nvtx_marker=False, enable_nccl_nvls=False, enable_symm_mem=False, triton_attention_reduce_in_fp32=False, triton_attention_num_kv_splits=8, triton_attention_split_tile_size=None, flashinfer_mla_disable_ragged=False, enable_fused_qk_norm_rope=False, enable_precise_embedding_interpolation=False, enable_fused_moe_sum_all_reduce=False, enable_deepseek_v4_fp4_indexer=False, disable_custom_all_reduce=False, enable_mscclpp=False, enable_torch_symm_mem=False, enable_scattered_sconv=False, pre_warm_nccl=False, enable_quant_communications=False, enable_flashinfer_allreduce_fusion=False, enforce_disable_flashinfer_allreduce_fusion=False, flashinfer_allreduce_fusion_backend=None, enable_aiter_allreduce_fusion=False, enable_torch_compile=False, enable_torch_compile_debug_mode=False, torch_compile_max_bs=32, torchao_config='', speculative_algorithm=None, speculative_draft_model_path=None, speculative_draft_model_revision=None, speculative_draft_load_format=None, speculative_num_steps=None, speculative_eagle_topk=None, speculative_num_draft_tokens=None, speculative_dflash_block_size=None, speculative_dspark_block_size=None, speculative_dspark_sps_table_path=None, speculative_dspark_confidence_sts_path=None, speculative_dspark_align_verify_tokens_to_graph_tier=False, speculative_accept_threshold_single=1.0, speculative_accept_threshold_acc=1.0, speculative_use_rejection_sampling=False, speculative_token_map=None, speculative_attention_mode='prefill', speculative_draft_attention_backend=None, speculative_draft_window_size=None, speculative_moe_runner_backend='marlin', speculative_moe_a2a_backend=None, speculative_draft_model_quantization=None, speculative_skip_dp_mlp_sync=False, enable_multi_layer_eagle=False, speculative_adaptive=False, speculative_adaptive_config=None, decoupled_spec_bind_endpoint=None, decoupled_spec_connect_endpoints=None, decoupled_spec_rank=None, decoupled_spec_role='null', spec_trace_dir=None, speculative_ngram_min_bfs_breadth=1, speculative_ngram_max_bfs_breadth=10, speculative_ngram_match_type='BFS', speculative_ngram_max_trie_depth=18, speculative_ngram_capacity=10000000, speculative_ngram_external_corpus_path=None, speculative_ngram_external_sam_budget=0, speculative_ngram_external_corpus_max_tokens=10000000, ep_size=32, moe_a2a_backend='none', moe_runner_backend='marlin', flashinfer_mxfp4_moe_precision='default', deepep_mode='auto', fuseep_mode=2, deepep_dispatcher_output_dtype='auto', ep_num_redundant_experts=0, ep_dispatch_algorithm=None, init_expert_location='trivial', enable_eplb=False, eplb_algorithm='auto', eplb_rebalance_num_iterations=1000, eplb_rebalance_layers_per_chunk=None, eplb_min_rebalancing_utilization_threshold=1.0, expert_distribution_recorder_mode=None, expert_distribution_recorder_buffer_size=1000, enable_expert_distribution_metrics=False, deepep_config=None, moe_dense_tp_size=None, elastic_ep_backend=None, enable_elastic_expert_backup=False, mooncake_ib_device=None, enable_waterfill=False, ep_join_mode=None, ep_join_rank_offset=0, elastic_ep_initial_size=None, max_ep_size=None, elastic_ep_scale_timeout=600, elastic_ep_rejoin=False, disable_flashinfer_cutlass_moe_fp4_allgather=False, disable_shared_experts_fusion=False, enforce_shared_experts_fusion=False, max_mamba_cache_size=None, mamba_ssm_dtype=None, mamba_max_states_per_path=-1, enable_mamba_cache_stochastic_rounding=False, mamba_cache_philox_rounds=0, mamba_full_memory_ratio=0.21, mamba_radix_cache_strategy='extra_buffer_lazy', uses_mamba_radix_cache=False, mamba_track_interval=256, enable_int8_mamba_checkpoint=False, int8_mamba_ckpt_size=None, linear_attn_backend='triton', linear_attn_decode_backend=None, linear_attn_prefill_backend=None, linear_attn_verify_backend=None, enable_linear_replayssm=False, linear_replayssm_cache_len=16, enable_linear_replayssm_spec=False, enable_hierarchical_cache=False, hicache_ratio=2.0, hicache_size=0, hicache_write_policy='write_through', hicache_io_backend='kernel', hicache_mem_layout='page_first', hicache_storage_backend=None, hicache_storage_prefetch_policy='timeout', hicache_storage_backend_extra_config=None, enable_hisparse=False, hisparse_config=None, enable_broadcast_mm_inputs_process=False, enable_prefix_mm_cache=False, mm_enable_dp_encoder=False, mm_process_config={}, mm_processor_worker_num=0, mm_io_worker_num=0, limit_mm_data_per_request=None, enable_mm_global_cache=False, disable_fast_image_processor=False, mm_feature_transport='cpu', keep_mm_feature_on_device=False, enable_lora=None, enable_lora_overlap_loading=None, max_lora_rank=None, lora_target_modules=None, lora_paths=None, max_loaded_loras=None, max_loras_per_batch=8, lora_eviction_policy='lru', lora_backend='csgmv', max_lora_chunk_size=16, experts_shared_outer_loras=None, lora_use_virtual_experts=False, lora_strict_loading=False, lora_drain_wait_threshold=0.0, enable_two_batch_overlap=False, enable_single_batch_overlap=False, tbo_token_distribution_threshold=0.48, cpu_offload_gb=0, offload_group_size=-1, offload_num_in_group=1, offload_prefetch_step=1, offload_mode='cpu', enable_lmcache=False, lmcache_config_file=None, enable_flexkv=False, flexkv_config_file=None, kt_weight_path=None, kt_method='AMXINT4', kt_cpuinfer=None, kt_threadpool_count=2, kt_num_gpu_experts=None, kt_max_deferred_experts_per_token=None, dllm_algorithm=None, dllm_algorithm_config=None, dllm_fdfo=True, disaggregation_mode='decode', disaggregation_transfer_backend='mooncake', disaggregation_bootstrap_port=28800, disaggregation_ib_device='mlx5_0,mlx5_1,mlx5_2,mlx5_3', disaggregation_decode_enable_radix_cache=False, disaggregation_decode_enable_offload_kvcache=False, num_reserved_decode_tokens=512, disaggregation_decode_extra_slots=0, disaggregation_decode_polling_interval=1, optimistic_prefill_attempts=0, encoder_only=False, language_only=False, encoder_transfer_backend='zmq_to_tokenizer', encoder_urls=[], encoder_bootstrap_port=8997, encoder_register_urls=[], enable_adaptive_dispatch_to_encoder=False, enable_pdmux=False, pdmux_config_path=None, sm_group_num=8, custom_weight_loader=[], weight_loader_disable_mmap=False, weight_loader_prefetch_checkpoints=False, weight_loader_prefetch_num_threads=4, weight_loader_drop_cache_after_load=False, remote_instance_weight_loader_seed_instance_ip=None, remote_instance_weight_loader_seed_instance_service_port=None, remote_instance_weight_loader_send_weights_group_ports=None, remote_instance_weight_loader_backend='nccl', remote_instance_weight_loader_start_seed_via_transfer_engine=False, engine_info_bootstrap_port=6789, modelexpress_config=None, download_dir=None, model_checksum=None, delete_ckpt_after_loading=False, decrypted_config_file=None, decrypted_draft_config_file=None, checkpoint_engine_wait_weights_before_ready=False, enable_prefill_delayer=False, prefill_delayer_max_delay_passes=30, prefill_delayer_token_usage_low_watermark=None, prefill_delayer_forward_passes_buckets=None, prefill_delayer_wait_seconds_buckets=None, prefill_delayer_queue_min_ratio=None, prefill_delayer_max_delay_ms=None, min_free_slots_delay=None, enable_deterministic_inference=False, rl_on_policy_target=None, kv_canary='none', kv_canary_real_data='none', kv_canary_sweep_interval=0, enable_dynamic_batch_tokenizer=False, dynamic_batch_tokenizer_batch_size=32, dynamic_batch_tokenizer_batch_timeout=0.002, enable_tokenizer_batch_encode=False, disable_tokenizer_batch_decode=False, debug_tensor_dump_output_folder=None, debug_tensor_dump_layers=None, debug_tensor_dump_input_file=None, enable_memory_saver=False, enable_weights_cpu_backup=False, enable_draft_weights_cpu_backup=False, enable_custom_logit_processor=False, enable_return_hidden_states=False, enable_return_routed_experts=False, enable_return_indexer_topk=False, disable_outlines_disk_cache=False, enable_mis=False, weight_cache_mode='off', weight_cache_socket=None, weight_cache_timeout=1800, forward_hooks=None, msprobe_dump_config=None)
+[2026-08-27 04:36:14] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:14] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:14] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:14] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:14] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:14] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:14] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:14] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:17] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:17] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:18] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:19] Multimodal processor concurrency enabled with 2 isolated worker threads (auto).
+[2026-08-27 04:36:19] No HuggingFace chat template found
+[2026-08-27 04:36:19] No chat template found, defaulting to 'string' content format
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:34 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:34 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:34 TP4 EP4] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:34 TP4 EP4] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:34 TP6 EP6] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:34 TP6 EP6] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:34] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:34] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:34 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:34 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+/sgl-workspace/sglang/python/sglang/srt/utils/hf_transformers/tokenizer.py:423: UserWarning: Using a slow tokenizer. This might cause a significant slowdown. Consider using a fast tokenizer instead.
+  warnings.warn(
+[2026-08-27 04:36:34] Applying special tokens cache patch for Kimi tokenizer: <class 'transformers_modules.Kimi_hyphen_K3.caf40b2773f9029e.tokenization_kimi.TikTokenTokenizer'>
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:34 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:34 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:34 TP7 EP7] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:34 TP7 EP7] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:34 TP5 EP5] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:34 TP5 EP5] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:35 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:35 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:36 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:36 TP6 EP6] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:36 TP4 EP4] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:36 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:36 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:36 TP7 EP7] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:36 TP5 EP5] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:37 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:37.190651   236 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:37.190778   236 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:37.190796   236 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.5 port: 12001
+I0827 04:36:37.190856   236 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.5:16500
+I0827 04:36:37.191038   236 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:37.194572   241 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:37.194692   241 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:37.194712   241 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.5 port: 12001
+I0827 04:36:37.194767   241 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.5:15678
+I0827 04:36:37.194965   241 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:37.219420   236 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:37.223201   241 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:37.227941   236 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:37.231628   241 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:37.236233   236 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:37.239850   241 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:37.244647   236 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:37.248145   241 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:37.248418   236 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:37.248464   236 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:37.248525   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.250921   241 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:37.250947   241 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:37.251004   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.255973   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:37.257668   236 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:05
+I0827 04:36:37.257705   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.283210   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:37.283238   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:37.284873   236 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:05
+I0827 04:36:37.284902   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.284904   241 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:05
+I0827 04:36:37.284934   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.293713   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:37.295406   241 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:05
+I0827 04:36:37.295433   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.311285   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:37.312896   236 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:05
+I0827 04:36:37.312925   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.327399   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:37.328944   241 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:05
+I0827 04:36:37.328972   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.339579   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:37.341301   236 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:05
+I0827 04:36:37.341434   236 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:37 TP1 EP1] Init torch distributed begin.
+I0827 04:36:37.359290   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:37.360814   241 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:05
+I0827 04:36:37.360936   241 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:37 TP6 EP6] Init torch distributed begin.
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:37.366210   239 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:37.366365   239 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:37.366390   239 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.5 port: 12001
+I0827 04:36:37.366467   239 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.5:15787
+I0827 04:36:37.366704   239 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:37.395485   239 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:37.402652   239 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:37.409775   239 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:37.434130   239 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:37.436257   239 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:37.436282   239 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:37.436331   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.442425   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:37.443729   239 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:05
+I0827 04:36:37.443749   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.448421   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:37.449750   239 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:05
+I0827 04:36:37.449769   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.454622   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:37.455952   239 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:05
+I0827 04:36:37.455972   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.463356   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:37.464902   239 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:05
+I0827 04:36:37.465021   239 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:37 TP4 EP4] Init torch distributed begin.
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:37.510618   238 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:37.510744   238 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:37.510763   238 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.5 port: 12001
+I0827 04:36:37.510824   238 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.5:16205
+I0827 04:36:37.511003   238 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:37.524268   242 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:37.524391   242 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:37.524411   242 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.5 port: 12001
+I0827 04:36:37.524524   242 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.5:16605
+I0827 04:36:37.524709   242 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:37.528693   235 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:37.528817   235 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:37.528836   235 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.5 port: 12001
+I0827 04:36:37.528893   235 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.5:16672
+I0827 04:36:37.529069   235 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:37.535192   238 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:37.535425   235 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:37.541376   235 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:37.542625   238 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:37.547586   235 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:37.549854   238 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:37.551366   242 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:37.553905   235 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:37.556406   235 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:37.556468   235 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:37.556523   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.558794   242 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:37.563313   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:37.566102   242 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:37.573271   242 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:37.574895   238 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:37.576385   242 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:37.576412   242 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:37.576478   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.577947   238 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:37.577976   238 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:37.578033   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.583841   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:37.585472   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:37.585623   242 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:05
+I0827 04:36:37.585656   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.587072   238 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:05
+I0827 04:36:37.587105   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.594069   235 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:05
+I0827 04:36:37.594128   235 rdma_context.cpp:205] Using SIEVE endpoint store
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:37.598918   240 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:37.599033   240 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:37.599050   240 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.5 port: 12001
+I0827 04:36:37.599107   240 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.5:15428
+I0827 04:36:37.599270   240 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:37.605341   240 topology.cpp:127] Device mlx5_0 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:37.606741   237 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:37.606841   237 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:37.606855   237 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.5 port: 12001
+I0827 04:36:37.606901   237 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.5:16531
+I0827 04:36:37.607028   237 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:37.611215   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:37.611989   237 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:37.612797   242 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:05
+I0827 04:36:37.612828   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.615110   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:37.616685   238 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:05
+I0827 04:36:37.616714   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.619258   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:37.620560   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:37.620851   235 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:05
+I0827 04:36:37.620885   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.622239   242 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:05
+I0827 04:36:37.622268   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.624459   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:37.625998   238 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:05
+I0827 04:36:37.626025   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.630085   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:37.630686   240 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:37.631680   242 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:05
+I0827 04:36:37.631806   242 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:37 TP7 EP7] Init torch distributed begin.
+I0827 04:36:37.633723   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:37.635268   238 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:05
+I0827 04:36:37.635390   238 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:37 TP3 EP3] Init torch distributed begin.
+I0827 04:36:37.642561   237 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:37.649834   237 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:37.651238   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:37.652786   235 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:05
+I0827 04:36:37.652813   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.657148   237 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:37.658787   240 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:37.659991   237 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:37.660017   237 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:37.660071   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.667523   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:37.669195   240 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:37.669226   237 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:05
+I0827 04:36:37.669257   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.672511   240 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:37.672547   240 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:37.672606   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.677067   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:37.678661   237 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:05
+I0827 04:36:37.678689   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.682178   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:37.683274   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:37.683740   240 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:05
+I0827 04:36:37.683773   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.684834   235 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:05
+I0827 04:36:37.684954   235 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:37 TP0 EP0] Init torch distributed begin.
+I0827 04:36:37.703369   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:37.704975   237 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:05
+I0827 04:36:37.705003   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.711591   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:37.713003   240 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:05
+I0827 04:36:37.713030   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.735574   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:37.737313   237 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:05
+I0827 04:36:37.737439   237 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:37 TP2 EP2] Init torch distributed begin.
+I0827 04:36:37.740485   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:37.741588   240 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:05
+I0827 04:36:37.741605   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.766494   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:37.767982   240 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:05
+I0827 04:36:37.768137   240 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:37 TP5 EP5] Init torch distributed begin.
+[2026-08-27 04:36:46 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:36:48 TP0 EP0] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP1 EP1] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP2 EP2] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP5 EP5] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP4 EP4] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP3 EP3] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP6 EP6] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP7 EP7] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP0 EP0] Init torch distributed ends. elapsed=10.55 s, mem usage=0.40 GB
+[2026-08-27 04:36:48 TP7 EP7] Init torch distributed ends. elapsed=10.61 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP6 EP6] Init torch distributed ends. elapsed=10.88 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP4 EP4] Init torch distributed ends. elapsed=10.77 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP3 EP3] Init torch distributed ends. elapsed=10.60 s, mem usage=0.40 GB
+[2026-08-27 04:36:48 TP1 EP1] Init torch distributed ends. elapsed=10.90 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP5 EP5] Init torch distributed ends. elapsed=10.47 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP2 EP2] Init torch distributed ends. elapsed=10.50 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP7 EP7] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP4 EP4] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP1 EP1] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP6 EP6] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP7 EP7] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP4 EP4] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP5 EP5] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP6 EP6] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP3 EP3] Load weight begin. avail mem=82.19 GB
+[2026-08-27 04:36:48 TP0 EP0] Load weight begin. avail mem=82.19 GB
+[2026-08-27 04:36:48 TP5 EP5] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP2 EP2] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP4 EP4] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP7 EP7] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP6 EP6] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP4 EP4] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP7 EP7] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP6 EP6] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP5 EP5] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP5 EP5] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:50 TP1 EP1] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP0 EP0] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP2 EP2] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP3 EP3] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP5 EP5] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP4 EP4] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP6 EP6] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP7 EP7] multimem all-gather disabled because the TP group spans across nodes.
+
Multi-thread loading shards:   0% Completed | 0/96 [00:00<?, ?it/s]
Multi-thread loading shards:   1% Completed | 1/96 [00:00<00:24,  3.85it/s]
Multi-thread loading shards:   2% Completed | 2/96 [00:07<07:13,  4.61s/it]
Multi-thread loading shards:   3% Completed | 3/96 [00:09<04:44,  3.06s/it]
Multi-thread loading shards:   4% Completed | 4/96 [00:10<03:33,  2.32s/it]
Multi-thread loading shards:   5% Completed | 5/96 [00:11<02:54,  1.92s/it]
Multi-thread loading shards:   6% Completed | 6/96 [00:12<02:29,  1.66s/it]
Multi-thread loading shards:   7% Completed | 7/96 [00:13<02:13,  1.50s/it]
Multi-thread loading shards:   8% Completed | 8/96 [00:15<02:04,  1.41s/it]
Multi-thread loading shards:   9% Completed | 9/96 [00:16<01:56,  1.34s/it]
Multi-thread loading shards:  10% Completed | 10/96 [00:17<01:47,  1.25s/it]
Multi-thread loading shards:  11% Completed | 11/96 [00:18<01:44,  1.23s/it]
Multi-thread loading shards:  12% Completed | 12/96 [00:19<01:43,  1.23s/it]
Multi-thread loading shards:  14% Completed | 13/96 [00:21<01:59,  1.44s/it]
Multi-thread loading shards:  15% Completed | 14/96 [00:22<01:51,  1.36s/it]
Multi-thread loading shards:  16% Completed | 15/96 [00:24<01:45,  1.30s/it]
Multi-thread loading shards:  17% Completed | 16/96 [00:25<01:42,  1.28s/it]
Multi-thread loading shards:  18% Completed | 17/96 [00:26<01:39,  1.26s/it]
Multi-thread loading shards:  19% Completed | 18/96 [00:27<01:36,  1.24s/it]
Multi-thread loading shards:  20% Completed | 19/96 [00:28<01:32,  1.20s/it]
Multi-thread loading shards:  21% Completed | 20/96 [00:29<01:30,  1.19s/it]
Multi-thread loading shards:  22% Completed | 21/96 [00:31<01:29,  1.19s/it]
Multi-thread loading shards:  23% Completed | 22/96 [00:32<01:27,  1.18s/it]
Multi-thread loading shards:  24% Completed | 23/96 [00:33<01:26,  1.18s/it]
Multi-thread loading shards:  25% Completed | 24/96 [00:34<01:23,  1.16s/it]
Multi-thread loading shards:  26% Completed | 25/96 [00:35<01:20,  1.14s/it]
Multi-thread loading shards:  27% Completed | 26/96 [00:36<01:18,  1.13s/it]
Multi-thread loading shards:  28% Completed | 27/96 [00:37<01:19,  1.15s/it]
Multi-thread loading shards:  29% Completed | 28/96 [00:39<01:18,  1.16s/it]
Multi-thread loading shards:  30% Completed | 29/96 [00:40<01:17,  1.16s/it]
Multi-thread loading shards:  31% Completed | 30/96 [00:41<01:17,  1.18s/it]
Multi-thread loading shards:  32% Completed | 31/96 [00:42<01:16,  1.18s/it]
Multi-thread loading shards:  33% Completed | 32/96 [00:44<01:30,  1.42s/it]
Multi-thread loading shards:  34% Completed | 33/96 [00:45<01:23,  1.33s/it]
Multi-thread loading shards:  35% Completed | 34/96 [00:46<01:17,  1.26s/it]
Multi-thread loading shards:  36% Completed | 35/96 [00:48<01:14,  1.23s/it]
Multi-thread loading shards:  38% Completed | 36/96 [00:49<01:13,  1.22s/it]
Multi-thread loading shards:  39% Completed | 37/96 [00:50<01:11,  1.21s/it]
Multi-thread loading shards:  40% Completed | 38/96 [00:51<01:09,  1.20s/it]
Multi-thread loading shards:  41% Completed | 39/96 [00:52<01:07,  1.18s/it]
Multi-thread loading shards:  42% Completed | 40/96 [00:53<01:05,  1.18s/it]
Multi-thread loading shards:  43% Completed | 41/96 [00:55<01:04,  1.17s/it]
Multi-thread loading shards:  44% Completed | 42/96 [00:56<01:02,  1.17s/it]
Multi-thread loading shards:  45% Completed | 43/96 [00:57<01:03,  1.19s/it]
Multi-thread loading shards:  46% Completed | 44/96 [00:58<01:01,  1.18s/it]
Multi-thread loading shards:  47% Completed | 45/96 [00:59<00:57,  1.13s/it]
Multi-thread loading shards:  48% Completed | 46/96 [01:00<00:57,  1.15s/it]
Multi-thread loading shards:  49% Completed | 47/96 [01:02<00:57,  1.17s/it]
Multi-thread loading shards:  50% Completed | 48/96 [01:03<00:55,  1.15s/it]
Multi-thread loading shards:  51% Completed | 49/96 [01:04<00:53,  1.13s/it]
Multi-thread loading shards:  52% Completed | 50/96 [01:06<01:01,  1.34s/it]
Multi-thread loading shards:  53% Completed | 51/96 [01:07<00:58,  1.29s/it]
Multi-thread loading shards:  54% Completed | 52/96 [01:08<00:55,  1.27s/it]
Multi-thread loading shards:  55% Completed | 53/96 [01:09<00:53,  1.24s/it]
Multi-thread loading shards:  56% Completed | 54/96 [01:10<00:50,  1.21s/it]
Multi-thread loading shards:  57% Completed | 55/96 [01:12<00:49,  1.22s/it]
Multi-thread loading shards:  58% Completed | 56/96 [01:13<00:48,  1.22s/it]
Multi-thread loading shards:  59% Completed | 57/96 [01:14<00:47,  1.21s/it]
Multi-thread loading shards:  60% Completed | 58/96 [01:15<00:45,  1.20s/it]
Multi-thread loading shards:  61% Completed | 59/96 [01:16<00:44,  1.19s/it]
Multi-thread loading shards:  62% Completed | 60/96 [01:17<00:43,  1.20s/it]
Multi-thread loading shards:  64% Completed | 61/96 [01:19<00:42,  1.21s/it]
Multi-thread loading shards:  65% Completed | 62/96 [01:20<00:40,  1.19s/it]
Multi-thread loading shards:  66% Completed | 63/96 [01:21<00:40,  1.21s/it]
Multi-thread loading shards:  67% Completed | 64/96 [01:22<00:38,  1.20s/it]
Multi-thread loading shards:  68% Completed | 65/96 [01:24<00:37,  1.20s/it]
Multi-thread loading shards:  69% Completed | 66/96 [01:25<00:36,  1.20s/it]
Multi-thread loading shards:  70% Completed | 67/96 [01:26<00:35,  1.22s/it]
Multi-thread loading shards:  71% Completed | 68/96 [01:28<00:41,  1.48s/it]
Multi-thread loading shards:  72% Completed | 69/96 [01:29<00:37,  1.39s/it]
Multi-thread loading shards:  73% Completed | 70/96 [01:30<00:34,  1.34s/it]
Multi-thread loading shards:  74% Completed | 71/96 [01:32<00:32,  1.29s/it]
Multi-thread loading shards:  75% Completed | 72/96 [01:33<00:31,  1.30s/it]
Multi-thread loading shards:  76% Completed | 73/96 [01:34<00:29,  1.29s/it]
Multi-thread loading shards:  77% Completed | 74/96 [01:35<00:27,  1.27s/it]
Multi-thread loading shards:  78% Completed | 75/96 [01:37<00:26,  1.26s/it]
Multi-thread loading shards:  79% Completed | 76/96 [01:38<00:24,  1.24s/it]
Multi-thread loading shards:  80% Completed | 77/96 [01:39<00:23,  1.23s/it]
Multi-thread loading shards:  81% Completed | 78/96 [01:40<00:21,  1.19s/it]
Multi-thread loading shards:  82% Completed | 79/96 [01:41<00:20,  1.20s/it]
Multi-thread loading shards:  83% Completed | 80/96 [01:43<00:19,  1.20s/it]
Multi-thread loading shards:  84% Completed | 81/96 [01:44<00:18,  1.21s/it]
Multi-thread loading shards:  85% Completed | 82/96 [01:45<00:16,  1.20s/it]
Multi-thread loading shards:  86% Completed | 83/96 [01:46<00:15,  1.17s/it]
Multi-thread loading shards:  88% Completed | 84/96 [01:47<00:14,  1.17s/it]
Multi-thread loading shards:  89% Completed | 85/96 [01:48<00:12,  1.10s/it]
Multi-thread loading shards:  90% Completed | 86/96 [01:49<00:10,  1.05s/it]
Multi-thread loading shards:  91% Completed | 87/96 [01:50<00:09,  1.02s/it]
Multi-thread loading shards:  92% Completed | 88/96 [01:51<00:07,  1.01it/s]
Multi-thread loading shards:  93% Completed | 89/96 [01:52<00:06,  1.03it/s]
Multi-thread loading shards:  94% Completed | 90/96 [01:53<00:05,  1.04it/s]
Multi-thread loading shards:  95% Completed | 91/96 [01:54<00:04,  1.06it/s]
Multi-thread loading shards:  96% Completed | 92/96 [01:55<00:03,  1.06it/s]
Multi-thread loading shards:  97% Completed | 93/96 [01:56<00:02,  1.08it/s]
Multi-thread loading shards:  98% Completed | 94/96 [01:56<00:01,  1.46it/s]
Multi-thread loading shards: 100% Completed | 96/96 [01:56<00:00,  2.28it/s]
Multi-thread loading shards: 100% Completed | 96/96 [01:56<00:00,  1.21s/it]
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:58 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP6 EP6] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP0 EP0] Load weight end. elapsed=130.28 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP6 EP6] Load weight end. elapsed=130.29 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP3 EP3] Load weight end. elapsed=130.28 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP2 EP2] Load weight end. elapsed=130.32 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP5 EP5] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP5 EP5] Load weight end. elapsed=130.38 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP4 EP4] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP4 EP4] Load weight end. elapsed=130.48 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP7 EP7] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP7 EP7] Load weight end. elapsed=130.48 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:39:01 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:39:01 TP1 EP1] Load weight end. elapsed=132.81 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:39:01 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP6 EP6] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP6 EP6] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP5 EP5] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP5 EP5] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP7 EP7] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP7 EP7] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP4 EP4] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP4 EP4] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:02 TP0 EP0] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP7 EP7] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP6 EP6] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP5 EP5] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP1 EP1] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP4 EP4] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP3 EP3] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP2 EP2] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP7 EP7] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP4 EP4] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP6 EP6] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP5 EP5] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP0 EP0] Memory pool end. avail mem=9.18 GB
+[2026-08-27 04:39:02 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP1 EP1] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP6 EP6] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP5 EP5] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP7 EP7] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP6 EP6] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP5 EP5] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP7 EP7] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP4 EP4] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP4 EP4] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP2 EP2] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP3 EP3] Memory pool end. avail mem=9.18 GB
+[2026-08-27 04:39:02 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:39:02 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:39:02 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP0 EP0] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.17 GB
+[2026-08-27 04:39:02 TP6 EP6] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP6 EP6] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP1 EP1] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP5 EP5] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP7 EP7] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP5 EP5] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP7 EP7] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:39:02 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP3 EP3] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.17 GB
+[2026-08-27 04:39:02 TP4 EP4] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP4 EP4] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:39:02 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP2 EP2] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+
  0%|          | 0/6 [00:00<?, ?it/s]
Capturing batches (bs=16 avail_mem=9.00 GB):   0%|          | 0/6 [00:00<?, ?it/s]
Capturing batches (bs=16 avail_mem=9.00 GB):  17%|█▋        | 1/6 [03:43<18:38, 223.66s/it]
Capturing batches (bs=12 avail_mem=8.84 GB):  17%|█▋        | 1/6 [03:43<18:38, 223.66s/it]
Capturing batches (bs=12 avail_mem=8.84 GB):  33%|███▎      | 2/6 [03:44<06:11, 92.84s/it] 
Capturing batches (bs=8 avail_mem=8.78 GB):  33%|███▎      | 2/6 [03:44<06:11, 92.84s/it] 
Capturing batches (bs=8 avail_mem=8.78 GB):  50%|█████     | 3/6 [03:49<02:36, 52.32s/it]
Capturing batches (bs=4 avail_mem=8.73 GB):  50%|█████     | 3/6 [03:49<02:36, 52.32s/it]
Capturing batches (bs=4 avail_mem=8.73 GB):  67%|██████▋   | 4/6 [03:49<01:03, 32.00s/it]
Capturing batches (bs=2 avail_mem=8.68 GB):  67%|██████▋   | 4/6 [03:49<01:03, 32.00s/it]
Capturing batches (bs=2 avail_mem=8.68 GB):  83%|████████▎ | 5/6 [03:50<00:20, 20.77s/it]
Capturing batches (bs=1 avail_mem=8.62 GB):  83%|████████▎ | 5/6 [03:50<00:20, 20.77s/it]
Capturing batches (bs=1 avail_mem=8.62 GB): 100%|██████████| 6/6 [03:52<00:00, 14.36s/it]
Capturing batches (bs=1 avail_mem=8.62 GB): 100%|██████████| 6/6 [03:52<00:00, 38.78s/it]
+[2026-08-27 04:42:56 TP4 EP4] Capture target decode CUDA graph end. elapsed=233.75 s, mem usage=0.60 GB, avail mem=8.52 GB.
+[2026-08-27 04:42:56 TP4 EP4] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP3 EP3] Capture target decode CUDA graph end. elapsed=233.75 s, mem usage=0.62 GB, avail mem=8.55 GB.
+[2026-08-27 04:42:56 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP0 EP0] Capture target decode CUDA graph end. elapsed=233.77 s, mem usage=0.60 GB, avail mem=8.56 GB.
+[2026-08-27 04:42:56 TP0 EP0] max_total_num_tokens=440256, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=181, context_len=1048576, available_gpu_mem=8.56 GB
+[2026-08-27 04:42:56 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP7 EP7] Capture target decode CUDA graph end. elapsed=233.77 s, mem usage=0.62 GB, avail mem=8.51 GB.
+[2026-08-27 04:42:56 TP7 EP7] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP6 EP6] Capture target decode CUDA graph end. elapsed=233.86 s, mem usage=0.56 GB, avail mem=8.59 GB.
+[2026-08-27 04:42:56 TP6 EP6] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP1 EP1] Capture target decode CUDA graph end. elapsed=233.86 s, mem usage=0.58 GB, avail mem=8.58 GB.
+[2026-08-27 04:42:56 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP2 EP2] Capture target decode CUDA graph end. elapsed=233.86 s, mem usage=0.56 GB, avail mem=8.59 GB.
+[2026-08-27 04:42:56 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP5 EP5] Capture target decode CUDA graph end. elapsed=233.90 s, mem usage=0.58 GB, avail mem=8.58 GB.
+[2026-08-27 04:42:56 TP5 EP5] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+W0827 04:42:57.550421  2569 memory_location.cpp:73] Failed to get NUMA node, addr: 0x599a0800, len: 23168: Operation not permitted [1]
+W0827 04:42:57.550521  2570 memory_location.cpp:73] Failed to get NUMA node, addr: 0x599cb380, len: 23168: Operation not permitted [1]
+W0827 04:42:57.550542  2571 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59e1b540, len: 23168: Operation not permitted [1]
+W0827 04:42:57.550628  2572 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59c8a2c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.550725  2573 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5be7a240, len: 185344: Operation not permitted [1]
+W0827 04:42:57.550769  2574 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5bea7680, len: 185344: Operation not permitted [1]
+W0827 04:42:57.550900  2575 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59cc48c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.551011  2576 memory_location.cpp:73] Failed to get NUMA node, addr: 0x45f39fc0, len: 46336: Operation not permitted [1]
+W0827 04:42:57.551070  2577 memory_location.cpp:73] Failed to get NUMA node, addr: 0x599bfbc0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.551105  2578 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59c69c80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.580384  2676 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59c988c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.580493  2678 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59a30c40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.580636  2677 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59d06340, len: 23168: Operation not permitted [1]
+W0827 04:42:57.580636  2679 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5bd91640, len: 185344: Operation not permitted [1]
+W0827 04:42:57.580685  2675 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59e30d80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.580811  2681 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59f53440, len: 23168: Operation not permitted [1]
+W0827 04:42:57.580950  2680 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5bdbea80, len: 185344: Operation not permitted [1]
+W0827 04:42:57.581368  2686 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59dca000, len: 23168: Operation not permitted [1]
+W0827 04:42:57.581421  2682 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5987fc00, len: 46336: Operation not permitted [1]
+W0827 04:42:57.581741  2684 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59e10480, len: 23168: Operation not permitted [1]
+W0827 04:42:57.597819  2845 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59224900, len: 23168: Operation not permitted [1]
+W0827 04:42:57.598558  2839 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5917e740, len: 23168: Operation not permitted [1]
+W0827 04:42:57.598562  2849 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58e6ce00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.598762  2838 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58e982c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.598834  2841 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5b32ec00, len: 185344: Operation not permitted [1]
+W0827 04:42:57.598897  2834 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5927b680, len: 23168: Operation not permitted [1]
+W0827 04:42:57.598922  2843 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5b35c040, len: 185344: Operation not permitted [1]
+W0827 04:42:57.598953  2836 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5939fc40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.599097  2850 memory_location.cpp:73] Failed to get NUMA node, addr: 0x592d3fc0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.599287  2847 memory_location.cpp:73] Failed to get NUMA node, addr: 0x453f4c40, len: 46336: Operation not permitted [1]
+W0827 04:42:57.612057  2918 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59f4a140, len: 23168: Operation not permitted [1]
+W0827 04:42:57.612175  2919 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59f098c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.612243  2920 memory_location.cpp:73] Failed to get NUMA node, addr: 0x46045dc0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.612710  2925 memory_location.cpp:73] Failed to get NUMA node, addr: 0x46106480, len: 46336: Operation not permitted [1]
+W0827 04:42:57.612715  2924 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59a7ab40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.612880  2922 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5bf25300, len: 185344: Operation not permitted [1]
+W0827 04:42:57.612898  2927 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59e8a9c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.612931  2921 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59fec840, len: 23168: Operation not permitted [1]
+W0827 04:42:57.612948  2923 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5bf52740, len: 185344: Operation not permitted [1]
+W0827 04:42:57.612996  2926 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59edf040, len: 23168: Operation not permitted [1]
+W0827 04:42:57.618498  2975 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a4cda80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.618582  2974 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a3878c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.618582  2973 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a3abc80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.618669  2976 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5c40da00, len: 185344: Operation not permitted [1]
+W0827 04:42:57.618669  2978 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59f66a40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.618741  2977 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5c43ae40, len: 185344: Operation not permitted [1]
+W0827 04:42:57.618854  2979 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59ce6c00, len: 46336: Operation not permitted [1]
+W0827 04:42:57.618898  2980 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a2ebe80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.619125  2982 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a37d0c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.632592  2972 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a371340, len: 23168: Operation not permitted [1]
+W0827 04:42:57.776260  3310 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58ee7080, len: 23168: Operation not permitted [1]
+W0827 04:42:57.776290  3311 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5854f480, len: 23168: Operation not permitted [1]
+W0827 04:42:57.776294  3312 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58edd900, len: 23168: Operation not permitted [1]
+W0827 04:42:57.776240  3309 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58f70e40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.776324  3313 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a63db40, len: 185344: Operation not permitted [1]
+W0827 04:42:57.776337  3314 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a66af80, len: 185344: Operation not permitted [1]
+W0827 04:42:57.776417  3315 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58e5ec40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.776444  3316 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58109800, len: 46336: Operation not permitted [1]
+W0827 04:42:57.776451  3317 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58ed2cc0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.776649  3318 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58316800, len: 23168: Operation not permitted [1]
+W0827 04:42:57.812515  3457 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58919a80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.812533  3458 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5892c840, len: 23168: Operation not permitted [1]
+W0827 04:42:57.812597  3460 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58063380, len: 23168: Operation not permitted [1]
+W0827 04:42:57.812603  3459 memory_location.cpp:73] Failed to get NUMA node, addr: 0x588ed280, len: 23168: Operation not permitted [1]
+W0827 04:42:57.812654  3461 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59fc0080, len: 185344: Operation not permitted [1]
+W0827 04:42:57.812734  3462 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59fed4c0, len: 185344: Operation not permitted [1]
+W0827 04:42:57.812816  3463 memory_location.cpp:73] Failed to get NUMA node, addr: 0x57ebf300, len: 23168: Operation not permitted [1]
+W0827 04:42:57.812877  3464 memory_location.cpp:73] Failed to get NUMA node, addr: 0x44a36bc0, len: 46336: Operation not permitted [1]
+W0827 04:42:57.812991  3466 memory_location.cpp:73] Failed to get NUMA node, addr: 0x44982040, len: 23168: Operation not permitted [1]
+W0827 04:42:57.812996  3465 memory_location.cpp:73] Failed to get NUMA node, addr: 0x588ba900, len: 23168: Operation not permitted [1]
+W0827 04:42:57.851508  3605 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5888adc0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.851550  3606 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58775700, len: 23168: Operation not permitted [1]
+W0827 04:42:57.851619  3607 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58aca800, len: 23168: Operation not permitted [1]
+W0827 04:42:57.851691  3608 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5886ea80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.851799  3609 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5aaacb40, len: 185344: Operation not permitted [1]
+W0827 04:42:57.851871  3611 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58ad7b00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.851959  3610 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5aad9f80, len: 185344: Operation not permitted [1]
+W0827 04:42:57.851996  3612 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58aa8840, len: 46336: Operation not permitted [1]
+W0827 04:42:57.852005  3614 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58ca01c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.852105  3613 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5879a9c0, len: 23168: Operation not permitted [1]
+[2026-08-27 04:43:00] INFO:     Started server process [1]
+[2026-08-27 04:43:00] INFO:     Waiting for application startup.
+[2026-08-27 04:43:00] Calling super().encode with {'add_special_tokens': False}
+[2026-08-27 04:43:00] Calling super().encode with {'add_special_tokens': False}
+[2026-08-27 04:43:00] INFO:     Application startup complete.
+[2026-08-27 04:43:00] INFO:     Uvicorn running on http://0.0.0.0:30000 (Press CTRL+C to quit)
+[2026-08-27 04:43:01] INFO:     174.1.60.1:55082 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:43:01] INFO:     127.0.0.1:38530 - "GET /model_info HTTP/1.1" 200 OK
+[2026-08-27 04:43:02] INFO:     174.1.60.1:55098 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:43:03] INFO:     127.0.0.1:38536 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 04:43:04] INFO:     174.1.60.1:55110 - "GET /health HTTP/1.1" 200 OK
+[2026-08-27 04:43:25] INFO:     174.1.60.1:41170 - "GET /health HTTP/1.1" 200 OK
+[2026-08-27 04:48:11] INFO:     174.1.60.5:38682 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 04:43:01] Start of pd disaggregation warmup ...
+[2026-08-27 04:43:03] Disaggregation warmup requests completed for all 1 DP ranks
+[2026-08-27 04:43:03] End of disaggregation warmup
+[2026-08-27 04:43:03] The server is fired up and ready to roll!
+I0827 04:48:13.997220  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:13.997509  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.013371  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.013677  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.042984  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.043259  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.046793  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.047015  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.057778  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.067556  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.076747  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.095343  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.099277  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.105573  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.105803  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.106056  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.109412  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.109645  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.109865  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.110114  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.110388  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.110597  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.110783  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.114174  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.114495  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.120798  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.136216  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.142400  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.142663  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.153999  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.154243  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.154534  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.154755  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.154990  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.158114  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.158325  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.170076  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.170320  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.170512  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.170779  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.185034  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.197149  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.197443  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.197665  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.197889  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.203205  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.203492  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.203737  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.203975  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.204202  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.215955  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.216298  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.216583  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.216789  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.222101  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.223455  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.232357  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.233336  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.233599  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.233934  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.248603  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.248853  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.249085  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.250525  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.252888  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.253137  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.253366  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.253639  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.256778  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.258342  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.259446  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.259649  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.259908  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.260114  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.260375  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.260576  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.262905  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.263399  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.263654  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.263896  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.264135  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.264392  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.264699  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.264971  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.276409  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.276680  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.280252  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.281527  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.289309  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.289914  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.290587  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.299579  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.299823  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.303650  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.303859  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.304064  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.307004  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.307231  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.307541  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.307801  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.308017  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.308123  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.308235  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.308485  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.308728  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.308988  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.309187  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.309360  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.309405  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.309656  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.309904  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.310132  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.310344  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.312103  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.312309  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.314198  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.315554  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.316985  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.320315  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.327411  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.337924  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.339138  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.344900  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.346967  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.348446  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.361662  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.361877  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.362105  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.365690  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.366845  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.373291  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.373983  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.374544  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.391561  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.391779  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.392053  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.392257  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.392505  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.392726  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.393024  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.393260  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.393772  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.395025  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.396075  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.396298  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.396525  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.396822  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.397025  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.397225  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.403301  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.404704  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.407682  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.417869  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.418104  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.418273  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.421886  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.422092  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.422293  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.422566  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.422824  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.424790  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.427910  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.429622  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.433352  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.433625  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.433885  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.434249  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.434464  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.436031  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.444931  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.445168  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.445376  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.445645  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.448356  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.448545  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.448810  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.449036  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.449236  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.450745  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.453661  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.455060  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.456887  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.457075  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.458276  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.463183  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.464545  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.473290  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.473594  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.473889  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.474606  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.478461  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.478672  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.478849  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.485463  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.485714  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.486359  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.486591  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.486837  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.486868  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.487639  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.492800  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.494125  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.496006  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.496238  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.496584  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.504025  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.504601  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.504830  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.505040  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.505227  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.505416  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.505596  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.505838  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.506070  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.506418  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.506714  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.506938  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.507185  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.507364  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.507618  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.507850  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.508181  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.509447  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.510848  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.510802  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.511059  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.511301  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.512179  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.513973  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.520185  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.521516  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.528406  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.528625  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.528868  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.532759  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.535007  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.535238  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.535440  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.535620  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.536624  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.538493  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.541709  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.541929  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.543128  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.543134  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.543255  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.543396  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.543586  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.543852  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.544047  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.544274  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.544514  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.544595  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.544750  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.544992  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.545264  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.545506  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.545732  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.549163  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.556982  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.557228  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.557428  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.558394  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.559183  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.563414  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.564719  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.564847  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.566068  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.568861  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.574138  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.575349  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.583217  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.583469  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.583667  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.584623  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.588591  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.588935  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.589197  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.590059  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.590425  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.591902  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.599382  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.599529  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.599663  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.599762  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.600073  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.600317  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.600823  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.605185  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.605659  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.605909  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.609126  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.610448  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.611255  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.614173  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.616300  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.616987  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.617319  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.617666  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.620188  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.620463  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.623186  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.623404  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.626320  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.629146  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.632516  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.632800  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.633054  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.634070  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.634683  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.637434  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.640316  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.640353  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.640614  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.641824  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.643349  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.646051  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.658253  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.662520  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.663751  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.663796  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.668447  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.674060  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.683264  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.683470  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.683725  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.684523  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.684583  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.684775  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.685092  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.685344  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.685595  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.685825  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.686066  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.694078  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.694272  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.694502  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.694731  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.696010  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.696235  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.696461  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.697254  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.702469  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.702711  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.702946  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.703189  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.712618  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.716657  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.717875  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.718667  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.718958  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.719236  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.722149  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.724561  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.724784  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.724995  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.730228  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.730405  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.730648  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.730953  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.731205  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.731441  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.736352  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.736553  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.736785  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.737946  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.739337  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.739604  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.743849  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.745306  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.747385  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.750119  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.755378  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.755673  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.755957  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.756151  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.756417  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.756649  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.756873  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.757055  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.757272  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.762418  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.762653  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.767700  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.767921  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.768116  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.769114  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.805779  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.806072  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.806393  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.816823  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.817051  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.817302  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.817519  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.822906  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.823174  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.823410  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.823618  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.828871  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.831584  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.834362  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.834635  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.845114  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.847841  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.853120  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.853320  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.853577  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.853799  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.856580  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.859321  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.859536  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.864822  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.875097  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.895438  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.895658  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.903518  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.903801  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.904052  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.904273  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.909722  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.911178  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.912268  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.912475  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.912720  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.912957  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.913183  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.913419  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.913677  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.913909  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.914156  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.914402  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.914625  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.917008  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.919989  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.920225  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.922858  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.923017  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.935199  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.935660  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.935937  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.939235  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.941046  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.943796  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.944716  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.946557  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.947388  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.949306  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.949542  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.949780  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.950085  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.950371  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.950598  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.950831  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.951049  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.951236  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.951502  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.955791  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.956029  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.964305  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.964589  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.964807  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.965030  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.965714  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.969712  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.971029  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.976545  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.978840  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.982290  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.993181  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.993423  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.993613  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.995042  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.995337  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.995555  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.996627  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.007792  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.008044  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.008275  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.009122  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.013960  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.014259  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.014477  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.014698  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.014875  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.015080  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.015322  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.015535  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.023283  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.024626  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.028729  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.028869  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.029064  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.029294  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.029582  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.030203  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.035056  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.047416  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.047739  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.047941  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.048739  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.054301  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.055744  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.056360  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.056833  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.057034  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.057294  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.057492  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.057713  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.057981  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.058207  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.058401  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.058617  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.069922  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.071324  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.075461  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.081681  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.093765  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.093961  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.094169  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.095338  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.101928  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.102181  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.102407  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.102629  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.102851  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.103041  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.103250  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.103484  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.103757  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.103947  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.104204  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.104463  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.104683  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.104951  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.105253  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.105489  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.105674  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.105890  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.106104  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.106318  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.107211  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.120297  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.124440  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.125627  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.131145  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.131403  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.131698  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.131965  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.132176  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.132416  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.132588  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.132663  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.132896  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.133141  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.133297  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.146152  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.146390  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.146620  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.146828  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.147426  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.156941  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.157151  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.178486  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.178771  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.179106  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.179323  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.208334  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.208683  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.208947  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.209216  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.209455  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.209815  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.210053  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.210210  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.210412  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.210608  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.210822  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.221958  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.222180  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.222395  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.223281  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.225317  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.226866  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.230994  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.237720  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.242835  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.243105  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.243345  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.243551  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.243834  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.244047  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.244271  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.250334  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.250538  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.250738  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.251941  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.257243  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.258617  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.273547  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.277629  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.278937  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.281635  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.281862  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.282078  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.282316  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.282523  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.282783  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.283066  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.283294  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.283550  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.283828  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.284119  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.284336  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.284598  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.285751  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.287382  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.298511  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.298730  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.298920  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.299866  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.301535  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.302943  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.305897  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.306208  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.306404  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.306623  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.306824  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.307075  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.307155  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.307289  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.307511  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.310241  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.310458  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.310705  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.311606  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.313308  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.320852  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.321092  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.321306  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.321498  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.322360  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.323149  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.323376  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.344715  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.362433  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.365234  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.365482  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.365670  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.365831  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.366075  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.366329  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.366532  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.369321  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.379413  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.379633  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.379848  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.380110  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.380908  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.388870  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.389055  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.393224  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.396981  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.397784  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.398291  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.404456  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.405587  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.405781  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.405988  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.416702  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.416770  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.416965  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.417004  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.417227  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.418077  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.419777  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.425088  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.425359  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.425642  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.425822  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.428227  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.428457  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.428719  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.429623  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.431046  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.433805  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.434010  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.441807  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.442006  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.442276  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.447495  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.447705  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.447944  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.453174  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.453361  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.456089  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.456282  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.457403  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.487757  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.487957  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.488181  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.488385  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.488687  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.489209  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.497211  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.497412  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.498611  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.506928  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.507129  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.508323  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.533521  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.533809  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.533984  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.534192  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.534411  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.534649  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.534850  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.535123  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.535336  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.538051  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.553102  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.555838  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.556046  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.556244  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.556519  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.556715  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.556900  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.559612  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.567348  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.570070  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+[2026-08-27 04:48:16] INFO:     174.1.60.5:49356 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 04:48:16] INFO:     174.1.60.5:49370 - "GET /server_info HTTP/1.1" 200 OK
+I0827 04:48:15.570282  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.570564  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.571498  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+[2026-08-27 04:57:13] INFO:     174.1.60.5:34722 - "POST /generate HTTP/1.1" 200 OK
+I0827 04:57:14.461794  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.462142  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.463397  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.464907  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.466337  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.472958  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.474522  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.477468  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.477753  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.479032  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.480643  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.498212  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.498531  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.498838  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.499125  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.499409  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.499795  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.505314  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.517073  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.518641  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.527722  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+[2026-08-27 04:57:14] INFO:     174.1.60.5:34732 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 04:57:14] INFO:     174.1.60.5:34748 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:00:07] INFO:     174.1.60.5:40816 - "POST /generate HTTP/1.1" 200 OK
+I0827 04:57:14.529148  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.534837  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.299574  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.301399  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.307986  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.310060  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.311648  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.317925  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.321113  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.321422  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.322642  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.325122  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.326697  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.332526  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.332682  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.334165  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.340161  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.341799  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.343374  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.345765  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.347450  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.349627  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.350939  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.352005  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.355481  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.355765  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.357254  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.357306  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.358845  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.363112  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.363651  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.368119  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.373224  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.374758  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.379626  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.379845  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.380074  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.381069  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.382846  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.384393  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.386190  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+[2026-08-27 05:04:22] INFO:     174.1.60.5:33422 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:04:22] INFO:     174.1.60.5:33424 - "GET /server_info HTTP/1.1" 200 OK
+I0827 05:04:21.387002  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.387748  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.388473  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.388860  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.399398  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.399649  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.400830  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.402611  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.402901  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.404011  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.405441  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.408586  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.409112  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.414987  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.416474  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.418708  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.420038  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.421981  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.424405  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.437287  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.437530  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.438683  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.443255  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+[2026-08-27 05:04:22 TP0 EP0] Decode batch, #running-req: 1, #full token: 0, full token usage: 0.00, mamba num: 0, mamba usage: 0.00, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 0.03, #queue-req: 0
+I0827 05:04:41.398729  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.399127  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.400347  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.401475  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.401768  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.403021  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.413998  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.415591  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.421803  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+[2026-08-27 05:04:42] INFO:     174.1.60.5:58198 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:05:38] INFO:     174.1.60.5:48352 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:07:16] INFO:     174.1.60.5:48172 - "POST /generate HTTP/1.1" 200 OK
+I0827 05:05:39.576606  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.576977  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.578258  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.600133  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.600483  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.600749  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.601066  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.601655  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.607283  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.615131  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.620574  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.620859  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.621142  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.622053  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.626708  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.148217  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.148566  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.149849  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.155894  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+[2026-08-27 05:07:33] INFO:     174.1.60.5:48616 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:07:40] INFO:     174.1.60.5:48620 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:07:48] INFO:     174.1.60.5:33014 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:07:56] INFO:     174.1.60.5:48046 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:08:03] INFO:     174.1.60.5:43662 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:08:11] INFO:     174.1.60.5:50736 - "POST /generate HTTP/1.1" 200 OK
+I0827 05:08:18.657902  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.659292  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.665387  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.682516  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.682862  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.683116  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.683411  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.683794  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.684041  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.690148  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.690267  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.691737  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.698587  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.699856  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.700106  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.701548  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.704351  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.705502  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.705786  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.706010  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.706919  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.708040  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.713306  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.714720  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.720657  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.722337  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.726626  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+[2026-08-27 05:08:18] INFO:     174.1.60.5:50748 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:08:26] INFO:     174.1.60.5:48502 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:08:26] INFO:     174.1.60.5:48506 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:11:44] INFO:     174.1.60.5:54256 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:11:46] INFO:     174.1.60.5:54260 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:11:46] INFO:     174.1.60.5:54274 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:11:46 TP0 EP0] Decode batch, #running-req: 1, #full token: 0, full token usage: 0.00, mamba num: 0, mamba usage: 0.00, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 0.09, #queue-req: 0
+I0827 05:14:42.491809  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+[2026-08-27 05:14:43] INFO:     174.1.60.5:52586 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:14:43] INFO:     174.1.60.5:52600 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:14:43] INFO:     174.1.60.5:52616 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:17:40] INFO:     174.1.60.5:47526 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:17:47] INFO:     174.1.60.5:33870 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:17:55] INFO:     174.1.60.5:33666 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:18:02] INFO:     174.1.60.5:59970 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:18:10] INFO:     174.1.60.5:59984 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:18:18] INFO:     174.1.60.5:35200 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:18:25] INFO:     174.1.60.5:47324 - "POST /generate HTTP/1.1" 200 OK
+I0827 05:14:42.493278  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.499468  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.158722  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.159102  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.160380  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+[2026-08-27 05:18:33] INFO:     174.1.60.5:60276 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:18:40] INFO:     174.1.60.5:60282 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:18:40] INFO:     174.1.60.5:60286 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:21:37] INFO:     174.1.60.5:39028 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:37] INFO:     174.1.60.5:39042 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:37] INFO:     174.1.60.5:39052 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:37] INFO:     174.1.60.5:39066 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:37] INFO:     174.1.60.5:39070 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:37] INFO:     174.1.60.5:39084 - "POST /generate HTTP/1.1" 200 OK
+I0827 05:18:33.167616  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.169173  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.173859  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.185045  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.185313  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.186507  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.192727  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+[2026-08-27 05:21:37] INFO:     174.1.60.5:39092 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:37] INFO:     174.1.60.5:39102 - "POST /generate HTTP/1.1" 200 OK
+I0827 05:21:47.564407  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:47.575759  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:47.585628  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:47.597601  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:47.599035  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:48.554785  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:48.565164  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:49.333541  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+[2026-08-27 05:21:51] INFO:     174.1.60.5:33930 - "POST /generate HTTP/1.1" 200 OK
+I0827 05:21:49.343402  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:49.352372  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:49.361497  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:50.265654  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:50.274874  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:50.283715  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.110808  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.124187  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.162127  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.173472  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.005925  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.016026  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.024489  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+[2026-08-27 05:21:53] INFO:     174.1.60.5:33942 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:55] INFO:     174.1.60.5:33946 - "POST /generate HTTP/1.1" 200 OK
+I0827 05:21:52.033867  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.856267  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.867897  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:53.712086  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:53.724234  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:54.551172  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:54.564998  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:55.326069  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:55.338083  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:56.186933  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:56.188442  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:56.200929  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:57.056439  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:57.056681  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:57.058399  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+[2026-08-27 05:21:57] INFO:     174.1.60.5:33950 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:57] INFO:     174.1.60.5:33960 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:57] INFO:     174.1.60.5:33968 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:59] INFO:     174.1.60.5:33980 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:00] INFO:     174.1.60.5:33990 - "POST /generate HTTP/1.1" 200 OK
+I0827 05:21:57.654704  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+[2026-08-27 05:22:11] INFO:     174.1.60.5:51774 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:13] INFO:     174.1.60.5:51778 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:15] INFO:     174.1.60.5:51794 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:17] INFO:     174.1.60.5:51800 - "POST /generate HTTP/1.1" 200 OK
+I0827 05:22:05.685521  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:05.687002  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:06.720834  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:07.458693  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:08.386162  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:08.387703  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:08.395984  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:10.976925  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:16.233013  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:17.990159  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:17.991626  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:17.999158  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:18.227823  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+[2026-08-27 05:22:18] INFO:     174.1.60.5:51812 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:18] INFO:     174.1.60.5:51824 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:20] INFO:     174.1.60.5:51834 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:21] INFO:     174.1.60.5:46458 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:31] INFO:     174.1.60.5:36970 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:34] INFO:     174.1.60.5:36980 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:35] INFO:     174.1.60.5:36990 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:37] INFO:     174.1.60.5:36996 - "POST /generate HTTP/1.1" 200 OK
+I0827 05:22:18.250644  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:18.360224  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:28.065888  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:36.003260  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:36.977586  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:38.816133  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+[2026-08-27 05:22:38] INFO:     174.1.60.5:37004 - "POST /generate HTTP/1.1" 200 OK
+I0827 05:22:38.914274  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+[2026-08-27 05:22:39] INFO:     174.1.60.5:37012 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:40] INFO:     174.1.60.5:37020 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:42] INFO:     174.1.60.5:36076 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:52] INFO:     174.1.60.5:50774 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:54] INFO:     174.1.60.5:50786 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:56] INFO:     174.1.60.5:50788 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:58] INFO:     174.1.60.5:50790 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:59] INFO:     174.1.60.5:50806 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:23:00] INFO:     174.1.60.5:50818 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:23:01] INFO:     174.1.60.5:49800 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:23:03] INFO:     174.1.60.5:49808 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:23:20] INFO:     174.1.60.5:35588 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:23:20] INFO:     174.1.60.5:35590 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:32:54] INFO:     174.1.60.1:40486 - "GET /health HTTP/1.1" 200 OK
+[2026-08-27 05:35:46] INFO:     174.1.60.5:41898 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:36:14] INFO:     174.1.60.5:52884 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:36:14] INFO:     174.1.60.5:52888 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:36:36] INFO:     174.1.60.1:41926 - "GET /health HTTP/1.1" 200 OK
+I0827 05:22:39.463928  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+[2026-08-27 05:35:55 TP0 EP0] Decode batch, #running-req: 1, #full token: 16896, full token usage: 0.04, mamba num: 1, mamba usage: 0.01, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 0.03, #queue-req: 0
+[2026-08-27 05:35:56 TP0 EP0] Decode batch, #running-req: 1, #full token: 16960, full token usage: 0.04, mamba num: 1, mamba usage: 0.01, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 25.04, #queue-req: 0
+[2026-08-27 05:35:58 TP0 EP0] Decode batch, #running-req: 1, #full token: 16960, full token usage: 0.04, mamba num: 1, mamba usage: 0.01, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 25.06, #queue-req: 0
+[2026-08-27 05:36:00 TP0 EP0] Decode batch, #running-req: 1, #full token: 17024, full token usage: 0.04, mamba num: 1, mamba usage: 0.01, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 25.08, #queue-req: 0
+[2026-08-27 05:36:01 TP0 EP0] Decode batch, #running-req: 1, #full token: 17088, full token usage: 0.04, mamba num: 1, mamba usage: 0.01, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 25.07, #queue-req: 0
+[2026-08-27 05:36:03 TP0 EP0] Decode batch, #running-req: 1, #full token: 17088, full token usage: 0.04, mamba num: 1, mamba usage: 0.01, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 25.83, #queue-req: 0
+[2026-08-27 05:36:04 TP0 EP0] Decode batch, #running-req: 1, #full token: 17152, full token usage: 0.04, mamba num: 1, mamba usage: 0.01, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 26.19, #queue-req: 0
+[2026-08-27 05:36:06 TP0 EP0] Decode batch, #running-req: 1, #full token: 17152, full token usage: 0.04, mamba num: 1, mamba usage: 0.01, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 26.20, #queue-req: 0
+[2026-08-27 05:36:07 TP0 EP0] Decode batch, #running-req: 1, #full token: 17216, full token usage: 0.04, mamba num: 1, mamba usage: 0.01, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 26.31, #queue-req: 0
+[2026-08-27 05:36:09 TP0 EP0] Decode batch, #running-req: 1, #full token: 17280, full token usage: 0.04, mamba num: 1, mamba usage: 0.01, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 26.34, #queue-req: 0
+[2026-08-27 05:36:10 TP0 EP0] Decode batch, #running-req: 1, #full token: 17280, full token usage: 0.04, mamba num: 1, mamba usage: 0.01, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 26.41, #queue-req: 0
+[2026-08-27 05:36:12 TP0 EP0] Decode batch, #running-req: 1, #full token: 17344, full token usage: 0.04, mamba num: 1, mamba usage: 0.01, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 26.42, #queue-req: 0
+[2026-08-27 05:36:13 TP0 EP0] Decode batch, #running-req: 1, #full token: 17408, full token usage: 0.04, mamba num: 1, mamba usage: 0.01, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 26.41, #queue-req: 0
+[2026-08-27 05:36:36] INFO:     174.1.60.5:41604 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:42:24] INFO:     174.1.60.5:58526 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:42:24] INFO:     174.1.60.5:58530 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:42:24] INFO:     174.1.60.5:58544 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:42:24] INFO:     174.1.60.5:58546 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:42:24] INFO:     174.1.60.5:58554 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:42:24] INFO:     174.1.60.5:58560 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:42:24] INFO:     174.1.60.5:58562 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:42:24] INFO:     174.1.60.5:58578 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:42:40 TP0 EP0] Decode batch, #running-req: 1, #full token: 135040, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.27, #prealloc-req: 0, #transfer-req: 7, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 0.10, #queue-req: 0
+[2026-08-27 05:42:42 TP0 EP0] Decode batch, #running-req: 2, #full token: 135104, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.23, #prealloc-req: 0, #transfer-req: 6, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 36.35, #queue-req: 0
+[2026-08-27 05:42:44 TP0 EP0] Decode batch, #running-req: 4, #full token: 135168, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.15, #prealloc-req: 0, #transfer-req: 4, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 53.06, #queue-req: 0
+[2026-08-27 05:42:46 TP0 EP0] Decode batch, #running-req: 8, #full token: 135616, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 99.41, #queue-req: 0
+[2026-08-27 05:43:11] INFO:     174.1.60.5:40798 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:42:49 TP0 EP0] Decode batch, #running-req: 8, #full token: 135808, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 113.77, #queue-req: 0
+[2026-08-27 05:42:52 TP0 EP0] Decode batch, #running-req: 8, #full token: 136192, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 114.49, #queue-req: 0
+[2026-08-27 05:42:55 TP0 EP0] Decode batch, #running-req: 8, #full token: 136576, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 116.05, #queue-req: 0
+[2026-08-27 05:42:58 TP0 EP0] Decode batch, #running-req: 8, #full token: 136768, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 117.94, #queue-req: 0
+[2026-08-27 05:43:00 TP0 EP0] Decode batch, #running-req: 8, #full token: 137152, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 118.51, #queue-req: 0
+[2026-08-27 05:43:03 TP0 EP0] Decode batch, #running-req: 8, #full token: 137408, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 120.72, #queue-req: 0
+[2026-08-27 05:43:05 TP0 EP0] Decode batch, #running-req: 8, #full token: 137728, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 122.88, #queue-req: 0
+[2026-08-27 05:43:08 TP0 EP0] Decode batch, #running-req: 8, #full token: 138176, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 123.83, #queue-req: 0
+[2026-08-27 05:43:13] INFO:     174.1.60.5:57674 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:43:16] INFO:     174.1.60.5:57684 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:43:18] INFO:     174.1.60.5:57700 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:43:18] INFO:     174.1.60.5:57706 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:43:18] INFO:     174.1.60.5:57714 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:43:11 TP0 EP0] Decode batch, #running-req: 8, #full token: 120960, full token usage: 0.27, mamba num: 7, mamba usage: 0.04, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 123.57, #queue-req: 0
+[2026-08-27 05:43:13 TP0 EP0] Decode batch, #running-req: 7, #full token: 138176, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.04, #prealloc-req: 0, #transfer-req: 1, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 106.75, #queue-req: 0
+[2026-08-27 05:43:16 TP0 EP0] Decode batch, #running-req: 6, #full token: 120512, full token usage: 0.27, mamba num: 7, mamba usage: 0.04, pre-allocated usage: 0.08, #prealloc-req: 0, #transfer-req: 2, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 93.83, #queue-req: 0
+[2026-08-27 05:43:18 TP0 EP0] Decode batch, #running-req: 4, #full token: 136896, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.15, #prealloc-req: 0, #transfer-req: 4, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 78.56, #queue-req: 0
+[2026-08-27 05:43:18] INFO:     174.1.60.5:57734 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:43:19] INFO:     174.1.60.5:57722 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:43:22 TP0 EP0] Decode batch, #running-req: 1, #full token: 134400, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.27, #prealloc-req: 0, #transfer-req: 7, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 13.68, #queue-req: 0
+[2026-08-27 05:43:23 TP0 EP0] Decode batch, #running-req: 1, #full token: 134400, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.27, #prealloc-req: 0, #transfer-req: 7, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 26.33, #queue-req: 0
+[2026-08-27 05:43:25 TP0 EP0] Decode batch, #running-req: 1, #full token: 134464, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.27, #prealloc-req: 0, #transfer-req: 7, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 26.29, #queue-req: 0
+[2026-08-27 05:43:26 TP0 EP0] Decode batch, #running-req: 1, #full token: 134464, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.27, #prealloc-req: 0, #transfer-req: 7, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 26.29, #queue-req: 0
+[2026-08-27 05:43:28 TP0 EP0] Decode batch, #running-req: 1, #full token: 134528, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.27, #prealloc-req: 0, #transfer-req: 7, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 26.27, #queue-req: 0
+[2026-08-27 05:43:42] INFO:     174.1.60.5:33074 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:44:01] INFO:     174.1.60.5:45114 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:43:29 TP0 EP0] Decode batch, #running-req: 1, #full token: 134592, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.27, #prealloc-req: 0, #transfer-req: 7, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 26.29, #queue-req: 0
+[2026-08-27 05:43:31 TP0 EP0] Decode batch, #running-req: 1, #full token: 134592, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.27, #prealloc-req: 0, #transfer-req: 7, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 26.27, #queue-req: 0
+[2026-08-27 05:43:32 TP0 EP0] Decode batch, #running-req: 2, #full token: 134720, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.23, #prealloc-req: 0, #transfer-req: 6, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 40.74, #queue-req: 0
+[2026-08-27 05:43:34 TP0 EP0] Decode batch, #running-req: 2, #full token: 134784, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.23, #prealloc-req: 0, #transfer-req: 6, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 47.93, #queue-req: 0
+[2026-08-27 05:43:36 TP0 EP0] Decode batch, #running-req: 3, #full token: 134912, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.19, #prealloc-req: 0, #transfer-req: 5, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 59.63, #queue-req: 0
+[2026-08-27 05:43:38 TP0 EP0] Decode batch, #running-req: 4, #full token: 135104, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.15, #prealloc-req: 0, #transfer-req: 4, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 79.43, #queue-req: 0
+[2026-08-27 05:43:40 TP0 EP0] Decode batch, #running-req: 8, #full token: 135168, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 91.87, #queue-req: 0
+[2026-08-27 05:43:43 TP0 EP0] Decode batch, #running-req: 7, #full token: 134976, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.04, #prealloc-req: 0, #transfer-req: 1, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 121.94, #queue-req: 0
+[2026-08-27 05:43:45 TP0 EP0] Decode batch, #running-req: 7, #full token: 135360, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.04, #prealloc-req: 0, #transfer-req: 1, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 108.10, #queue-req: 0
+[2026-08-27 05:43:48 TP0 EP0] Decode batch, #running-req: 7, #full token: 135552, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.04, #prealloc-req: 0, #transfer-req: 1, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 108.10, #queue-req: 0
+[2026-08-27 05:43:50 TP0 EP0] Decode batch, #running-req: 8, #full token: 135872, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 111.27, #queue-req: 0
+[2026-08-27 05:43:53 TP0 EP0] Decode batch, #running-req: 8, #full token: 136128, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 124.23, #queue-req: 0
+[2026-08-27 05:43:55 TP0 EP0] Decode batch, #running-req: 8, #full token: 136512, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 124.71, #queue-req: 0
+[2026-08-27 05:43:58 TP0 EP0] Decode batch, #running-req: 8, #full token: 136832, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 124.62, #queue-req: 0
+[2026-08-27 05:44:01 TP0 EP0] Decode batch, #running-req: 8, #full token: 137088, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 124.28, #queue-req: 0
+[2026-08-27 05:44:03 TP0 EP0] Decode batch, #running-req: 7, #full token: 136704, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.04, #prealloc-req: 0, #transfer-req: 1, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 109.28, #queue-req: 0
+[2026-08-27 05:44:06 TP0 EP0] Decode batch, #running-req: 7, #full token: 137088, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.04, #prealloc-req: 0, #transfer-req: 1, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 107.23, #queue-req: 0
+[2026-08-27 05:44:06] INFO:     174.1.60.5:45118 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:44:08] INFO:     174.1.60.5:45120 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:44:13] INFO:     174.1.60.5:48824 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:44:13] INFO:     174.1.60.5:48838 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:44:13] INFO:     174.1.60.5:48864 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:44:13] INFO:     174.1.60.5:48848 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:44:19] INFO:     174.1.60.5:48872 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:44:08 TP0 EP0] Decode batch, #running-req: 6, #full token: 119296, full token usage: 0.27, mamba num: 7, mamba usage: 0.04, pre-allocated usage: 0.08, #prealloc-req: 0, #transfer-req: 2, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 90.81, #queue-req: 0
+[2026-08-27 05:44:11 TP0 EP0] Decode batch, #running-req: 6, #full token: 136448, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.08, #prealloc-req: 0, #transfer-req: 2, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 93.69, #queue-req: 0
+[2026-08-27 05:44:13 TP0 EP0] Decode batch, #running-req: 2, #full token: 134272, full token usage: 0.30, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.23, #prealloc-req: 0, #transfer-req: 6, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 78.18, #queue-req: 0
+[2026-08-27 05:44:15 TP0 EP0] Decode batch, #running-req: 2, #full token: 134336, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.23, #prealloc-req: 0, #transfer-req: 6, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 48.09, #queue-req: 0
+[2026-08-27 05:44:17 TP0 EP0] Decode batch, #running-req: 2, #full token: 134464, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.23, #prealloc-req: 0, #transfer-req: 6, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 48.12, #queue-req: 0
+[2026-08-27 05:44:18 TP0 EP0] Decode batch, #running-req: 2, #full token: 134528, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.23, #prealloc-req: 0, #transfer-req: 6, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 48.12, #queue-req: 0
+[2026-08-27 05:44:20 TP0 EP0] Decode batch, #running-req: 2, #full token: 134144, full token usage: 0.30, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.23, #prealloc-req: 0, #transfer-req: 6, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 43.53, #queue-req: 0
+[2026-08-27 05:44:22 TP0 EP0] Decode batch, #running-req: 2, #full token: 134208, full token usage: 0.30, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.23, #prealloc-req: 0, #transfer-req: 6, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 47.97, #queue-req: 0
+[2026-08-27 05:44:23 TP0 EP0] Decode batch, #running-req: 2, #full token: 134336, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.23, #prealloc-req: 0, #transfer-req: 6, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 48.02, #queue-req: 0
+[2026-08-27 05:44:25 TP0 EP0] Decode batch, #running-req: 2, #full token: 134400, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.23, #prealloc-req: 0, #transfer-req: 6, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 48.03, #queue-req: 0
+[2026-08-27 05:44:27 TP0 EP0] Decode batch, #running-req: 2, #full token: 134464, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.23, #prealloc-req: 0, #transfer-req: 6, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 48.09, #queue-req: 0
+[2026-08-27 05:44:28 TP0 EP0] Decode batch, #running-req: 2, #full token: 134528, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.23, #prealloc-req: 0, #transfer-req: 6, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 48.07, #queue-req: 0
+[2026-08-27 05:44:30 TP0 EP0] Decode batch, #running-req: 3, #full token: 134656, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.19, #prealloc-req: 0, #transfer-req: 5, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 52.87, #queue-req: 0
+[2026-08-27 05:44:32] INFO:     174.1.60.5:46810 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:44:45] INFO:     174.1.60.5:60838 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:45:00] INFO:     174.1.60.5:53994 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:44:32 TP0 EP0] Decode batch, #running-req: 2, #full token: 134400, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.23, #prealloc-req: 0, #transfer-req: 6, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 60.67, #queue-req: 0
+[2026-08-27 05:44:34 TP0 EP0] Decode batch, #running-req: 2, #full token: 134400, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.23, #prealloc-req: 0, #transfer-req: 6, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 48.06, #queue-req: 0
+[2026-08-27 05:44:35 TP0 EP0] Decode batch, #running-req: 3, #full token: 134528, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.19, #prealloc-req: 0, #transfer-req: 5, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 57.01, #queue-req: 0
+I0827 05:44:36.401998  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+[2026-08-27 05:44:38 TP0 EP0] Decode batch, #running-req: 6, #full token: 134784, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.08, #prealloc-req: 0, #transfer-req: 2, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 87.24, #queue-req: 0
+[2026-08-27 05:44:40 TP0 EP0] Decode batch, #running-req: 7, #full token: 135104, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.04, #prealloc-req: 0, #transfer-req: 1, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 107.12, #queue-req: 0
+[2026-08-27 05:44:43 TP0 EP0] Decode batch, #running-req: 7, #full token: 135296, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.04, #prealloc-req: 0, #transfer-req: 1, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 107.40, #queue-req: 0
+[2026-08-27 05:44:46 TP0 EP0] Decode batch, #running-req: 7, #full token: 134976, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.04, #prealloc-req: 0, #transfer-req: 1, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 116.53, #queue-req: 0
+[2026-08-27 05:44:48 TP0 EP0] Decode batch, #running-req: 7, #full token: 135232, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.04, #prealloc-req: 0, #transfer-req: 1, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 107.72, #queue-req: 0
+[2026-08-27 05:44:51 TP0 EP0] Decode batch, #running-req: 7, #full token: 135616, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.04, #prealloc-req: 0, #transfer-req: 1, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 107.60, #queue-req: 0
+[2026-08-27 05:44:53 TP0 EP0] Decode batch, #running-req: 8, #full token: 135872, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 115.45, #queue-req: 0
+[2026-08-27 05:44:56 TP0 EP0] Decode batch, #running-req: 8, #full token: 136192, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 124.20, #queue-req: 0
+[2026-08-27 05:44:59 TP0 EP0] Decode batch, #running-req: 8, #full token: 136512, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 124.15, #queue-req: 0
+[2026-08-27 05:45:01 TP0 EP0] Decode batch, #running-req: 7, #full token: 136384, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.04, #prealloc-req: 0, #transfer-req: 1, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 115.66, #queue-req: 0
+[2026-08-27 05:45:07] INFO:     174.1.60.5:53288 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:45:09] INFO:     174.1.60.5:53292 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:45:09] INFO:     174.1.60.5:53306 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:45:09] INFO:     174.1.60.5:53316 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:45:11] INFO:     174.1.60.5:48244 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:45:04 TP0 EP0] Decode batch, #running-req: 7, #full token: 136576, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.04, #prealloc-req: 0, #transfer-req: 1, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 107.73, #queue-req: 0
+[2026-08-27 05:45:06 TP0 EP0] Decode batch, #running-req: 7, #full token: 136832, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.04, #prealloc-req: 0, #transfer-req: 1, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 107.71, #queue-req: 0
+[2026-08-27 05:45:09 TP0 EP0] Decode batch, #running-req: 7, #full token: 102080, full token usage: 0.23, mamba num: 6, mamba usage: 0.03, pre-allocated usage: 0.04, #prealloc-req: 0, #transfer-req: 1, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 101.40, #queue-req: 0
+[2026-08-27 05:45:11 TP0 EP0] Decode batch, #running-req: 3, #full token: 135360, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.19, #prealloc-req: 0, #transfer-req: 5, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 80.53, #queue-req: 0
+[2026-08-27 05:45:14] INFO:     174.1.60.5:48254 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:45:13 TP0 EP0] Decode batch, #running-req: 3, #full token: 135360, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.19, #prealloc-req: 0, #transfer-req: 5, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 63.12, #queue-req: 0
+[2026-08-27 05:45:15 TP0 EP0] Decode batch, #running-req: 2, #full token: 135040, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.23, #prealloc-req: 0, #transfer-req: 6, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 60.56, #queue-req: 0
+[2026-08-27 05:45:16 TP0 EP0] Decode batch, #running-req: 2, #full token: 135168, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.23, #prealloc-req: 0, #transfer-req: 6, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 48.20, #queue-req: 0
+[2026-08-27 05:45:18 TP0 EP0] Decode batch, #running-req: 3, #full token: 135168, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.19, #prealloc-req: 0, #transfer-req: 5, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 51.84, #queue-req: 0
+[2026-08-27 05:45:20 TP0 EP0] Decode batch, #running-req: 3, #full token: 135360, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.19, #prealloc-req: 0, #transfer-req: 5, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 63.13, #queue-req: 0
+[2026-08-27 05:45:21] INFO:     174.1.60.5:41400 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:45:22 TP0 EP0] Decode batch, #running-req: 2, #full token: 134976, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.23, #prealloc-req: 0, #transfer-req: 6, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 52.84, #queue-req: 0
+[2026-08-27 05:45:23 TP0 EP0] Decode batch, #running-req: 2, #full token: 135104, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.23, #prealloc-req: 0, #transfer-req: 6, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 48.14, #queue-req: 0
+[2026-08-27 05:45:25 TP0 EP0] Decode batch, #running-req: 2, #full token: 135168, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.23, #prealloc-req: 0, #transfer-req: 6, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 48.14, #queue-req: 0
+[2026-08-27 05:45:31] INFO:     174.1.60.5:57408 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:45:44] INFO:     174.1.60.5:53926 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:46:01] INFO:     174.1.60.5:58032 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:46:02] INFO:     174.1.60.5:58044 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:46:04] INFO:     174.1.60.5:58048 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:45:27 TP0 EP0] Decode batch, #running-req: 2, #full token: 135232, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.23, #prealloc-req: 0, #transfer-req: 6, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 47.97, #queue-req: 0
+[2026-08-27 05:45:28 TP0 EP0] Decode batch, #running-req: 2, #full token: 135360, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.23, #prealloc-req: 0, #transfer-req: 6, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 47.96, #queue-req: 0
+[2026-08-27 05:45:30 TP0 EP0] Decode batch, #running-req: 4, #full token: 135424, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.15, #prealloc-req: 0, #transfer-req: 4, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 66.98, #queue-req: 0
+[2026-08-27 05:45:32 TP0 EP0] Decode batch, #running-req: 5, #full token: 134976, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.11, #prealloc-req: 0, #transfer-req: 3, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 79.27, #queue-req: 0
+[2026-08-27 05:45:35 TP0 EP0] Decode batch, #running-req: 6, #full token: 135232, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.08, #prealloc-req: 0, #transfer-req: 2, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 80.46, #queue-req: 0
+[2026-08-27 05:45:38 TP0 EP0] Decode batch, #running-req: 6, #full token: 135424, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.08, #prealloc-req: 0, #transfer-req: 2, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 92.05, #queue-req: 0
+[2026-08-27 05:45:40 TP0 EP0] Decode batch, #running-req: 7, #full token: 135680, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.04, #prealloc-req: 0, #transfer-req: 1, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 99.21, #queue-req: 0
+[2026-08-27 05:45:43 TP0 EP0] Decode batch, #running-req: 7, #full token: 135936, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.04, #prealloc-req: 0, #transfer-req: 1, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 107.21, #queue-req: 0
+[2026-08-27 05:45:45 TP0 EP0] Decode batch, #running-req: 7, #full token: 135744, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.04, #prealloc-req: 0, #transfer-req: 1, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 111.83, #queue-req: 0
+[2026-08-27 05:45:48 TP0 EP0] Decode batch, #running-req: 7, #full token: 136064, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.04, #prealloc-req: 0, #transfer-req: 1, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 107.18, #queue-req: 0
+[2026-08-27 05:45:51 TP0 EP0] Decode batch, #running-req: 7, #full token: 136384, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.04, #prealloc-req: 0, #transfer-req: 1, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 106.89, #queue-req: 0
+[2026-08-27 05:45:53 TP0 EP0] Decode batch, #running-req: 8, #full token: 136640, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 115.54, #queue-req: 0
+[2026-08-27 05:45:56 TP0 EP0] Decode batch, #running-req: 8, #full token: 137024, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 123.48, #queue-req: 0
+[2026-08-27 05:45:58 TP0 EP0] Decode batch, #running-req: 8, #full token: 137280, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 123.53, #queue-req: 0
+[2026-08-27 05:46:01 TP0 EP0] Decode batch, #running-req: 8, #full token: 137600, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 123.92, #queue-req: 0
+[2026-08-27 05:46:04 TP0 EP0] Decode batch, #running-req: 6, #full token: 136896, full token usage: 0.31, mamba num: 8, mamba usage: 0.04, pre-allocated usage: 0.08, #prealloc-req: 0, #transfer-req: 2, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 100.02, #queue-req: 0
+[2026-08-27 05:46:06 TP0 EP0] Decode batch, #running-req: 4, #full token: 119040, full token usage: 0.27, mamba num: 7, mamba usage: 0.04, pre-allocated usage: 0.11, #prealloc-req: 0, #transfer-req: 3, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 81.96, #queue-req: 0
+[2026-08-27 05:46:08 TP0 EP0] Decode batch, #running-req: 3, #full token: 101824, full token usage: 0.23, mamba num: 6, mamba usage: 0.03, pre-allocated usage: 0.11, #prealloc-req: 0, #transfer-req: 3, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 75.43, #queue-req: 0
+[2026-08-27 05:46:10 TP0 EP0] Decode batch, #running-req: 3, #full token: 101952, full token usage: 0.23, mamba num: 6, mamba usage: 0.03, pre-allocated usage: 0.11, #prealloc-req: 0, #transfer-req: 3, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 63.25, #queue-req: 0
+[2026-08-27 05:46:11 TP0 EP0] Decode batch, #running-req: 2, #full token: 84800, full token usage: 0.19, mamba num: 5, mamba usage: 0.03, pre-allocated usage: 0.11, #prealloc-req: 0, #transfer-req: 3, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 53.37, #queue-req: 0
+[2026-08-27 05:46:13 TP0 EP0] Decode batch, #running-req: 3, #full token: 84864, full token usage: 0.19, mamba num: 5, mamba usage: 0.03, pre-allocated usage: 0.08, #prealloc-req: 0, #transfer-req: 2, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 56.43, #queue-req: 0
+[2026-08-27 05:46:15 TP0 EP0] Decode batch, #running-req: 2, #full token: 67648, full token usage: 0.15, mamba num: 4, mamba usage: 0.02, pre-allocated usage: 0.08, #prealloc-req: 0, #transfer-req: 2, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 49.36, #queue-req: 0
+[2026-08-27 05:46:17 TP0 EP0] Decode batch, #running-req: 2, #full token: 67712, full token usage: 0.15, mamba num: 4, mamba usage: 0.02, pre-allocated usage: 0.08, #prealloc-req: 0, #transfer-req: 2, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 48.17, #queue-req: 0
+[2026-08-27 05:46:18 TP0 EP0] Decode batch, #running-req: 4, #full token: 67840, full token usage: 0.15, mamba num: 4, mamba usage: 0.02, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 58.51, #queue-req: 0
+[2026-08-27 05:46:20 TP0 EP0] Decode batch, #running-req: 3, #full token: 50752, full token usage: 0.12, mamba num: 3, mamba usage: 0.02, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 69.21, #queue-req: 0
+[2026-08-27 05:46:22 TP0 EP0] Decode batch, #running-req: 3, #full token: 50880, full token usage: 0.12, mamba num: 3, mamba usage: 0.02, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 63.44, #queue-req: 0
+[2026-08-27 05:46:24 TP0 EP0] Decode batch, #running-req: 3, #full token: 50944, full token usage: 0.12, mamba num: 3, mamba usage: 0.02, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 63.55, #queue-req: 0
+[2026-08-27 05:46:26 TP0 EP0] Decode batch, #running-req: 3, #full token: 51072, full token usage: 0.12, mamba num: 3, mamba usage: 0.02, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 63.54, #queue-req: 0
+[2026-08-27 05:46:28 TP0 EP0] Decode batch, #running-req: 3, #full token: 51136, full token usage: 0.12, mamba num: 3, mamba usage: 0.02, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 63.46, #queue-req: 0
+[2026-08-27 05:46:30 TP0 EP0] Decode batch, #running-req: 3, #full token: 51328, full token usage: 0.12, mamba num: 3, mamba usage: 0.02, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 63.48, #queue-req: 0
+[2026-08-27 05:46:32 TP0 EP0] Decode batch, #running-req: 3, #full token: 51456, full token usage: 0.12, mamba num: 3, mamba usage: 0.02, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 63.53, #queue-req: 0
+[2026-08-27 05:46:33 TP0 EP0] Decode batch, #running-req: 3, #full token: 51520, full token usage: 0.12, mamba num: 3, mamba usage: 0.02, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 63.47, #queue-req: 0
+[2026-08-27 05:46:35 TP0 EP0] Decode batch, #running-req: 3, #full token: 51712, full token usage: 0.12, mamba num: 3, mamba usage: 0.02, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 63.36, #queue-req: 0
+[2026-08-27 05:46:41] INFO:     174.1.60.5:33486 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:46:41] INFO:     174.1.60.5:33500 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:46:37 TP0 EP0] Decode batch, #running-req: 2, #full token: 34560, full token usage: 0.08, mamba num: 2, mamba usage: 0.01, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 52.83, #queue-req: 0
+[2026-08-27 05:46:39 TP0 EP0] Decode batch, #running-req: 2, #full token: 34560, full token usage: 0.08, mamba num: 2, mamba usage: 0.01, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 48.26, #queue-req: 0
+[2026-08-27 05:46:40 TP0 EP0] Decode batch, #running-req: 2, #full token: 34688, full token usage: 0.08, mamba num: 2, mamba usage: 0.01, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 48.26, #queue-req: 0
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-c8-20260827-1340/logs/manual_d_node1.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-c8-20260827-1340/logs/manual_d_node1.log
new file mode 100644
index 0000000..4ae8fab
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-c8-20260827-1340/logs/manual_d_node1.log
@@ -0,0 +1,1458 @@
+/usr/local/lib/python3.12/dist-packages/torchao/quantization/quant_api.py:1731: SyntaxWarning: invalid escape sequence '\.'
+  """Configuration class for applying different quantization configs to modules or parameters based on their fully qualified names (FQNs).
+/sgl-workspace/sglang/python/sglang/launch_server.py:56: UserWarning: 'python -m sglang.launch_server' is still supported, but 'sglang serve' is the recommended entrypoint.
+  Example: sglang serve --model-path <model> [options]
+  warnings.warn(
+[2026-08-27 04:35:58] KV cache is forced as chunk cache for decode server
+[2026-08-27 04:36:02] Breakable CUDA graph is incompatible with MLA attention; disabling prefill CUDA graph.
+[2026-08-27 04:36:02] Attention backend not specified. Use triton backend by default.
+[2026-08-27 04:36:05] server_args=ServerArgs(model_path='/data/hf_models/Kimi-K3', tokenizer_path='/data/hf_models/Kimi-K3', tokenizer_mode='auto', tokenizer_backend='huggingface', tokenizer_worker_num=1, detokenizer_worker_num=1, skip_tokenizer_init=False, load_format='auto', model_loader_extra_config='{}', trust_remote_code=True, context_length=None, is_embedding=False, enable_multimodal=None, revision=None, model_impl='auto', model_config_parser='auto', json_model_override_args='{}', dtype='auto', quantization=None, quantization_param_path=None, kv_cache_dtype='auto', enable_fp32_lm_head=False, modelopt_quant=None, modelopt_checkpoint_restore_path=None, modelopt_checkpoint_save_path=None, modelopt_export_path=None, quantize_and_serve=False, rl_quant_profile=None, enable_tf32_matmul=False, mem_fraction_static=0.88, max_running_requests=None, max_queued_requests=None, max_total_tokens=None, chunked_prefill_size=8192, enable_dynamic_chunking=False, max_prefill_tokens=16384, prefill_max_requests=None, schedule_policy='fcfs', enable_priority_scheduling=False, disable_priority_preemption=False, default_priority_value=None, abort_on_priority_when_disabled=False, schedule_low_priority_values_first=False, priority_scheduling_preemption_threshold=10, retraction_policy='length', schedule_conservativeness=1.0, page_size=64, swa_full_tokens_ratio=0.8, disable_hybrid_swa_memory=False, radix_eviction_policy='lru', prefill_only_disable_kv_cache=False, disable_radix_cache=True, enable_page_major_kv_layout=False, enable_unified_memory=False, disable_chunked_prefix_cache=False, disable_overlap_schedule=False, num_continuous_decode_steps=1, scheduler_recv_interval=1, enable_mixed_chunk=False, nccl_port=None, dist_timeout=3600, dist_init_addr='174.1.60.5:20000', nnodes=4, node_rank=1, tp_size=32, dcp_size=1, pp_size=1, pp_max_micro_batch_size=None, pp_async_batch_depth=0, dp_size=1, load_balance_method='round_robin', attn_cp_size=1, moe_dp_size=1, dcp_comm_backend='ag_rs', dcp_replicate_q_proj=None, dwdp_size=1, enable_prefill_cp=False, cp_strategy=None, enable_dsa_cache_layer_split=False, enable_dsa_prefill_context_parallel=False, dsa_prefill_cp_mode='round-robin-split', enable_prefill_context_parallel=False, prefill_cp_mode='in-seq-split', enable_cp_decode_attn_tp=False, enable_dp_attention=False, enable_dp_attention_local_control_broadcast=False, enable_dp_lm_head=False, enable_attn_tp_input_scattered=False, disable_attn_tp_gather=False, enable_p2p_check=False, device='cuda', base_gpu_id=0, gpu_id_step=1, random_seed=434112381, watchdog_timeout=300, soft_watchdog_timeout=None, sleep_on_idle=False, use_ray=False, custom_sigquit_handler=None, numa_node=None, gc_threshold=None, host='0.0.0.0', port=30000, fastapi_root_path='', smg_grpc_mode=False, grpc_mode=False, grpc_port=None, sidecar=None, sidecar_args=None, skip_server_warmup=False, warmups=None, enable_http2=False, ssl_keyfile=None, ssl_certfile=None, ssl_ca_certs=None, ssl_keyfile_password=None, enable_ssl_refresh=False, api_key=None, admin_api_key=None, served_model_name='kimi-k3', weight_version='default', chat_template=None, hf_chat_template_name=None, completion_template=None, file_storage_path='sglang_storage', enable_cache_report=False, reasoning_parser=None, default_chat_template_kwargs=None, strip_thinking_cache=False, enable_strict_thinking=False, tool_call_parser=None, tool_server=None, sampling_defaults='model', asr_max_buffer_seconds=60, asr_max_concurrent_sessions=32, preferred_sampling_params=None, allow_auto_truncate=False, stream_interval=1, batch_notify_size=16, stream_response_default_include_usage=False, incremental_streaming_output=False, enable_streaming_session=False, enable_session_radix_cache=False, log_level='info', log_level_http=None, log_requests=False, log_requests_level=2, log_requests_format='text', log_requests_target=None, uvicorn_access_log_exclude_prefixes=[], crash_dump_folder=None, show_time_cost=False, enable_metrics=False, smg_http_sidecar_port=None, enable_mfu_metrics=False, enable_metrics_for_all_schedulers=False, load_snapshot_publish_interval=15, tokenizer_metrics_custom_labels_header='x-custom-labels', tokenizer_metrics_allowed_custom_labels=None, extra_metric_labels=None, bucket_time_to_first_token=None, bucket_inter_token_latency=None, bucket_e2e_request_latency=None, prompt_tokens_buckets=None, generation_tokens_buckets=None, gc_warning_threshold_secs=0.0, decode_log_interval=40, enable_request_time_stats_logging=False, kv_events_config=None, enable_forward_pass_metrics=False, forward_pass_metrics_worker_id='', forward_pass_metrics_ipc_name=None, enable_trace=False, trace_modules='request', otlp_traces_endpoint='localhost:4317', export_metrics_to_file=False, export_metrics_to_file_dir=None, stat_loggers=None, constrained_json_whitespace_pattern=None, constrained_json_disable_any_whitespace=False, attention_backend='triton', decode_attention_backend=None, prefill_attention_backend=None, sampling_backend='flashinfer', grammar_backend='xgrammar', radix_cache_backend=None, mm_attention_backend=None, fp8_gemm_runner_backend='auto', fp4_gemm_runner_backend='auto', bf16_gemm_backend='auto', dsa_prefill_backend=None, dsa_decode_backend=None, dsa_paged_mqa_logits_backend='auto', dsa_topk_backend='sgl-kernel', disable_flashinfer_autotune=False, flashinfer_autotune_skip_ops=None, mamba_backend='triton', cuda_graph_config=CudaGraphConfig(decode=PhaseConfig(backend='full', max_bs=16, bs=[1, 2, 4, 8, 12, 16], tc_compiler='eager', full_prefill_max_req=None), prefill=PhaseConfig(backend='disabled', max_bs=2048, bs=[4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], tc_compiler='eager', full_prefill_max_req=None)), cuda_graph_backend_decode=None, cuda_graph_backend_prefill=None, cuda_graph_max_bs_decode=16, cuda_graph_max_bs_prefill=None, cuda_graph_bs_decode=None, cuda_graph_bs_prefill=None, cuda_graph_tc_compiler=None, disable_prefill_cuda_graph=False, disable_decode_cuda_graph=False, disable_cuda_graph=False, disable_cuda_graph_padding=False, enable_profile_cuda_graph=False, enable_cudagraph_gc=False, debug_cuda_graph=False, enable_layerwise_nvtx_marker=False, enable_nccl_nvls=False, enable_symm_mem=False, triton_attention_reduce_in_fp32=False, triton_attention_num_kv_splits=8, triton_attention_split_tile_size=None, flashinfer_mla_disable_ragged=False, enable_fused_qk_norm_rope=False, enable_precise_embedding_interpolation=False, enable_fused_moe_sum_all_reduce=False, enable_deepseek_v4_fp4_indexer=False, disable_custom_all_reduce=False, enable_mscclpp=False, enable_torch_symm_mem=False, enable_scattered_sconv=False, pre_warm_nccl=False, enable_quant_communications=False, enable_flashinfer_allreduce_fusion=False, enforce_disable_flashinfer_allreduce_fusion=False, flashinfer_allreduce_fusion_backend=None, enable_aiter_allreduce_fusion=False, enable_torch_compile=False, enable_torch_compile_debug_mode=False, torch_compile_max_bs=32, torchao_config='', speculative_algorithm=None, speculative_draft_model_path=None, speculative_draft_model_revision=None, speculative_draft_load_format=None, speculative_num_steps=None, speculative_eagle_topk=None, speculative_num_draft_tokens=None, speculative_dflash_block_size=None, speculative_dspark_block_size=None, speculative_dspark_sps_table_path=None, speculative_dspark_confidence_sts_path=None, speculative_dspark_align_verify_tokens_to_graph_tier=False, speculative_accept_threshold_single=1.0, speculative_accept_threshold_acc=1.0, speculative_use_rejection_sampling=False, speculative_token_map=None, speculative_attention_mode='prefill', speculative_draft_attention_backend=None, speculative_draft_window_size=None, speculative_moe_runner_backend='marlin', speculative_moe_a2a_backend=None, speculative_draft_model_quantization=None, speculative_skip_dp_mlp_sync=False, enable_multi_layer_eagle=False, speculative_adaptive=False, speculative_adaptive_config=None, decoupled_spec_bind_endpoint=None, decoupled_spec_connect_endpoints=None, decoupled_spec_rank=None, decoupled_spec_role='null', spec_trace_dir=None, speculative_ngram_min_bfs_breadth=1, speculative_ngram_max_bfs_breadth=10, speculative_ngram_match_type='BFS', speculative_ngram_max_trie_depth=18, speculative_ngram_capacity=10000000, speculative_ngram_external_corpus_path=None, speculative_ngram_external_sam_budget=0, speculative_ngram_external_corpus_max_tokens=10000000, ep_size=32, moe_a2a_backend='none', moe_runner_backend='marlin', flashinfer_mxfp4_moe_precision='default', deepep_mode='auto', fuseep_mode=2, deepep_dispatcher_output_dtype='auto', ep_num_redundant_experts=0, ep_dispatch_algorithm=None, init_expert_location='trivial', enable_eplb=False, eplb_algorithm='auto', eplb_rebalance_num_iterations=1000, eplb_rebalance_layers_per_chunk=None, eplb_min_rebalancing_utilization_threshold=1.0, expert_distribution_recorder_mode=None, expert_distribution_recorder_buffer_size=1000, enable_expert_distribution_metrics=False, deepep_config=None, moe_dense_tp_size=None, elastic_ep_backend=None, enable_elastic_expert_backup=False, mooncake_ib_device=None, enable_waterfill=False, ep_join_mode=None, ep_join_rank_offset=0, elastic_ep_initial_size=None, max_ep_size=None, elastic_ep_scale_timeout=600, elastic_ep_rejoin=False, disable_flashinfer_cutlass_moe_fp4_allgather=False, disable_shared_experts_fusion=False, enforce_shared_experts_fusion=False, max_mamba_cache_size=None, mamba_ssm_dtype=None, mamba_max_states_per_path=-1, enable_mamba_cache_stochastic_rounding=False, mamba_cache_philox_rounds=0, mamba_full_memory_ratio=0.21, mamba_radix_cache_strategy='extra_buffer_lazy', uses_mamba_radix_cache=False, mamba_track_interval=256, enable_int8_mamba_checkpoint=False, int8_mamba_ckpt_size=None, linear_attn_backend='triton', linear_attn_decode_backend=None, linear_attn_prefill_backend=None, linear_attn_verify_backend=None, enable_linear_replayssm=False, linear_replayssm_cache_len=16, enable_linear_replayssm_spec=False, enable_hierarchical_cache=False, hicache_ratio=2.0, hicache_size=0, hicache_write_policy='write_through', hicache_io_backend='kernel', hicache_mem_layout='page_first', hicache_storage_backend=None, hicache_storage_prefetch_policy='timeout', hicache_storage_backend_extra_config=None, enable_hisparse=False, hisparse_config=None, enable_broadcast_mm_inputs_process=False, enable_prefix_mm_cache=False, mm_enable_dp_encoder=False, mm_process_config={}, mm_processor_worker_num=0, mm_io_worker_num=0, limit_mm_data_per_request=None, enable_mm_global_cache=False, disable_fast_image_processor=False, mm_feature_transport='cpu', keep_mm_feature_on_device=False, enable_lora=None, enable_lora_overlap_loading=None, max_lora_rank=None, lora_target_modules=None, lora_paths=None, max_loaded_loras=None, max_loras_per_batch=8, lora_eviction_policy='lru', lora_backend='csgmv', max_lora_chunk_size=16, experts_shared_outer_loras=None, lora_use_virtual_experts=False, lora_strict_loading=False, lora_drain_wait_threshold=0.0, enable_two_batch_overlap=False, enable_single_batch_overlap=False, tbo_token_distribution_threshold=0.48, cpu_offload_gb=0, offload_group_size=-1, offload_num_in_group=1, offload_prefetch_step=1, offload_mode='cpu', enable_lmcache=False, lmcache_config_file=None, enable_flexkv=False, flexkv_config_file=None, kt_weight_path=None, kt_method='AMXINT4', kt_cpuinfer=None, kt_threadpool_count=2, kt_num_gpu_experts=None, kt_max_deferred_experts_per_token=None, dllm_algorithm=None, dllm_algorithm_config=None, dllm_fdfo=True, disaggregation_mode='decode', disaggregation_transfer_backend='mooncake', disaggregation_bootstrap_port=28800, disaggregation_ib_device='mlx5_0,mlx5_1,mlx5_2,mlx5_3', disaggregation_decode_enable_radix_cache=False, disaggregation_decode_enable_offload_kvcache=False, num_reserved_decode_tokens=512, disaggregation_decode_extra_slots=0, disaggregation_decode_polling_interval=1, optimistic_prefill_attempts=0, encoder_only=False, language_only=False, encoder_transfer_backend='zmq_to_tokenizer', encoder_urls=[], encoder_bootstrap_port=8997, encoder_register_urls=[], enable_adaptive_dispatch_to_encoder=False, enable_pdmux=False, pdmux_config_path=None, sm_group_num=8, custom_weight_loader=[], weight_loader_disable_mmap=False, weight_loader_prefetch_checkpoints=False, weight_loader_prefetch_num_threads=4, weight_loader_drop_cache_after_load=False, remote_instance_weight_loader_seed_instance_ip=None, remote_instance_weight_loader_seed_instance_service_port=None, remote_instance_weight_loader_send_weights_group_ports=None, remote_instance_weight_loader_backend='nccl', remote_instance_weight_loader_start_seed_via_transfer_engine=False, engine_info_bootstrap_port=6789, modelexpress_config=None, download_dir=None, model_checksum=None, delete_ckpt_after_loading=False, decrypted_config_file=None, decrypted_draft_config_file=None, checkpoint_engine_wait_weights_before_ready=False, enable_prefill_delayer=False, prefill_delayer_max_delay_passes=30, prefill_delayer_token_usage_low_watermark=None, prefill_delayer_forward_passes_buckets=None, prefill_delayer_wait_seconds_buckets=None, prefill_delayer_queue_min_ratio=None, prefill_delayer_max_delay_ms=None, min_free_slots_delay=None, enable_deterministic_inference=False, rl_on_policy_target=None, kv_canary='none', kv_canary_real_data='none', kv_canary_sweep_interval=0, enable_dynamic_batch_tokenizer=False, dynamic_batch_tokenizer_batch_size=32, dynamic_batch_tokenizer_batch_timeout=0.002, enable_tokenizer_batch_encode=False, disable_tokenizer_batch_decode=False, debug_tensor_dump_output_folder=None, debug_tensor_dump_layers=None, debug_tensor_dump_input_file=None, enable_memory_saver=False, enable_weights_cpu_backup=False, enable_draft_weights_cpu_backup=False, enable_custom_logit_processor=False, enable_return_hidden_states=False, enable_return_routed_experts=False, enable_return_indexer_topk=False, disable_outlines_disk_cache=False, enable_mis=False, weight_cache_mode='off', weight_cache_socket=None, weight_cache_timeout=1800, forward_hooks=None, msprobe_dump_config=None)
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:22 TP15 EP15] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:22 TP15 EP15] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:23 TP9 EP9] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:23 TP9 EP9] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:23 TP12 EP12] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:23 TP12 EP12] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:23 TP13 EP13] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:23 TP13 EP13] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:23 TP14 EP14] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:23 TP14 EP14] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:23 TP8 EP8] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:23 TP8 EP8] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:23 TP10 EP10] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:23 TP10 EP10] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:23 TP11 EP11] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:23 TP11 EP11] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:24 TP15 EP15] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:24 TP9 EP9] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:24 TP13 EP13] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:24 TP12 EP12] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:24 TP14 EP14] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:25 TP10 EP10] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:25 TP8 EP8] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:25 TP11 EP11] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:26.087226   242 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:26.087333   242 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:26.087350   242 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.6 port: 12001
+I0827 04:36:26.087409   242 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.6:16298
+I0827 04:36:26.087569   242 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:26.095995   236 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:26.096091   236 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:26.096107   236 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.6 port: 12001
+I0827 04:36:26.096150   236 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.6:15958
+I0827 04:36:26.096269   236 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:26.096457   240 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:26.096560   240 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:26.096577   240 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.6 port: 12001
+I0827 04:36:26.096626   240 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.6:16447
+I0827 04:36:26.096773   240 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:26.096876   241 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:26.096975   241 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:26.096992   241 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.6 port: 12001
+I0827 04:36:26.097035   241 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.6:16035
+I0827 04:36:26.097170   241 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:26.103180   241 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:26.107820   240 topology.cpp:127] Device mlx5_0 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:26.109326   239 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:26.109416   239 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:26.109428   239 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.6 port: 12001
+I0827 04:36:26.109464   239 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.6:16843
+I0827 04:36:26.109562   239 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:26.114297   242 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:26.114653   239 topology.cpp:127] Device mlx5_0 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:26.115540   237 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:26.115630   237 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:26.115645   237 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.6 port: 12001
+I0827 04:36:26.115686   237 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.6:15544
+I0827 04:36:26.115813   237 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:26.118220   240 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:26.121481   242 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:26.122418   237 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:26.126497   236 topology.cpp:127] Device mlx5_0 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:26.126574   235 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:26.126684   235 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:26.126703   235 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.6 port: 12001
+I0827 04:36:26.126758   235 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.6:15006
+I0827 04:36:26.126933   235 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:26.127116   240 topology.cpp:127] Device mlx5_2 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:26.127162   238 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:26.127270   238 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:26.127285   238 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.6 port: 12001
+I0827 04:36:26.127342   238 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.6:16140
+I0827 04:36:26.127489   238 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:26.128777   237 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:26.129559   242 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:26.130432   241 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:26.133821   238 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:26.134943   235 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:26.135799   240 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:26.136080   237 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:26.138710   241 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:26.139173   242 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:26.139978   240 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:26.140033   240 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:26.140090   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.140504   235 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:26.142936   242 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:26.142961   242 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:26.143014   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.144655   237 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:26.146834   235 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:26.147805   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:26.147867   241 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:26.148357   239 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:26.148370   237 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:26.148394   237 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:26.148444   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.149503   240 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:06
+I0827 04:36:26.149536   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.150290   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:26.150381   236 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:26.151773   242 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:06
+I0827 04:36:26.151803   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.152055   241 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:26.152087   241 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:26.152148   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.153594   235 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:26.156044   235 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:26.156075   235 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:26.156124   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.157910   238 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:26.159448   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:26.159547   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:26.160912   242 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:06
+I0827 04:36:26.160940   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.161144   241 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:06
+I0827 04:36:26.161176   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.162882   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:26.164348   235 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:06
+I0827 04:36:26.164378   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.165159   238 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:26.168772   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:26.170400   241 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:06
+I0827 04:36:26.170428   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.172459   238 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:26.174376   239 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:26.175243   238 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:26.175272   238 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:26.175325   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.178375   236 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:26.178763   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:26.178785   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:26.180411   237 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:06
+I0827 04:36:26.180462   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.180604   240 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:06
+I0827 04:36:26.180672   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.181643   239 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:26.182818   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:26.184404   238 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:06
+I0827 04:36:26.184437   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.184495   239 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:26.184537   239 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:26.184595   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.185601   236 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:26.186753   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:26.188311   242 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:06
+I0827 04:36:26.188344   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.188390   236 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:26.188416   236 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:26.188470   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.190752   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:26.192051   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:26.192406   235 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:06
+I0827 04:36:26.192444   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.193856   240 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:06
+I0827 04:36:26.193897   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.194197   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:26.195964   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:26.195997   238 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:06
+I0827 04:36:26.196039   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.196040   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:26.197574   242 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:06
+I0827 04:36:26.197599   236 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:06
+I0827 04:36:26.197629   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.197691   242 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:26 TP15 EP15] Init torch distributed begin.
+I0827 04:36:26.198544   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:26.200238   241 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:06
+I0827 04:36:26.200268   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.204000   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:26.205737   238 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:06
+I0827 04:36:26.205766   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.210784   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:26.212373   237 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:06
+I0827 04:36:26.212417   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.218751   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:26.220386   235 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:06
+I0827 04:36:26.220422   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.222854   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:26.223088   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:26.223217   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:26.223345   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:26.224500   241 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:06
+I0827 04:36:26.224536   240 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:06
+I0827 04:36:26.224612   241 transfer_engine_impl.cpp:388] installTransport, type=rdma
+I0827 04:36:26.224682   240 transfer_engine_impl.cpp:388] installTransport, type=rdma
+I0827 04:36:26.224973   236 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:06
+I0827 04:36:26.225018   236 rdma_context.cpp:205] Using SIEVE endpoint store
+[2026-08-27 04:36:26 TP14 EP14] Init torch distributed begin.
+[2026-08-27 04:36:26 TP13 EP13] Init torch distributed begin.
+I0827 04:36:26.225459   239 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:06
+I0827 04:36:26.225523   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.230858   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:26.232468   238 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:06
+I0827 04:36:26.232614   238 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:26 TP11 EP11] Init torch distributed begin.
+I0827 04:36:26.233065   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:26.234685   236 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:06
+I0827 04:36:26.234727   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.238961   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:26.240710   237 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:06
+I0827 04:36:26.240758   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.242511   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:26.244089   236 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:06
+I0827 04:36:26.244236   236 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:26 TP9 EP9] Init torch distributed begin.
+I0827 04:36:26.247059   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:26.248764   235 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:06
+I0827 04:36:26.248920   235 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:26 TP8 EP8] Init torch distributed begin.
+I0827 04:36:26.250787   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:26.252344   239 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:06
+I0827 04:36:26.252393   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.266752   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:26.268360   237 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:06
+I0827 04:36:26.268492   237 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:26 TP10 EP10] Init torch distributed begin.
+I0827 04:36:26.278649   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:26.280236   239 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:06
+I0827 04:36:26.280287   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.306850   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:26.308480   239 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:06
+I0827 04:36:26.308605   239 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:26 TP12 EP12] Init torch distributed begin.
+[2026-08-27 04:36:48 TP14 EP14] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP15 EP15] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP8 EP8] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP11 EP11] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP9 EP9] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP13 EP13] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP10 EP10] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP12 EP12] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP15 EP15] Init torch distributed ends. elapsed=22.03 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP14 EP14] Init torch distributed ends. elapsed=22.01 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP13 EP13] Init torch distributed ends. elapsed=22.01 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP12 EP12] Init torch distributed ends. elapsed=21.92 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP11 EP11] Init torch distributed ends. elapsed=22.00 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP9 EP9] Init torch distributed ends. elapsed=21.99 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP10 EP10] Init torch distributed ends. elapsed=21.96 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP8 EP8] Init torch distributed ends. elapsed=21.98 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP14 EP14] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP15 EP15] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP9 EP9] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP13 EP13] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP8 EP8] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP10 EP10] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP12 EP12] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP14 EP14] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP15 EP15] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP9 EP9] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP10 EP10] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP13 EP13] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP8 EP8] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP12 EP12] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP11 EP11] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP11 EP11] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP13 EP13] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP14 EP14] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP15 EP15] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP12 EP12] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP10 EP10] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP8 EP8] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP9 EP9] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP13 EP13] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP12 EP12] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP14 EP14] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP15 EP15] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP10 EP10] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP8 EP8] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP9 EP9] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP11 EP11] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP11 EP11] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:50 TP8 EP8] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP9 EP9] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP11 EP11] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP12 EP12] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP13 EP13] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP14 EP14] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP10 EP10] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP15 EP15] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:38:57 TP14 EP14] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:57 TP14 EP14] Load weight end. elapsed=129.27 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP9 EP9] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP9 EP9] Load weight end. elapsed=129.52 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP10 EP10] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP10 EP10] Load weight end. elapsed=129.56 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP8 EP8] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP8 EP8] Load weight end. elapsed=129.57 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP13 EP13] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP13 EP13] Load weight end. elapsed=129.58 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP11 EP11] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP11 EP11] Load weight end. elapsed=129.64 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP15 EP15] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP15 EP15] Load weight end. elapsed=129.77 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP12 EP12] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP12 EP12] Load weight end. elapsed=129.90 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:39:01 TP14 EP14] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP14 EP14] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP11 EP11] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP11 EP11] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP8 EP8] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP8 EP8] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP13 EP13] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP13 EP13] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP9 EP9] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP9 EP9] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP15 EP15] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP15 EP15] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP12 EP12] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP12 EP12] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP10 EP10] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP10 EP10] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:02 TP14 EP14] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP15 EP15] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP12 EP12] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP13 EP13] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP10 EP10] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP9 EP9] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP11 EP11] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP8 EP8] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP15 EP15] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP12 EP12] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP10 EP10] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP8 EP8] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP11 EP11] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP13 EP13] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP9 EP9] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP14 EP14] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP9 EP9] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP12 EP12] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP9 EP9] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP10 EP10] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP12 EP12] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP13 EP13] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP15 EP15] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP10 EP10] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP11 EP11] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP13 EP13] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP14 EP14] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP15 EP15] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP11 EP11] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP8 EP8] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP14 EP14] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP8 EP8] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP14 EP14] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP14 EP14] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP9 EP9] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP9 EP9] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP12 EP12] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP12 EP12] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:39:02 TP10 EP10] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP10 EP10] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP8 EP8] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP8 EP8] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:39:02 TP11 EP11] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP11 EP11] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:39:02 TP13 EP13] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP13 EP13] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP15 EP15] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP15 EP15] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:42:56 TP8 EP8] Capture target decode CUDA graph end. elapsed=233.78 s, mem usage=0.60 GB, avail mem=8.52 GB.
+[2026-08-27 04:42:56 TP13 EP13] Capture target decode CUDA graph end. elapsed=233.78 s, mem usage=0.58 GB, avail mem=8.58 GB.
+[2026-08-27 04:42:56 TP8 EP8] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP13 EP13] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP14 EP14] Capture target decode CUDA graph end. elapsed=233.81 s, mem usage=0.56 GB, avail mem=8.59 GB.
+[2026-08-27 04:42:56 TP10 EP10] Capture target decode CUDA graph end. elapsed=233.79 s, mem usage=0.56 GB, avail mem=8.59 GB.
+[2026-08-27 04:42:56 TP14 EP14] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP12 EP12] Capture target decode CUDA graph end. elapsed=233.80 s, mem usage=0.60 GB, avail mem=8.52 GB.
+[2026-08-27 04:42:56 TP10 EP10] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP12 EP12] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP11 EP11] Capture target decode CUDA graph end. elapsed=233.80 s, mem usage=0.62 GB, avail mem=8.51 GB.
+[2026-08-27 04:42:56 TP11 EP11] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP9 EP9] Capture target decode CUDA graph end. elapsed=233.91 s, mem usage=0.58 GB, avail mem=8.58 GB.
+[2026-08-27 04:42:56 TP9 EP9] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP15 EP15] Capture target decode CUDA graph end. elapsed=233.94 s, mem usage=0.62 GB, avail mem=8.51 GB.
+[2026-08-27 04:42:56 TP15 EP15] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+W0827 04:42:57.337543  2432 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5972aec0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.337602  2433 memory_location.cpp:73] Failed to get NUMA node, addr: 0x594a1440, len: 23168: Operation not permitted [1]
+W0827 04:42:57.337656  2434 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59f51680, len: 23168: Operation not permitted [1]
+W0827 04:42:57.337674  2435 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59ebed40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.337788  2436 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5b797200, len: 185344: Operation not permitted [1]
+W0827 04:42:57.337911  2437 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5b7c4640, len: 185344: Operation not permitted [1]
+W0827 04:42:57.337891  2438 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59639800, len: 23168: Operation not permitted [1]
+W0827 04:42:57.337962  2439 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59435700, len: 23168: Operation not permitted [1]
+W0827 04:42:57.338008  2440 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59660c80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.338057  2441 memory_location.cpp:73] Failed to get NUMA node, addr: 0x594c9400, len: 46336: Operation not permitted [1]
+W0827 04:42:57.338071  2442 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59bb1f00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.338107  2443 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59f5fc80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.338161  2445 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5953b700, len: 23168: Operation not permitted [1]
+W0827 04:42:57.338165  2444 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59812080, len: 23168: Operation not permitted [1]
+W0827 04:42:57.338204  2446 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5baddac0, len: 185344: Operation not permitted [1]
+W0827 04:42:57.338325  2448 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59994940, len: 23168: Operation not permitted [1]
+W0827 04:42:57.338340  2447 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5bb0af00, len: 185344: Operation not permitted [1]
+W0827 04:42:57.338538  2449 memory_location.cpp:73] Failed to get NUMA node, addr: 0x45ba2780, len: 46336: Operation not permitted [1]
+W0827 04:42:57.338596  2450 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59ac5f00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.338649  2451 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59984000, len: 23168: Operation not permitted [1]
+W0827 04:42:57.381100  2729 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59997640, len: 23168: Operation not permitted [1]
+W0827 04:42:57.381177  2728 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59a855c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.381280  2730 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59953e40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.381347  2732 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5b87eac0, len: 185344: Operation not permitted [1]
+W0827 04:42:57.381466  2731 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59971000, len: 23168: Operation not permitted [1]
+W0827 04:42:57.381542  2733 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5b8abf00, len: 185344: Operation not permitted [1]
+W0827 04:42:57.381620  2734 memory_location.cpp:73] Failed to get NUMA node, addr: 0x599b8540, len: 23168: Operation not permitted [1]
+W0827 04:42:57.381628  2735 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5927e3c0, len: 46336: Operation not permitted [1]
+W0827 04:42:57.381863  2736 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59900b40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.381878  2737 memory_location.cpp:73] Failed to get NUMA node, addr: 0x598b4f80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.421048  2876 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58bb5680, len: 23168: Operation not permitted [1]
+W0827 04:42:57.421196  2878 memory_location.cpp:73] Failed to get NUMA node, addr: 0x592f7a00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.421193  2877 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59321cc0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.421262  2881 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ab93340, len: 185344: Operation not permitted [1]
+W0827 04:42:57.421321  2879 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5929c2c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.421382  2880 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ab65f00, len: 185344: Operation not permitted [1]
+W0827 04:42:57.421423  2882 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58aa4dc0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.421440  2883 memory_location.cpp:73] Failed to get NUMA node, addr: 0x585179c0, len: 46336: Operation not permitted [1]
+W0827 04:42:57.421434  2884 memory_location.cpp:73] Failed to get NUMA node, addr: 0x588b0240, len: 23168: Operation not permitted [1]
+W0827 04:42:57.421594  2885 memory_location.cpp:73] Failed to get NUMA node, addr: 0x588d0800, len: 23168: Operation not permitted [1]
+W0827 04:42:57.423333  2900 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58bc9040, len: 23168: Operation not permitted [1]
+W0827 04:42:57.423418  2903 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58a57e40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.423974  2913 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ae07a80, len: 185344: Operation not permitted [1]
+W0827 04:42:57.424072  2911 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5adda640, len: 185344: Operation not permitted [1]
+W0827 04:42:57.424093  2908 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58f8c900, len: 23168: Operation not permitted [1]
+W0827 04:42:57.424189  2924 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58d795c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.424269  2906 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58e657c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.424296  2915 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58c46ac0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.424311  2922 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58e39680, len: 23168: Operation not permitted [1]
+W0827 04:42:57.424809  2930 memory_location.cpp:73] Failed to get NUMA node, addr: 0x591a7f40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.424866  2932 memory_location.cpp:73] Failed to get NUMA node, addr: 0x593dee00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.425096  2934 memory_location.cpp:73] Failed to get NUMA node, addr: 0x596af5c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.425186  2935 memory_location.cpp:73] Failed to get NUMA node, addr: 0x595b2ac0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.425271  2938 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5b4f5f80, len: 185344: Operation not permitted [1]
+W0827 04:42:57.425292  2937 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5b4c8b40, len: 185344: Operation not permitted [1]
+W0827 04:42:57.425374  2939 memory_location.cpp:73] Failed to get NUMA node, addr: 0x594e07c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.425433  2942 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58ec6dc0, len: 46336: Operation not permitted [1]
+W0827 04:42:57.425568  2944 memory_location.cpp:73] Failed to get NUMA node, addr: 0x596a8180, len: 23168: Operation not permitted [1]
+W0827 04:42:57.425750  2919 memory_location.cpp:73] Failed to get NUMA node, addr: 0x45850900, len: 46336: Operation not permitted [1]
+W0827 04:42:57.425791  2946 memory_location.cpp:73] Failed to get NUMA node, addr: 0x592c6f00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.427150  2998 memory_location.cpp:73] Failed to get NUMA node, addr: 0x599c15c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.427410  2989 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59750440, len: 23168: Operation not permitted [1]
+W0827 04:42:57.427572  3010 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5bad0cc0, len: 185344: Operation not permitted [1]
+W0827 04:42:57.427641  3014 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a39cd80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.427647  3024 memory_location.cpp:73] Failed to get NUMA node, addr: 0x464ff140, len: 23168: Operation not permitted [1]
+W0827 04:42:57.427655  3002 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a377380, len: 23168: Operation not permitted [1]
+W0827 04:42:57.427663  2993 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59876080, len: 23168: Operation not permitted [1]
+W0827 04:42:57.427681  3007 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5baa3880, len: 185344: Operation not permitted [1]
+W0827 04:42:57.427703  3025 memory_location.cpp:73] Failed to get NUMA node, addr: 0x463cd480, len: 23168: Operation not permitted [1]
+W0827 04:42:57.427706  3018 memory_location.cpp:73] Failed to get NUMA node, addr: 0x464e6b00, len: 46336: Operation not permitted [1]
+W0827 04:42:57.586287  3470 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a4fe500, len: 23168: Operation not permitted [1]
+W0827 04:42:57.586316  3471 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a8129c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.586287  3468 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a530a00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.586329  3469 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a72a280, len: 23168: Operation not permitted [1]
+W0827 04:42:57.586364  3472 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5c756140, len: 185344: Operation not permitted [1]
+W0827 04:42:57.586422  3474 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a287d00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.586429  3473 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5c783580, len: 185344: Operation not permitted [1]
+W0827 04:42:57.586477  3476 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a731080, len: 23168: Operation not permitted [1]
+W0827 04:42:57.586475  3475 memory_location.cpp:73] Failed to get NUMA node, addr: 0x46a69900, len: 46336: Operation not permitted [1]
+W0827 04:42:57.586767  3477 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a2cd580, len: 23168: Operation not permitted [1]
+[2026-08-27 04:42:59] Dummy health check server started in background thread at 0.0.0.0:30000
+I0827 04:48:13.995834  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:13.996141  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.010807  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.011080  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.031366  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.039373  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.042748  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.043025  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.057039  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.060467  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.074244  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.093904  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.097144  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.097398  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.100539  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.103674  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.106997  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.107286  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.107595  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.107802  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.108084  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.108319  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.111485  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.114651  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.114880  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.118037  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.127089  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.130350  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.136305  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.145159  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.145349  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.145601  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.148783  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.161775  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.162027  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.165246  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.165525  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.165712  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.171312  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.182595  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.182830  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.183091  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.183334  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.197396  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.202782  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.202994  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.211761  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.212010  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.212227  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.212466  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.218309  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.218585  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.218844  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.219072  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.219323  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.221719  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.223042  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.230329  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.230581  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.233455  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.235435  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.239619  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.239846  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.243142  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.248512  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.248768  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.249244  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.249539  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.249773  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.257058  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.259995  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.260232  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.260426  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.260643  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.263525  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.263789  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.264045  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.270218  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.272343  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.272601  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.272828  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.273056  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.273649  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.273857  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.274083  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.274338  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.274350  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.274650  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.274857  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.280273  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.281651  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.283277  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.288719  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.291597  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.291850  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.294591  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.297427  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.297646  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.297865  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.298131  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.298350  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.298571  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.300019  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.300335  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.301384  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.303872  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.304067  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.306974  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.307229  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.307489  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.308355  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.311895  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.312120  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.312359  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.312578  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.313377  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.317476  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.317679  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.319007  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.323084  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.323379  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.323665  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.333745  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.333979  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.334159  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.334322  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.337862  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.351508  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.351702  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.351871  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.352072  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.352252  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.355370  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.364216  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.367312  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.367552  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.367769  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.368913  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.375226  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.375468  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.375686  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.375947  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.376187  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.376428  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.376683  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.376904  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.377125  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.380695  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.382211  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.384042  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.384289  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.384522  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.390187  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.390600  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.391651  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.392228  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.398043  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.404222  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.404719  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.404901  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.405130  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.405673  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.405961  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.406175  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.406375  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.406561  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.406761  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.406991  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.407531  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.411696  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.411933  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.412912  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.416914  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.417143  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.417341  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.418166  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.418262  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.423719  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.425009  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.431804  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.432065  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.432313  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.432547  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.432744  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.432958  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.433151  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.433398  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.433662  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.433943  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.435259  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.439250  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.440708  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.443462  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.444864  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.446599  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.453094  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.457774  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.458002  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.458236  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.461966  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.463891  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.464155  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.465241  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.467888  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.468128  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.472088  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.473263  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.474021  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.474298  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.475373  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.479022  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.479259  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.479537  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.479795  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.480019  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.480264  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.480456  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.480696  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.480752  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.480891  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.481091  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.481326  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.481606  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.481833  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.482064  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.482175  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.482331  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.482553  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.482764  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.491822  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.493273  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.497345  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.498744  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.501601  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.502969  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.502988  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.509712  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.516039  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.516237  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.516439  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.519500  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.519733  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.519950  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.520223  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.520452  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.520711  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.520921  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.521140  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.521364  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.521629  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.521858  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.522101  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.522328  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.522553  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.522855  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.523010  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.523072  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.523293  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.523304  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.524191  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.524235  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.528468  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.535353  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.535600  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.535769  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.535825  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.536702  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.537087  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.542068  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.543424  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.545435  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.546203  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.546453  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.546715  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.546816  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.546933  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.547155  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.547389  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.547611  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.547857  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.548101  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.551676  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.553035  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.553618  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.554762  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.555172  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.556280  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.558871  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.565316  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.568676  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.569031  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.569290  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.569475  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.570151  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.572939  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.574455  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.575088  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.575302  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.576429  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.578527  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.578720  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.580932  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.582515  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.583039  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.583253  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.583478  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.583778  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.583832  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.584945  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.585202  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.586391  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.591900  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.593173  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.593271  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.593396  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.593664  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.596908  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.601369  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.602131  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.602725  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.604811  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.605091  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.606220  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.608003  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.610472  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.610689  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.610699  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.610924  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.611121  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.611315  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.611521  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.611792  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.612001  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.612202  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.612247  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.612334  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.612481  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.613652  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.615125  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.619858  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.620216  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.621151  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.624950  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.625170  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.625332  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.625414  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.625568  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.625614  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.625829  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.626011  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.626319  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.631613  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.632304  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.632570  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.632856  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.633095  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.633306  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.633574  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.634308  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.637017  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.637279  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.637583  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.638610  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.639593  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.640806  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.640900  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.641085  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.643601  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.644850  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.645069  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.646034  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.646308  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.646575  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.647399  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.648111  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.649897  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.651033  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.651146  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.653847  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.655079  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.659225  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.662467  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.663832  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.664858  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.665107  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.665309  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.666141  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.669166  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.673557  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.673764  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.673919  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.674009  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.674209  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.674503  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.674647  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.674710  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.674896  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.674955  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.675308  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.675336  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.675500  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.675552  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.675832  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.676005  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.676035  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.676223  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.676425  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.676607  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.680454  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.681128  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.685020  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.685290  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.686386  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.686959  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.694617  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.694845  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.695986  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.697674  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.698963  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.699177  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.699368  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.701349  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.701767  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.702576  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.702847  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.703204  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.706663  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.706753  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.706854  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.707039  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.707223  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.707465  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.707770  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.708069  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.708083  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.708248  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.708472  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.708722  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.708920  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.709123  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.710752  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.710954  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.712257  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.719622  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.719846  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.720033  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.720993  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.725260  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.725533  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.725705  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.725924  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.726168  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.726495  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.726595  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.727191  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.728030  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.728752  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.732066  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.734992  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.740563  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.746973  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.750195  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.750399  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.750788  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.752028  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.753922  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.756233  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.756840  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.759518  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.760934  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.766074  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.767369  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.769282  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.769475  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.769667  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.770648  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.774008  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.774199  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.774485  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.774672  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.774865  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.775053  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.775226  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.777894  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.778448  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.778666  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.779709  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.780627  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.781956  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.783814  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.785252  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.785427  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.785633  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.786656  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.787170  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.792593  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.797327  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.797559  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.797796  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.798789  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.801191  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.804433  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.804607  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.804807  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.805258  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.805953  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.806377  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.810664  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.811244  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.812598  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.824426  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.824638  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.824924  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.825793  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.831593  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.833058  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.833375  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.833634  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.833873  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.834122  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.834323  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.834601  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.834823  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.835043  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.835248  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.835427  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.835631  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.835824  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.836011  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.838608  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.838699  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.854580  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.858219  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.859412  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.862950  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.864025  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.865686  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.865891  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.866111  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.866333  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.866657  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.866854  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.867084  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.867278  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.867471  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.867661  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.872786  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.876446  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.876659  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.876896  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.877848  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.880227  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.880436  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.883129  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.883190  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.883363  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.883567  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.884621  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.884654  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.888731  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.926294  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.926565  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.926801  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.926995  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.927213  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.927462  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.927661  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.927874  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.937942  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.938158  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.938393  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.939355  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.949986  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.950210  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.950419  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.951480  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.985823  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.986092  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.986321  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.986529  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.986739  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.986924  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.987135  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.987387  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.987639  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.987892  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.988083  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.988299  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.988473  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.988667  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.012990  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.013230  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.013480  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.013671  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.013873  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.014089  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.014310  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.014530  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.014705  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.014922  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.029297  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.029497  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.029675  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.029886  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.030081  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.030721  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.075974  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.076195  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.076392  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.076560  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.076789  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.076989  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.077189  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.077405  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.086928  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.087107  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.087375  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.088318  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.098666  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.098937  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.099119  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.100116  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.134526  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.134721  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.134938  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.135179  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.135361  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.135597  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.135814  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.136029  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.136237  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.136453  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.136703  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.136907  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.137085  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.139734  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.164215  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.166940  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.167219  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.167455  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.167656  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.167850  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.168042  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.168241  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.168471  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.168686  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.168931  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.169119  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.183355  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.183578  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.183832  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.184046  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.184273  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.184649  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.231909  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.232179  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.232437  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.232667  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.232929  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.233191  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.233395  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.236016  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.241250  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.243832  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.244118  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.246883  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.252156  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.254787  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.257608  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.265419  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.273245  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.283512  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.283768  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.286342  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.291661  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.291878  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.292133  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.292326  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.292569  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.292784  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.293030  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.298199  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.298434  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.298682  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.311589  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.311841  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.319455  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.319670  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.319875  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.320118  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.320379  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.323133  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.323354  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.323570  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.331285  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.336499  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.336781  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.336977  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.337196  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.337783  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.375766  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.375975  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.376159  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.376356  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.376547  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.376745  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.376950  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.379487  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.384498  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.387058  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.387269  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.389864  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.394851  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.397480  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.397692  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.407253  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.424064  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.424232  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.424441  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.424628  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.432085  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.432324  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.432541  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.432767  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.432996  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.433199  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.433389  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.438321  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.438519  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.438731  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.450695  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.450888  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.458081  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.458277  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.458539  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.458791  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.458984  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.461611  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.461808  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.462006  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.469441  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.474292  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.474494  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.474697  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.474973  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.475522  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.457798  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.458177  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.459460  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.463181  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.464798  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.469339  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.469461  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.470870  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.475363  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.479321  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.480788  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.481803  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.483294  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.485772  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.494769  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.496369  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.500792  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.299279  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.299625  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.301414  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.309536  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.311267  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.312475  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.314007  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.319712  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.324035  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.325511  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.330739  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.330972  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.331518  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.332414  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.337031  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.339011  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.339269  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.340338  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.342085  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.343655  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.344139  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.345620  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.348776  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.349041  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.349591  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.350200  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.360395  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.360633  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.361879  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.362035  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.362210  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.363549  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.366465  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.370512  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.371948  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.373085  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.373347  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.374653  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.376161  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.378137  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.378718  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.378968  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.379657  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.384268  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.385306  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.385583  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.386750  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.386739  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.386987  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.388157  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.394881  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.398934  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.400285  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.402129  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.402400  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.403517  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.406072  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.406335  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.406554  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.406766  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.407377  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.416814  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.417057  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.418246  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.426201  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.426426  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.426638  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.426827  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.427093  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.427573  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.430253  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.430498  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.431689  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.391251  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.393167  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.399513  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.593394  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.593744  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.595059  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.599748  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.607136  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.608918  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.613543  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:40.735651  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:40.737267  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:40.743311  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:40.749338  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:40.750874  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:48.322273  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:48.323827  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:48.330022  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:55.890759  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:55.891063  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:55.892275  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:55.900822  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:55.902331  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:55.907302  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:55.915869  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:55.917403  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.662242  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.663878  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.676903  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.677158  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.677393  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.678658  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.690680  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.691715  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.692193  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.693166  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.699034  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.706213  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.707717  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.708302  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.710057  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.710531  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.725970  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.726244  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.727491  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.460754  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.462435  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.468483  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.477201  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.477489  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.478799  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.514531  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.514875  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.516037  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.170460  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.172108  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.178090  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.187213  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.187461  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.188678  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.201197  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.201504  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.202839  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.215639  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.215888  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.217134  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:47.594174  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:47.607700  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:48.578758  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:48.588865  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:48.597946  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:49.367239  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:49.380311  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:50.292033  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:50.302100  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:50.303560  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:50.312023  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:50.320772  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.137686  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.152026  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.162048  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.026280  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.032133  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.038511  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.041432  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.047593  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.050129  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.058367  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.826843  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.838438  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.847219  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:53.739216  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:54.582928  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:55.353269  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:55.481009  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:56.209721  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:57.074601  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:57.089560  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:57.620945  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:57.639796  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:57.643188  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:57.764745  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:05.707906  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:05.720738  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:06.742728  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:06.752833  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:07.478389  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:08.273787  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:09.240689  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:09.253865  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:10.164781  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:10.170979  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:11.012079  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:11.870416  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:18.023379  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:18.250713  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:18.926174  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:30.803402  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:30.804934  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:32.463899  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:52.158495  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:56.508555  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:56.518028  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:58.313472  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-c8-20260827-1340/logs/manual_d_node2.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-c8-20260827-1340/logs/manual_d_node2.log
new file mode 100644
index 0000000..cf2064d
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-c8-20260827-1340/logs/manual_d_node2.log
@@ -0,0 +1,1470 @@
+/usr/local/lib/python3.12/dist-packages/torchao/quantization/quant_api.py:1731: SyntaxWarning: invalid escape sequence '\.'
+  """Configuration class for applying different quantization configs to modules or parameters based on their fully qualified names (FQNs).
+/sgl-workspace/sglang/python/sglang/launch_server.py:56: UserWarning: 'python -m sglang.launch_server' is still supported, but 'sglang serve' is the recommended entrypoint.
+  Example: sglang serve --model-path <model> [options]
+  warnings.warn(
+[2026-08-27 04:35:59] KV cache is forced as chunk cache for decode server
+[2026-08-27 04:36:03] Breakable CUDA graph is incompatible with MLA attention; disabling prefill CUDA graph.
+[2026-08-27 04:36:03] Attention backend not specified. Use triton backend by default.
+[2026-08-27 04:36:06] server_args=ServerArgs(model_path='/data/hf_models/Kimi-K3', tokenizer_path='/data/hf_models/Kimi-K3', tokenizer_mode='auto', tokenizer_backend='huggingface', tokenizer_worker_num=1, detokenizer_worker_num=1, skip_tokenizer_init=False, load_format='auto', model_loader_extra_config='{}', trust_remote_code=True, context_length=None, is_embedding=False, enable_multimodal=None, revision=None, model_impl='auto', model_config_parser='auto', json_model_override_args='{}', dtype='auto', quantization=None, quantization_param_path=None, kv_cache_dtype='auto', enable_fp32_lm_head=False, modelopt_quant=None, modelopt_checkpoint_restore_path=None, modelopt_checkpoint_save_path=None, modelopt_export_path=None, quantize_and_serve=False, rl_quant_profile=None, enable_tf32_matmul=False, mem_fraction_static=0.88, max_running_requests=None, max_queued_requests=None, max_total_tokens=None, chunked_prefill_size=8192, enable_dynamic_chunking=False, max_prefill_tokens=16384, prefill_max_requests=None, schedule_policy='fcfs', enable_priority_scheduling=False, disable_priority_preemption=False, default_priority_value=None, abort_on_priority_when_disabled=False, schedule_low_priority_values_first=False, priority_scheduling_preemption_threshold=10, retraction_policy='length', schedule_conservativeness=1.0, page_size=64, swa_full_tokens_ratio=0.8, disable_hybrid_swa_memory=False, radix_eviction_policy='lru', prefill_only_disable_kv_cache=False, disable_radix_cache=True, enable_page_major_kv_layout=False, enable_unified_memory=False, disable_chunked_prefix_cache=False, disable_overlap_schedule=False, num_continuous_decode_steps=1, scheduler_recv_interval=1, enable_mixed_chunk=False, nccl_port=None, dist_timeout=3600, dist_init_addr='174.1.60.5:20000', nnodes=4, node_rank=2, tp_size=32, dcp_size=1, pp_size=1, pp_max_micro_batch_size=None, pp_async_batch_depth=0, dp_size=1, load_balance_method='round_robin', attn_cp_size=1, moe_dp_size=1, dcp_comm_backend='ag_rs', dcp_replicate_q_proj=None, dwdp_size=1, enable_prefill_cp=False, cp_strategy=None, enable_dsa_cache_layer_split=False, enable_dsa_prefill_context_parallel=False, dsa_prefill_cp_mode='round-robin-split', enable_prefill_context_parallel=False, prefill_cp_mode='in-seq-split', enable_cp_decode_attn_tp=False, enable_dp_attention=False, enable_dp_attention_local_control_broadcast=False, enable_dp_lm_head=False, enable_attn_tp_input_scattered=False, disable_attn_tp_gather=False, enable_p2p_check=False, device='cuda', base_gpu_id=0, gpu_id_step=1, random_seed=212084647, watchdog_timeout=300, soft_watchdog_timeout=None, sleep_on_idle=False, use_ray=False, custom_sigquit_handler=None, numa_node=None, gc_threshold=None, host='0.0.0.0', port=30000, fastapi_root_path='', smg_grpc_mode=False, grpc_mode=False, grpc_port=None, sidecar=None, sidecar_args=None, skip_server_warmup=False, warmups=None, enable_http2=False, ssl_keyfile=None, ssl_certfile=None, ssl_ca_certs=None, ssl_keyfile_password=None, enable_ssl_refresh=False, api_key=None, admin_api_key=None, served_model_name='kimi-k3', weight_version='default', chat_template=None, hf_chat_template_name=None, completion_template=None, file_storage_path='sglang_storage', enable_cache_report=False, reasoning_parser=None, default_chat_template_kwargs=None, strip_thinking_cache=False, enable_strict_thinking=False, tool_call_parser=None, tool_server=None, sampling_defaults='model', asr_max_buffer_seconds=60, asr_max_concurrent_sessions=32, preferred_sampling_params=None, allow_auto_truncate=False, stream_interval=1, batch_notify_size=16, stream_response_default_include_usage=False, incremental_streaming_output=False, enable_streaming_session=False, enable_session_radix_cache=False, log_level='info', log_level_http=None, log_requests=False, log_requests_level=2, log_requests_format='text', log_requests_target=None, uvicorn_access_log_exclude_prefixes=[], crash_dump_folder=None, show_time_cost=False, enable_metrics=False, smg_http_sidecar_port=None, enable_mfu_metrics=False, enable_metrics_for_all_schedulers=False, load_snapshot_publish_interval=15, tokenizer_metrics_custom_labels_header='x-custom-labels', tokenizer_metrics_allowed_custom_labels=None, extra_metric_labels=None, bucket_time_to_first_token=None, bucket_inter_token_latency=None, bucket_e2e_request_latency=None, prompt_tokens_buckets=None, generation_tokens_buckets=None, gc_warning_threshold_secs=0.0, decode_log_interval=40, enable_request_time_stats_logging=False, kv_events_config=None, enable_forward_pass_metrics=False, forward_pass_metrics_worker_id='', forward_pass_metrics_ipc_name=None, enable_trace=False, trace_modules='request', otlp_traces_endpoint='localhost:4317', export_metrics_to_file=False, export_metrics_to_file_dir=None, stat_loggers=None, constrained_json_whitespace_pattern=None, constrained_json_disable_any_whitespace=False, attention_backend='triton', decode_attention_backend=None, prefill_attention_backend=None, sampling_backend='flashinfer', grammar_backend='xgrammar', radix_cache_backend=None, mm_attention_backend=None, fp8_gemm_runner_backend='auto', fp4_gemm_runner_backend='auto', bf16_gemm_backend='auto', dsa_prefill_backend=None, dsa_decode_backend=None, dsa_paged_mqa_logits_backend='auto', dsa_topk_backend='sgl-kernel', disable_flashinfer_autotune=False, flashinfer_autotune_skip_ops=None, mamba_backend='triton', cuda_graph_config=CudaGraphConfig(decode=PhaseConfig(backend='full', max_bs=16, bs=[1, 2, 4, 8, 12, 16], tc_compiler='eager', full_prefill_max_req=None), prefill=PhaseConfig(backend='disabled', max_bs=2048, bs=[4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], tc_compiler='eager', full_prefill_max_req=None)), cuda_graph_backend_decode=None, cuda_graph_backend_prefill=None, cuda_graph_max_bs_decode=16, cuda_graph_max_bs_prefill=None, cuda_graph_bs_decode=None, cuda_graph_bs_prefill=None, cuda_graph_tc_compiler=None, disable_prefill_cuda_graph=False, disable_decode_cuda_graph=False, disable_cuda_graph=False, disable_cuda_graph_padding=False, enable_profile_cuda_graph=False, enable_cudagraph_gc=False, debug_cuda_graph=False, enable_layerwise_nvtx_marker=False, enable_nccl_nvls=False, enable_symm_mem=False, triton_attention_reduce_in_fp32=False, triton_attention_num_kv_splits=8, triton_attention_split_tile_size=None, flashinfer_mla_disable_ragged=False, enable_fused_qk_norm_rope=False, enable_precise_embedding_interpolation=False, enable_fused_moe_sum_all_reduce=False, enable_deepseek_v4_fp4_indexer=False, disable_custom_all_reduce=False, enable_mscclpp=False, enable_torch_symm_mem=False, enable_scattered_sconv=False, pre_warm_nccl=False, enable_quant_communications=False, enable_flashinfer_allreduce_fusion=False, enforce_disable_flashinfer_allreduce_fusion=False, flashinfer_allreduce_fusion_backend=None, enable_aiter_allreduce_fusion=False, enable_torch_compile=False, enable_torch_compile_debug_mode=False, torch_compile_max_bs=32, torchao_config='', speculative_algorithm=None, speculative_draft_model_path=None, speculative_draft_model_revision=None, speculative_draft_load_format=None, speculative_num_steps=None, speculative_eagle_topk=None, speculative_num_draft_tokens=None, speculative_dflash_block_size=None, speculative_dspark_block_size=None, speculative_dspark_sps_table_path=None, speculative_dspark_confidence_sts_path=None, speculative_dspark_align_verify_tokens_to_graph_tier=False, speculative_accept_threshold_single=1.0, speculative_accept_threshold_acc=1.0, speculative_use_rejection_sampling=False, speculative_token_map=None, speculative_attention_mode='prefill', speculative_draft_attention_backend=None, speculative_draft_window_size=None, speculative_moe_runner_backend='marlin', speculative_moe_a2a_backend=None, speculative_draft_model_quantization=None, speculative_skip_dp_mlp_sync=False, enable_multi_layer_eagle=False, speculative_adaptive=False, speculative_adaptive_config=None, decoupled_spec_bind_endpoint=None, decoupled_spec_connect_endpoints=None, decoupled_spec_rank=None, decoupled_spec_role='null', spec_trace_dir=None, speculative_ngram_min_bfs_breadth=1, speculative_ngram_max_bfs_breadth=10, speculative_ngram_match_type='BFS', speculative_ngram_max_trie_depth=18, speculative_ngram_capacity=10000000, speculative_ngram_external_corpus_path=None, speculative_ngram_external_sam_budget=0, speculative_ngram_external_corpus_max_tokens=10000000, ep_size=32, moe_a2a_backend='none', moe_runner_backend='marlin', flashinfer_mxfp4_moe_precision='default', deepep_mode='auto', fuseep_mode=2, deepep_dispatcher_output_dtype='auto', ep_num_redundant_experts=0, ep_dispatch_algorithm=None, init_expert_location='trivial', enable_eplb=False, eplb_algorithm='auto', eplb_rebalance_num_iterations=1000, eplb_rebalance_layers_per_chunk=None, eplb_min_rebalancing_utilization_threshold=1.0, expert_distribution_recorder_mode=None, expert_distribution_recorder_buffer_size=1000, enable_expert_distribution_metrics=False, deepep_config=None, moe_dense_tp_size=None, elastic_ep_backend=None, enable_elastic_expert_backup=False, mooncake_ib_device=None, enable_waterfill=False, ep_join_mode=None, ep_join_rank_offset=0, elastic_ep_initial_size=None, max_ep_size=None, elastic_ep_scale_timeout=600, elastic_ep_rejoin=False, disable_flashinfer_cutlass_moe_fp4_allgather=False, disable_shared_experts_fusion=False, enforce_shared_experts_fusion=False, max_mamba_cache_size=None, mamba_ssm_dtype=None, mamba_max_states_per_path=-1, enable_mamba_cache_stochastic_rounding=False, mamba_cache_philox_rounds=0, mamba_full_memory_ratio=0.21, mamba_radix_cache_strategy='extra_buffer_lazy', uses_mamba_radix_cache=False, mamba_track_interval=256, enable_int8_mamba_checkpoint=False, int8_mamba_ckpt_size=None, linear_attn_backend='triton', linear_attn_decode_backend=None, linear_attn_prefill_backend=None, linear_attn_verify_backend=None, enable_linear_replayssm=False, linear_replayssm_cache_len=16, enable_linear_replayssm_spec=False, enable_hierarchical_cache=False, hicache_ratio=2.0, hicache_size=0, hicache_write_policy='write_through', hicache_io_backend='kernel', hicache_mem_layout='page_first', hicache_storage_backend=None, hicache_storage_prefetch_policy='timeout', hicache_storage_backend_extra_config=None, enable_hisparse=False, hisparse_config=None, enable_broadcast_mm_inputs_process=False, enable_prefix_mm_cache=False, mm_enable_dp_encoder=False, mm_process_config={}, mm_processor_worker_num=0, mm_io_worker_num=0, limit_mm_data_per_request=None, enable_mm_global_cache=False, disable_fast_image_processor=False, mm_feature_transport='cpu', keep_mm_feature_on_device=False, enable_lora=None, enable_lora_overlap_loading=None, max_lora_rank=None, lora_target_modules=None, lora_paths=None, max_loaded_loras=None, max_loras_per_batch=8, lora_eviction_policy='lru', lora_backend='csgmv', max_lora_chunk_size=16, experts_shared_outer_loras=None, lora_use_virtual_experts=False, lora_strict_loading=False, lora_drain_wait_threshold=0.0, enable_two_batch_overlap=False, enable_single_batch_overlap=False, tbo_token_distribution_threshold=0.48, cpu_offload_gb=0, offload_group_size=-1, offload_num_in_group=1, offload_prefetch_step=1, offload_mode='cpu', enable_lmcache=False, lmcache_config_file=None, enable_flexkv=False, flexkv_config_file=None, kt_weight_path=None, kt_method='AMXINT4', kt_cpuinfer=None, kt_threadpool_count=2, kt_num_gpu_experts=None, kt_max_deferred_experts_per_token=None, dllm_algorithm=None, dllm_algorithm_config=None, dllm_fdfo=True, disaggregation_mode='decode', disaggregation_transfer_backend='mooncake', disaggregation_bootstrap_port=28800, disaggregation_ib_device='mlx5_0,mlx5_1,mlx5_2,mlx5_3', disaggregation_decode_enable_radix_cache=False, disaggregation_decode_enable_offload_kvcache=False, num_reserved_decode_tokens=512, disaggregation_decode_extra_slots=0, disaggregation_decode_polling_interval=1, optimistic_prefill_attempts=0, encoder_only=False, language_only=False, encoder_transfer_backend='zmq_to_tokenizer', encoder_urls=[], encoder_bootstrap_port=8997, encoder_register_urls=[], enable_adaptive_dispatch_to_encoder=False, enable_pdmux=False, pdmux_config_path=None, sm_group_num=8, custom_weight_loader=[], weight_loader_disable_mmap=False, weight_loader_prefetch_checkpoints=False, weight_loader_prefetch_num_threads=4, weight_loader_drop_cache_after_load=False, remote_instance_weight_loader_seed_instance_ip=None, remote_instance_weight_loader_seed_instance_service_port=None, remote_instance_weight_loader_send_weights_group_ports=None, remote_instance_weight_loader_backend='nccl', remote_instance_weight_loader_start_seed_via_transfer_engine=False, engine_info_bootstrap_port=6789, modelexpress_config=None, download_dir=None, model_checksum=None, delete_ckpt_after_loading=False, decrypted_config_file=None, decrypted_draft_config_file=None, checkpoint_engine_wait_weights_before_ready=False, enable_prefill_delayer=False, prefill_delayer_max_delay_passes=30, prefill_delayer_token_usage_low_watermark=None, prefill_delayer_forward_passes_buckets=None, prefill_delayer_wait_seconds_buckets=None, prefill_delayer_queue_min_ratio=None, prefill_delayer_max_delay_ms=None, min_free_slots_delay=None, enable_deterministic_inference=False, rl_on_policy_target=None, kv_canary='none', kv_canary_real_data='none', kv_canary_sweep_interval=0, enable_dynamic_batch_tokenizer=False, dynamic_batch_tokenizer_batch_size=32, dynamic_batch_tokenizer_batch_timeout=0.002, enable_tokenizer_batch_encode=False, disable_tokenizer_batch_decode=False, debug_tensor_dump_output_folder=None, debug_tensor_dump_layers=None, debug_tensor_dump_input_file=None, enable_memory_saver=False, enable_weights_cpu_backup=False, enable_draft_weights_cpu_backup=False, enable_custom_logit_processor=False, enable_return_hidden_states=False, enable_return_routed_experts=False, enable_return_indexer_topk=False, disable_outlines_disk_cache=False, enable_mis=False, weight_cache_mode='off', weight_cache_socket=None, weight_cache_timeout=1800, forward_hooks=None, msprobe_dump_config=None)
+[2026-08-27 04:36:06] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:06] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:06] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:06] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:06] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:06] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:06] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:06] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:24 TP18 EP18] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:24 TP18 EP18] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:24 TP21 EP21] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:24 TP21 EP21] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:24 TP20 EP20] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:24 TP20 EP20] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:25 TP19 EP19] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:25 TP19 EP19] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:25 TP22 EP22] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:25 TP22 EP22] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:25 TP23 EP23] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:25 TP23 EP23] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:26 TP17 EP17] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:26 TP17 EP17] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:26 TP16 EP16] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:26 TP16 EP16] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:27 TP19 EP19] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:27 TP21 EP21] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:27 TP18 EP18] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:27 TP20 EP20] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:27 TP23 EP23] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:27 TP22 EP22] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:27 TP17 EP17] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:27 TP16 EP16] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:29.187384   239 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:29.187532   239 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:29.187558   239 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.7 port: 12001
+I0827 04:36:29.187624   239 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.7:16248
+I0827 04:36:29.187814   239 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:29.191661   235 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:29.191802   235 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:29.191823   235 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.7 port: 12001
+I0827 04:36:29.191901   235 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.7:16408
+I0827 04:36:29.192099   235 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:29.195215   242 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:29.195319   242 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:29.195333   242 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.7 port: 12001
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:29.195333   238 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:29.195374   242 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.7:15148
+I0827 04:36:29.195490   238 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:29.195513   238 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.7 port: 12001
+I0827 04:36:29.195519   242 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:29.195581   238 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.7:16909
+I0827 04:36:29.195815   238 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:29.198207   236 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:29.198346   236 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:29.198370   236 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.7 port: 12001
+I0827 04:36:29.198427   236 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.7:16738
+I0827 04:36:29.198611   236 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:29.200070   242 topology.cpp:127] Device mlx5_0 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:29.201086   237 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:29.201221   237 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:29.201239   237 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.7 port: 12001
+I0827 04:36:29.201300   237 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.7:16522
+I0827 04:36:29.201495   237 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:29.201511   240 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:29.201632   240 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:29.201648   240 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.7 port: 12001
+I0827 04:36:29.201701   240 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.7:16431
+I0827 04:36:29.201871   240 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:29.204236   241 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:29.204355   242 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:29.204380   241 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:29.204393   241 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.7 port: 12001
+I0827 04:36:29.204454   241 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.7:16508
+I0827 04:36:29.204633   241 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:29.204896   238 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:29.207675   237 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:29.208006   240 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:29.208549   242 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:29.212560   242 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:29.214218   242 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:29.214243   242 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:29.214290   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.214713   240 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:29.214757   237 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:29.219102   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:29.219870   239 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:29.220057   235 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:29.220343   242 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:07
+I0827 04:36:29.220369   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.222033   240 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:29.222069   237 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:29.223913   236 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:29.228577   235 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:29.228634   239 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:29.230391   237 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:29.230446   240 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:29.232455   236 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:29.233546   237 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:29.233578   237 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:29.233631   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.234635   240 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:29.234665   240 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:29.234716   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.235774   238 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:29.235827   241 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:29.237140   235 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:29.241117   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:29.241130   236 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:29.242069   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:29.242725   237 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:07
+I0827 04:36:29.242758   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.243661   240 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:07
+I0827 04:36:29.243691   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.245620   235 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:29.247138   238 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:29.247233   241 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:29.248172   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:29.248514   235 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:29.248550   235 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:29.248612   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.248719   236 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:29.249847   242 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:07
+I0827 04:36:29.249886   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.251354   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:29.251672   236 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:29.251719   236 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:29.251781   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.252889   240 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:07
+I0827 04:36:29.252921   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.255584   239 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:29.256263   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:29.257805   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:29.257980   235 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:07
+I0827 04:36:29.258016   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.258512   238 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:29.258567   241 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:29.259447   242 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:07
+I0827 04:36:29.259477   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.259642   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:29.261375   236 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:07
+I0827 04:36:29.261410   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.261518   238 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:29.261557   238 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:29.261615   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.261615   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:29.262753   239 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:29.263202   240 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:07
+I0827 04:36:29.263229   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.265475   239 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:29.265504   239 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:29.265558   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.269449   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:29.271157   236 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:07
+I0827 04:36:29.271186   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.271761   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:29.272920   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:29.273422   237 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:07
+I0827 04:36:29.273456   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.274566   239 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:07
+I0827 04:36:29.274600   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.281176   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:29.282341   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:29.282708   237 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:07
+I0827 04:36:29.282739   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.284137   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:29.284178   239 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:07
+I0827 04:36:29.284219   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.285904   235 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:07
+I0827 04:36:29.285953   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.288185   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:29.289810   242 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:07
+I0827 04:36:29.289940   242 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:29 TP23 EP23] Init torch distributed begin.
+I0827 04:36:29.291602   241 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:29.291882   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:29.291936   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:29.293576   240 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:07
+I0827 04:36:29.293699   240 transfer_engine_impl.cpp:388] installTransport, type=rdma
+I0827 04:36:29.293921   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+[2026-08-27 04:36:29 TP21 EP21] Init torch distributed begin.
+I0827 04:36:29.294121   238 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:07
+I0827 04:36:29.294169   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.294241   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:29.295475   239 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:07
+I0827 04:36:29.295507   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.295553   241 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:29.295591   241 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:29.295648   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.296002   235 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:07
+I0827 04:36:29.296041   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.296232   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:29.298113   236 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:07
+I0827 04:36:29.298154   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.303915   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:29.305596   235 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:07
+I0827 04:36:29.305743   235 transfer_engine_impl.cpp:388] installTransport, type=rdma
+I0827 04:36:29.306217   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+[2026-08-27 04:36:29 TP16 EP16] Init torch distributed begin.
+I0827 04:36:29.307994   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:29.308105   236 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:07
+I0827 04:36:29.308277   236 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:29 TP17 EP17] Init torch distributed begin.
+I0827 04:36:29.310081   237 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:07
+I0827 04:36:29.310261   237 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:29 TP18 EP18] Init torch distributed begin.
+I0827 04:36:29.320421   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:29.321812   239 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:07
+I0827 04:36:29.321919   239 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:29 TP20 EP20] Init torch distributed begin.
+I0827 04:36:29.323829   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:29.323894   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:29.325821   238 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:07
+I0827 04:36:29.325820   241 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:07
+I0827 04:36:29.325871   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.325872   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.360142   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:29.360253   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:29.362092   241 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:07
+I0827 04:36:29.362147   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.362201   238 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:07
+I0827 04:36:29.362243   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.392091   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:29.392130   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:29.394111   238 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:07
+I0827 04:36:29.394127   241 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:07
+I0827 04:36:29.394166   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.394286   238 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:29 TP19 EP19] Init torch distributed begin.
+I0827 04:36:29.420166   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:29.421864   241 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:07
+I0827 04:36:29.421993   241 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:29 TP22 EP22] Init torch distributed begin.
+[2026-08-27 04:36:48 TP23 EP23] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP22 EP22] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP21 EP21] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP20 EP20] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP19 EP19] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP18 EP18] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP17 EP17] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP16 EP16] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP22 EP22] Init torch distributed ends. elapsed=18.81 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP23 EP23] Init torch distributed ends. elapsed=18.94 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP21 EP21] Init torch distributed ends. elapsed=18.93 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP20 EP20] Init torch distributed ends. elapsed=18.91 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP19 EP19] Init torch distributed ends. elapsed=18.83 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP18 EP18] Init torch distributed ends. elapsed=18.92 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP17 EP17] Init torch distributed ends. elapsed=18.92 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP16 EP16] Init torch distributed ends. elapsed=18.92 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP21 EP21] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP16 EP16] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP22 EP22] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP20 EP20] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP23 EP23] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP18 EP18] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP19 EP19] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP22 EP22] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP20 EP20] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP21 EP21] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP16 EP16] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP23 EP23] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP19 EP19] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP18 EP18] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP17 EP17] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP17 EP17] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP20 EP20] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP22 EP22] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP18 EP18] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP23 EP23] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP19 EP19] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP21 EP21] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP20 EP20] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP16 EP16] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP18 EP18] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP22 EP22] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP19 EP19] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP21 EP21] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP23 EP23] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP16 EP16] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP17 EP17] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP17 EP17] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:50 TP23 EP23] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP22 EP22] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP21 EP21] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP19 EP19] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP18 EP18] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP20 EP20] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP16 EP16] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP17 EP17] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:38:53 TP16 EP16] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:53 TP16 EP16] Load weight end. elapsed=124.66 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:53 TP17 EP17] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:53 TP17 EP17] Load weight end. elapsed=124.69 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:53 TP23 EP23] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:53 TP23 EP23] Load weight end. elapsed=124.71 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:53 TP21 EP21] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:53 TP21 EP21] Load weight end. elapsed=124.74 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:53 TP18 EP18] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:53 TP18 EP18] Load weight end. elapsed=124.82 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:53 TP20 EP20] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:53 TP20 EP20] Load weight end. elapsed=124.82 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:53 TP22 EP22] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:53 TP22 EP22] Load weight end. elapsed=124.92 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:53 TP19 EP19] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:53 TP19 EP19] Load weight end. elapsed=125.03 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:39:01 TP17 EP17] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP17 EP17] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP23 EP23] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP23 EP23] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP16 EP16] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP16 EP16] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP18 EP18] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP18 EP18] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP22 EP22] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP22 EP22] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP21 EP21] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP21 EP21] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP20 EP20] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP20 EP20] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP19 EP19] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP19 EP19] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:02 TP23 EP23] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP18 EP18] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP21 EP21] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP22 EP22] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP20 EP20] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP17 EP17] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP19 EP19] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP16 EP16] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP21 EP21] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP22 EP22] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP23 EP23] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP17 EP17] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP16 EP16] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP20 EP20] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP18 EP18] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP19 EP19] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP21 EP21] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP23 EP23] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP22 EP22] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP17 EP17] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP23 EP23] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP21 EP21] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP20 EP20] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP16 EP16] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP22 EP22] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP17 EP17] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP20 EP20] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP16 EP16] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP18 EP18] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP19 EP19] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP18 EP18] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP19 EP19] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP23 EP23] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP23 EP23] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:39:02 TP17 EP17] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP17 EP17] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP16 EP16] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP16 EP16] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:39:02 TP21 EP21] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP21 EP21] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP22 EP22] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP22 EP22] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP20 EP20] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP20 EP20] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:39:02 TP18 EP18] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP18 EP18] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP19 EP19] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP19 EP19] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:42:56 TP19 EP19] Capture target decode CUDA graph end. elapsed=233.80 s, mem usage=0.62 GB, avail mem=8.51 GB.
+[2026-08-27 04:42:56 TP19 EP19] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP16 EP16] Capture target decode CUDA graph end. elapsed=233.84 s, mem usage=0.60 GB, avail mem=8.52 GB.
+[2026-08-27 04:42:56 TP16 EP16] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP22 EP22] Capture target decode CUDA graph end. elapsed=233.84 s, mem usage=0.56 GB, avail mem=8.59 GB.
+[2026-08-27 04:42:56 TP22 EP22] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP17 EP17] Capture target decode CUDA graph end. elapsed=233.88 s, mem usage=0.58 GB, avail mem=8.58 GB.
+[2026-08-27 04:42:56 TP17 EP17] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP21 EP21] Capture target decode CUDA graph end. elapsed=233.90 s, mem usage=0.58 GB, avail mem=8.58 GB.
+[2026-08-27 04:42:56 TP21 EP21] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP20 EP20] Capture target decode CUDA graph end. elapsed=233.91 s, mem usage=0.60 GB, avail mem=8.52 GB.
+[2026-08-27 04:42:56 TP20 EP20] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP23 EP23] Capture target decode CUDA graph end. elapsed=234.01 s, mem usage=0.62 GB, avail mem=8.51 GB.
+[2026-08-27 04:42:56 TP23 EP23] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP18 EP18] Capture target decode CUDA graph end. elapsed=234.00 s, mem usage=0.56 GB, avail mem=8.59 GB.
+[2026-08-27 04:42:56 TP18 EP18] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+W0827 04:42:57.619849  2423 memory_location.cpp:73] Failed to get NUMA node, addr: 0x9680d380, len: 23168: Operation not permitted [1]
+W0827 04:42:57.619855  2422 memory_location.cpp:73] Failed to get NUMA node, addr: 0x9696f4c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.619967  2424 memory_location.cpp:73] Failed to get NUMA node, addr: 0x966bf4c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.620213  2425 memory_location.cpp:73] Failed to get NUMA node, addr: 0x82b481c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.620281  2426 memory_location.cpp:73] Failed to get NUMA node, addr: 0x988cf580, len: 185344: Operation not permitted [1]
+W0827 04:42:57.620285  2427 memory_location.cpp:73] Failed to get NUMA node, addr: 0x988fc9c0, len: 185344: Operation not permitted [1]
+W0827 04:42:57.620325  2428 memory_location.cpp:73] Failed to get NUMA node, addr: 0x96988ec0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.620384  2429 memory_location.cpp:73] Failed to get NUMA node, addr: 0x96949540, len: 46336: Operation not permitted [1]
+W0827 04:42:57.620546  2430 memory_location.cpp:73] Failed to get NUMA node, addr: 0x96790780, len: 23168: Operation not permitted [1]
+W0827 04:42:57.620606  2431 memory_location.cpp:73] Failed to get NUMA node, addr: 0x967ccac0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.651996  2570 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6c8ac200, len: 23168: Operation not permitted [1]
+W0827 04:42:57.652108  2571 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6c6f5f80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.652212  2572 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6c8ea380, len: 23168: Operation not permitted [1]
+W0827 04:42:57.652417  2573 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6d2e8ec0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.652568  2574 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6ea4eb80, len: 185344: Operation not permitted [1]
+W0827 04:42:57.652587  2576 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6cb20d80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.652642  2577 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6c987340, len: 46336: Operation not permitted [1]
+W0827 04:42:57.652667  2575 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6ea7bfc0, len: 185344: Operation not permitted [1]
+W0827 04:42:57.652815  2578 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6c6eb200, len: 23168: Operation not permitted [1]
+W0827 04:42:57.652889  2579 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6cb0a2c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.713306  2681 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8a7feb00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.713331  2680 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ad55240, len: 23168: Operation not permitted [1]
+W0827 04:42:57.713303  2679 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ad28280, len: 23168: Operation not permitted [1]
+W0827 04:42:57.713424  2682 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8abc1200, len: 23168: Operation not permitted [1]
+W0827 04:42:57.713523  2683 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ca64980, len: 185344: Operation not permitted [1]
+W0827 04:42:57.713629  2685 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ca91dc0, len: 185344: Operation not permitted [1]
+W0827 04:42:57.713903  2687 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ac54c00, len: 46336: Operation not permitted [1]
+W0827 04:42:57.714316  2688 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ab93e40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.714335  2689 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8abe8700, len: 23168: Operation not permitted [1]
+W0827 04:42:57.714352  2686 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8a7f7880, len: 23168: Operation not permitted [1]
+W0827 04:42:57.717235  2716 memory_location.cpp:73] Failed to get NUMA node, addr: 0x739fc3c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.717247  2718 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ff359c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.717484  2725 memory_location.cpp:73] Failed to get NUMA node, addr: 0x75e8fa00, len: 185344: Operation not permitted [1]
+W0827 04:42:57.717736  2729 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ff46700, len: 46336: Operation not permitted [1]
+W0827 04:42:57.718014  2728 memory_location.cpp:73] Failed to get NUMA node, addr: 0x73b6a280, len: 23168: Operation not permitted [1]
+W0827 04:42:57.718112  2730 memory_location.cpp:73] Failed to get NUMA node, addr: 0x73dda340, len: 23168: Operation not permitted [1]
+W0827 04:42:57.718163  2731 memory_location.cpp:73] Failed to get NUMA node, addr: 0x73dbf000, len: 23168: Operation not permitted [1]
+W0827 04:42:57.718196  2721 memory_location.cpp:73] Failed to get NUMA node, addr: 0x73a8f640, len: 23168: Operation not permitted [1]
+W0827 04:42:57.718261  2723 memory_location.cpp:73] Failed to get NUMA node, addr: 0x73c5ba40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.724411  2727 memory_location.cpp:73] Failed to get NUMA node, addr: 0x75ebce40, len: 185344: Operation not permitted [1]
+W0827 04:42:57.755818  2858 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5e996800, len: 23168: Operation not permitted [1]
+W0827 04:42:57.755931  2859 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ec52140, len: 23168: Operation not permitted [1]
+W0827 04:42:57.756026  2861 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5e767ec0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.756074  2862 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60bbb440, len: 185344: Operation not permitted [1]
+W0827 04:42:57.756379  2865 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4aecca80, len: 46336: Operation not permitted [1]
+W0827 04:42:57.756517  2863 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60be8880, len: 185344: Operation not permitted [1]
+W0827 04:42:57.756536  2864 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5eb0d880, len: 23168: Operation not permitted [1]
+W0827 04:42:57.756650  2866 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ea7fe80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.757373  2860 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5eb4d800, len: 23168: Operation not permitted [1]
+W0827 04:42:57.757380  2867 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ea97dc0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.818434  3070 memory_location.cpp:73] Failed to get NUMA node, addr: 0x965568c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.818459  3075 memory_location.cpp:73] Failed to get NUMA node, addr: 0x96700fc0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.818470  3072 memory_location.cpp:73] Failed to get NUMA node, addr: 0x96541bc0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.820174  3079 memory_location.cpp:73] Failed to get NUMA node, addr: 0x984fb440, len: 185344: Operation not permitted [1]
+W0827 04:42:57.822142  3084 memory_location.cpp:73] Failed to get NUMA node, addr: 0x962df200, len: 23168: Operation not permitted [1]
+W0827 04:42:57.822369  3086 memory_location.cpp:73] Failed to get NUMA node, addr: 0x831c0780, len: 46336: Operation not permitted [1]
+W0827 04:42:57.822384  3110 memory_location.cpp:73] Failed to get NUMA node, addr: 0x965cd3c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.822431  3082 memory_location.cpp:73] Failed to get NUMA node, addr: 0x98528880, len: 185344: Operation not permitted [1]
+W0827 04:42:57.822527  3127 memory_location.cpp:73] Failed to get NUMA node, addr: 0x9642bbc0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.829262  3077 memory_location.cpp:73] Failed to get NUMA node, addr: 0x9640b580, len: 23168: Operation not permitted [1]
+W0827 04:42:57.832078  3193 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ce567c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.832136  3194 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8cbe8d00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.832149  3195 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8cc94640, len: 23168: Operation not permitted [1]
+W0827 04:42:57.832294  3196 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8d79f140, len: 23168: Operation not permitted [1]
+W0827 04:42:57.832419  3199 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ce4bec0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.832562  3201 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8d7e4b40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.832683  3202 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8cbbd940, len: 23168: Operation not permitted [1]
+W0827 04:42:57.833209  3198 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ef88080, len: 185344: Operation not permitted [1]
+W0827 04:42:57.833267  3200 memory_location.cpp:73] Failed to get NUMA node, addr: 0x79c278c0, len: 46336: Operation not permitted [1]
+W0827 04:42:57.837312  3197 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ef5ac40, len: 185344: Operation not permitted [1]
+W0827 04:42:58.193480  3458 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59a23f40, len: 23168: Operation not permitted [1]
+W0827 04:42:58.193507  3459 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59aa70c0, len: 23168: Operation not permitted [1]
+W0827 04:42:58.193545  3460 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5999c100, len: 23168: Operation not permitted [1]
+W0827 04:42:58.193585  3461 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59bc3040, len: 23168: Operation not permitted [1]
+W0827 04:42:58.193631  3462 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5bd1e5c0, len: 185344: Operation not permitted [1]
+W0827 04:42:58.193667  3463 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5bd4ba00, len: 185344: Operation not permitted [1]
+W0827 04:42:58.193720  3464 memory_location.cpp:73] Failed to get NUMA node, addr: 0x599fa280, len: 23168: Operation not permitted [1]
+W0827 04:42:58.193746  3466 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59dd6100, len: 23168: Operation not permitted [1]
+W0827 04:42:58.193764  3467 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59cbd400, len: 23168: Operation not permitted [1]
+W0827 04:42:58.193776  3465 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59db0340, len: 46336: Operation not permitted [1]
+[2026-08-27 04:43:00] Dummy health check server started in background thread at 0.0.0.0:30000
+I0827 04:48:13.990782  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:13.991083  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.005937  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.006215  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.026726  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.034708  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.038136  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.038393  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.052722  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.052948  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.069641  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.089646  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.092875  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.099138  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.099377  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.099617  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.102869  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.103116  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.103413  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.103614  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.103897  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.104102  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.104316  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.107553  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.107847  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.113867  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.125564  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.131559  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.131772  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.140426  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.140640  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.140889  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.141057  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.144480  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.144747  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.160434  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.160689  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.160965  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.177232  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.188812  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.189050  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.189292  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.189566  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.189749  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.197841  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.198038  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.198292  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.198523  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.198736  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.198998  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.210616  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.210834  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.211103  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.214471  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.216403  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.217782  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.227747  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.228065  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.228319  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.228518  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.230293  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.244758  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.244995  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.245216  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.246076  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.247792  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.248039  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.248274  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.251454  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.255178  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.255391  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.256364  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.256808  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.260069  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.260272  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.260452  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.260666  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.260878  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.266151  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.266335  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.266537  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.266744  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.266952  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.267238  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.267465  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.268203  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.270165  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.272876  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.273144  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.275923  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.276090  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.281610  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.281848  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.285796  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.285949  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.287339  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.289616  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.289852  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.290091  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.290292  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.290467  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.290657  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.290853  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.291211  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.295532  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.297179  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.297688  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.299065  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.299283  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.300338  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.309764  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.310025  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.310344  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.310527  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.310770  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.314235  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.317247  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.318601  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.324312  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.338757  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.343169  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.344022  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.344889  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.353394  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.354696  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.354852  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.360512  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.363333  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.363579  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.363832  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.364054  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.364255  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.364542  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.364830  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.368968  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.369174  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.369390  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.373200  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.374004  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.374215  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.374486  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.379066  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.379289  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.383756  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.384938  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.393990  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.394248  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.394496  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.400096  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.400372  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.400589  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.400769  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.400981  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.401257  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.401494  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.401587  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.402990  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.409467  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.409701  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.410995  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.412505  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.423301  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.423527  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.423753  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.425726  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.426009  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.426232  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.429703  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.432900  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.436066  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.437069  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.437340  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.437568  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.437778  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.438095  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.441175  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.442378  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.443518  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.443746  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.443964  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.444185  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.444403  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.453183  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.454483  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.458822  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.460139  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.460391  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.461473  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.468333  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.469964  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.474977  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.475178  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.475373  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.475994  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.476274  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.482146  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.483319  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.483541  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.483824  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.487998  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.491807  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.492008  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.492403  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.492631  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.492843  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.493103  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.493304  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.493522  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.493757  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.493925  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.493986  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.494186  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.494228  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.494501  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.494679  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.494771  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.494979  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.495187  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.495486  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.495695  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.495930  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.496155  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.496392  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.498075  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.499614  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.499698  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.507068  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.507290  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.515178  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.515554  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.515731  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.515975  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.516929  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.517546  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.519863  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.520967  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.522394  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.522583  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.522799  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.523052  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.523262  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.523473  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.523695  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.523783  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.523902  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.524128  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.524420  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.524629  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.525529  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.525698  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.526990  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.535063  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.535307  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.535513  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.536439  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.536978  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.537204  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.537415  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.537595  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.539846  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.540122  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.540378  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.540510  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.540766  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.541023  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.541293  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.544811  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.547500  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.547943  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.548959  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.551965  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.553215  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.563808  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.564100  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.568046  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.569260  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.569701  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.571031  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.573542  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.579483  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.581025  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.587301  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.587486  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.587728  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.588649  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.594245  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.594436  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.594633  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.594759  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.596094  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.596485  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.596745  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.597023  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.597223  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.597502  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.597754  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.598001  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.598238  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.598425  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.598681  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.598914  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.599107  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.599298  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.601323  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.601538  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.602514  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.606913  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.611363  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.611377  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.612835  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.615579  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.616807  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.616865  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.616990  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.617111  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.617369  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.617590  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.617800  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.617998  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.618206  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.621685  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.623104  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.633111  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.633315  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.634286  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.634464  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.634523  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.634792  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.634966  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.635149  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.635445  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.635624  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.635821  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.636071  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.637221  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.643270  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.644847  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.648780  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.648998  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.656845  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.657044  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.657281  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.658142  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.660220  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.661625  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.664076  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.665403  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.667707  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.669119  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.670938  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.677649  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.677932  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.677983  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.678122  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.678347  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.678530  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.678735  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.678929  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.679240  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.679463  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.679695  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.679894  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.680079  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.680286  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.680457  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.680696  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.681843  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.683002  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.683204  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.683274  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.683362  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.683475  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.683499  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.683585  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.683856  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.684543  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.687750  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.689368  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.690239  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.691855  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.693795  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.696607  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.696817  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.698160  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.703644  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.706352  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.706555  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.706766  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.707012  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.707203  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.707451  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.707659  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.707913  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.708099  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.708461  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.708745  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.708987  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.709229  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.713573  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.714908  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.718238  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.718487  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.718716  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.719660  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.720849  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.722200  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.724702  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.725086  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.725322  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.725628  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.725817  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.726325  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.731271  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.736089  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.736286  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.736464  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.737399  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.740564  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.742311  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.743717  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.746318  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.746944  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.747165  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.748356  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.749109  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.749446  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.749665  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.750972  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.755626  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.756073  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.756265  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.757405  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.762998  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.764305  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.765604  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.765775  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.766307  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.766513  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.766744  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.766980  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.767177  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.767390  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.767621  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.767817  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.775632  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.778723  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.778945  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.779145  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.780236  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.783447  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.786221  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.786506  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.789175  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.789419  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.790586  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.790643  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.790793  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.790974  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.791939  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.796165  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.797827  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.802098  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.813362  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.814442  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.814667  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.814668  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.814859  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.820125  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.820905  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.821156  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.822324  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.832154  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.832475  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.832693  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.832880  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.833066  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.833330  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.833523  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.833789  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.834038  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.834266  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.834484  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.834671  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.834861  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.835052  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.835216  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.835410  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.838090  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.848765  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.849062  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.849324  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.849581  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.849845  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.850172  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.858408  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.858647  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.859768  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.860574  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.860812  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.861027  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.861253  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.861526  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.861723  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.861910  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.862110  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.862299  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.862488  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.864478  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.872546  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.872754  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.872980  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.873731  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.873880  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.873962  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.874325  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.875612  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.877731  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.879717  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.880618  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.885710  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.886065  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.886333  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.890219  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.891563  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.895009  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.903152  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.905467  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.913136  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.915786  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.916013  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.918720  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.918928  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.919982  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.920882  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.921152  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.921317  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.921556  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.921797  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.921981  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.922211  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.924968  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.925161  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.926721  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.930300  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.931375  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.932979  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.933200  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.935876  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.941114  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.943547  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.943744  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.943778  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.943972  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.944057  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.944993  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.954442  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.955266  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.955477  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.955673  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.956692  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.974392  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.974649  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.974893  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.975108  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.985150  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.985353  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.985610  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.985852  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.986055  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.986230  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.986440  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.986611  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.991794  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.991982  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.992164  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.992372  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.004820  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.007462  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.012593  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.012835  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.013056  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.013247  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.013500  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.013749  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.016324  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.016539  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.021533  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.024173  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.024427  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.025575  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.067353  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.067580  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.067747  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.067951  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.068106  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.068390  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.068591  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.068789  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.078953  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.079149  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.079314  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.080277  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.090672  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.090881  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.091123  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.092058  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.132000  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.132241  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.132424  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.132637  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.132885  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.133169  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.133404  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.133606  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.133805  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.133991  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.134207  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.134473  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.134696  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.134949  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.135172  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.135368  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.160171  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.160442  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.160633  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.160794  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.161010  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.161175  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.161370  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.161597  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.161856  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.162032  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.171926  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.172152  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.172313  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.173470  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.217530  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.217757  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.217985  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.218160  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.218354  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.218546  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.218766  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.218966  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.229063  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.229290  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.229496  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.230377  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.240563  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.240742  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.240916  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.241938  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.282371  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.282605  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.282783  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.282986  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.283192  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.283366  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.283558  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.283744  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.283931  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.284118  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.284356  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.284595  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.284804  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.285007  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.285272  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.290516  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.310380  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.310582  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.310774  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.310997  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.311203  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.311424  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.311609  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.311781  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.311981  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.312191  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.322089  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.322285  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.322484  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.323439  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.366820  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.367026  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.367223  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.367421  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.367637  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.367813  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.368016  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.368258  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.378153  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.378351  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.378584  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.379503  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.389467  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.389665  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.389904  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.390815  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.430491  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.430742  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.430959  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.431205  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.431394  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.431623  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.431800  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.432049  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.432286  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.432466  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.432688  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.432901  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.433138  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.433328  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.433526  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.436129  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.458094  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.458309  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.458535  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.458752  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.458948  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.459110  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.459302  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.459517  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.459699  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.459882  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.469916  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.470148  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.470368  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.471292  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.448522  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.450094  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.460287  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.460638  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.462020  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.466852  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.473448  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.481580  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.482398  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.482719  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.483039  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.483215  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.483292  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.483968  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.488898  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.493845  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.494148  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.495286  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.501287  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.504213  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.512477  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.512745  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.513012  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.513972  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.297312  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.299115  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.304275  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.314786  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.316474  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.328058  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.328325  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.328570  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.329540  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.338943  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.339195  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.339458  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.340605  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.341210  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.346589  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.357823  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.358044  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.358083  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.358332  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.358369  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.359782  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.363950  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.366775  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.367130  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.368332  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.369814  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.374266  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.375430  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.378304  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.380679  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.381263  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.381505  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.381752  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.382050  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.382229  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.382687  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.386806  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.386886  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.388373  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.392937  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.394059  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.395437  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.395885  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.396198  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.397209  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.397620  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.398900  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.399183  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.399243  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.400318  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.401305  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.403239  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.404074  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.404656  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.409005  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.411810  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.414434  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.415934  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.417624  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.417887  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.418118  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.420249  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.423964  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.424198  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.424408  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.430177  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.433308  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.433637  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.436582  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.439628  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.440941  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.394086  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.394385  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.395802  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.408921  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.409214  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.410487  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.420368  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.420595  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.421932  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.426435  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.573661  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.574067  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.575153  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.109197  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.110759  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.117308  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.127852  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.129487  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.134262  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.139751  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.141630  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.146772  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.148381  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.148725  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.149887  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.154366  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.161194  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.162659  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.168562  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.654505  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.655917  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.669334  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.669617  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.669867  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.670902  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.679297  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.680840  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.687160  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.692883  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.693225  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.693225  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.693482  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.693727  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.694342  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.694828  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.699385  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.700112  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.701340  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.702790  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.708935  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.719955  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.720265  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.720535  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.721411  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.727005  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.728433  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.462263  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.462824  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.463836  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.475934  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.477473  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.481695  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.487771  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.489451  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.495987  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.185277  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.186878  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.193236  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:47.566026  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:47.575708  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:47.584995  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:48.551697  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:48.561645  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:48.570935  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:49.336337  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:49.347584  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:49.348870  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:49.353668  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:49.362033  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:50.265223  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:50.274431  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:50.283521  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.110039  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.122382  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.127803  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.131213  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.163244  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.171993  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.179919  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.997390  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.007871  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.013643  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.018814  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.023959  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.033485  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.799268  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.810883  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.819684  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:53.708249  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:53.717978  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:53.727499  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:53.735821  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:54.567750  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:54.576763  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:55.320358  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:55.447818  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:55.457465  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:57.588092  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:57.606285  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:57.611370  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:57.639269  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:57.640733  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:57.646224  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:57.660796  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:57.732604  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:57.734110  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:58.280643  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:58.297982  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:59.042151  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:59.051748  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:08.244721  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:08.253723  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:09.194365  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:10.135797  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:10.203588  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:10.219120  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:10.975451  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:10.985549  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:11.843453  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:11.845074  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:12.790898  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:13.633699  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:13.643146  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:15.418679  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:17.010130  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:17.183473  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:17.197549  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:17.986510  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:18.003332  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:18.214159  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:18.224337  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:18.241684  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:18.243786  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:18.258334  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:18.350939  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:18.364851  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:18.893000  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:29.860329  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:46.812764  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-c8-20260827-1340/logs/manual_d_node3.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-c8-20260827-1340/logs/manual_d_node3.log
new file mode 100644
index 0000000..a9edef0
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-c8-20260827-1340/logs/manual_d_node3.log
@@ -0,0 +1,1462 @@
+/usr/local/lib/python3.12/dist-packages/torchao/quantization/quant_api.py:1731: SyntaxWarning: invalid escape sequence '\.'
+  """Configuration class for applying different quantization configs to modules or parameters based on their fully qualified names (FQNs).
+/sgl-workspace/sglang/python/sglang/launch_server.py:56: UserWarning: 'python -m sglang.launch_server' is still supported, but 'sglang serve' is the recommended entrypoint.
+  Example: sglang serve --model-path <model> [options]
+  warnings.warn(
+[2026-08-27 04:36:00] KV cache is forced as chunk cache for decode server
+[2026-08-27 04:36:03] Breakable CUDA graph is incompatible with MLA attention; disabling prefill CUDA graph.
+[2026-08-27 04:36:03] Attention backend not specified. Use triton backend by default.
+[2026-08-27 04:36:05] server_args=ServerArgs(model_path='/data/hf_models/Kimi-K3', tokenizer_path='/data/hf_models/Kimi-K3', tokenizer_mode='auto', tokenizer_backend='huggingface', tokenizer_worker_num=1, detokenizer_worker_num=1, skip_tokenizer_init=False, load_format='auto', model_loader_extra_config='{}', trust_remote_code=True, context_length=None, is_embedding=False, enable_multimodal=None, revision=None, model_impl='auto', model_config_parser='auto', json_model_override_args='{}', dtype='auto', quantization=None, quantization_param_path=None, kv_cache_dtype='auto', enable_fp32_lm_head=False, modelopt_quant=None, modelopt_checkpoint_restore_path=None, modelopt_checkpoint_save_path=None, modelopt_export_path=None, quantize_and_serve=False, rl_quant_profile=None, enable_tf32_matmul=False, mem_fraction_static=0.88, max_running_requests=None, max_queued_requests=None, max_total_tokens=None, chunked_prefill_size=8192, enable_dynamic_chunking=False, max_prefill_tokens=16384, prefill_max_requests=None, schedule_policy='fcfs', enable_priority_scheduling=False, disable_priority_preemption=False, default_priority_value=None, abort_on_priority_when_disabled=False, schedule_low_priority_values_first=False, priority_scheduling_preemption_threshold=10, retraction_policy='length', schedule_conservativeness=1.0, page_size=64, swa_full_tokens_ratio=0.8, disable_hybrid_swa_memory=False, radix_eviction_policy='lru', prefill_only_disable_kv_cache=False, disable_radix_cache=True, enable_page_major_kv_layout=False, enable_unified_memory=False, disable_chunked_prefix_cache=False, disable_overlap_schedule=False, num_continuous_decode_steps=1, scheduler_recv_interval=1, enable_mixed_chunk=False, nccl_port=None, dist_timeout=3600, dist_init_addr='174.1.60.5:20000', nnodes=4, node_rank=3, tp_size=32, dcp_size=1, pp_size=1, pp_max_micro_batch_size=None, pp_async_batch_depth=0, dp_size=1, load_balance_method='round_robin', attn_cp_size=1, moe_dp_size=1, dcp_comm_backend='ag_rs', dcp_replicate_q_proj=None, dwdp_size=1, enable_prefill_cp=False, cp_strategy=None, enable_dsa_cache_layer_split=False, enable_dsa_prefill_context_parallel=False, dsa_prefill_cp_mode='round-robin-split', enable_prefill_context_parallel=False, prefill_cp_mode='in-seq-split', enable_cp_decode_attn_tp=False, enable_dp_attention=False, enable_dp_attention_local_control_broadcast=False, enable_dp_lm_head=False, enable_attn_tp_input_scattered=False, disable_attn_tp_gather=False, enable_p2p_check=False, device='cuda', base_gpu_id=0, gpu_id_step=1, random_seed=297832107, watchdog_timeout=300, soft_watchdog_timeout=None, sleep_on_idle=False, use_ray=False, custom_sigquit_handler=None, numa_node=None, gc_threshold=None, host='0.0.0.0', port=30000, fastapi_root_path='', smg_grpc_mode=False, grpc_mode=False, grpc_port=None, sidecar=None, sidecar_args=None, skip_server_warmup=False, warmups=None, enable_http2=False, ssl_keyfile=None, ssl_certfile=None, ssl_ca_certs=None, ssl_keyfile_password=None, enable_ssl_refresh=False, api_key=None, admin_api_key=None, served_model_name='kimi-k3', weight_version='default', chat_template=None, hf_chat_template_name=None, completion_template=None, file_storage_path='sglang_storage', enable_cache_report=False, reasoning_parser=None, default_chat_template_kwargs=None, strip_thinking_cache=False, enable_strict_thinking=False, tool_call_parser=None, tool_server=None, sampling_defaults='model', asr_max_buffer_seconds=60, asr_max_concurrent_sessions=32, preferred_sampling_params=None, allow_auto_truncate=False, stream_interval=1, batch_notify_size=16, stream_response_default_include_usage=False, incremental_streaming_output=False, enable_streaming_session=False, enable_session_radix_cache=False, log_level='info', log_level_http=None, log_requests=False, log_requests_level=2, log_requests_format='text', log_requests_target=None, uvicorn_access_log_exclude_prefixes=[], crash_dump_folder=None, show_time_cost=False, enable_metrics=False, smg_http_sidecar_port=None, enable_mfu_metrics=False, enable_metrics_for_all_schedulers=False, load_snapshot_publish_interval=15, tokenizer_metrics_custom_labels_header='x-custom-labels', tokenizer_metrics_allowed_custom_labels=None, extra_metric_labels=None, bucket_time_to_first_token=None, bucket_inter_token_latency=None, bucket_e2e_request_latency=None, prompt_tokens_buckets=None, generation_tokens_buckets=None, gc_warning_threshold_secs=0.0, decode_log_interval=40, enable_request_time_stats_logging=False, kv_events_config=None, enable_forward_pass_metrics=False, forward_pass_metrics_worker_id='', forward_pass_metrics_ipc_name=None, enable_trace=False, trace_modules='request', otlp_traces_endpoint='localhost:4317', export_metrics_to_file=False, export_metrics_to_file_dir=None, stat_loggers=None, constrained_json_whitespace_pattern=None, constrained_json_disable_any_whitespace=False, attention_backend='triton', decode_attention_backend=None, prefill_attention_backend=None, sampling_backend='flashinfer', grammar_backend='xgrammar', radix_cache_backend=None, mm_attention_backend=None, fp8_gemm_runner_backend='auto', fp4_gemm_runner_backend='auto', bf16_gemm_backend='auto', dsa_prefill_backend=None, dsa_decode_backend=None, dsa_paged_mqa_logits_backend='auto', dsa_topk_backend='sgl-kernel', disable_flashinfer_autotune=False, flashinfer_autotune_skip_ops=None, mamba_backend='triton', cuda_graph_config=CudaGraphConfig(decode=PhaseConfig(backend='full', max_bs=16, bs=[1, 2, 4, 8, 12, 16], tc_compiler='eager', full_prefill_max_req=None), prefill=PhaseConfig(backend='disabled', max_bs=2048, bs=[4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], tc_compiler='eager', full_prefill_max_req=None)), cuda_graph_backend_decode=None, cuda_graph_backend_prefill=None, cuda_graph_max_bs_decode=16, cuda_graph_max_bs_prefill=None, cuda_graph_bs_decode=None, cuda_graph_bs_prefill=None, cuda_graph_tc_compiler=None, disable_prefill_cuda_graph=False, disable_decode_cuda_graph=False, disable_cuda_graph=False, disable_cuda_graph_padding=False, enable_profile_cuda_graph=False, enable_cudagraph_gc=False, debug_cuda_graph=False, enable_layerwise_nvtx_marker=False, enable_nccl_nvls=False, enable_symm_mem=False, triton_attention_reduce_in_fp32=False, triton_attention_num_kv_splits=8, triton_attention_split_tile_size=None, flashinfer_mla_disable_ragged=False, enable_fused_qk_norm_rope=False, enable_precise_embedding_interpolation=False, enable_fused_moe_sum_all_reduce=False, enable_deepseek_v4_fp4_indexer=False, disable_custom_all_reduce=False, enable_mscclpp=False, enable_torch_symm_mem=False, enable_scattered_sconv=False, pre_warm_nccl=False, enable_quant_communications=False, enable_flashinfer_allreduce_fusion=False, enforce_disable_flashinfer_allreduce_fusion=False, flashinfer_allreduce_fusion_backend=None, enable_aiter_allreduce_fusion=False, enable_torch_compile=False, enable_torch_compile_debug_mode=False, torch_compile_max_bs=32, torchao_config='', speculative_algorithm=None, speculative_draft_model_path=None, speculative_draft_model_revision=None, speculative_draft_load_format=None, speculative_num_steps=None, speculative_eagle_topk=None, speculative_num_draft_tokens=None, speculative_dflash_block_size=None, speculative_dspark_block_size=None, speculative_dspark_sps_table_path=None, speculative_dspark_confidence_sts_path=None, speculative_dspark_align_verify_tokens_to_graph_tier=False, speculative_accept_threshold_single=1.0, speculative_accept_threshold_acc=1.0, speculative_use_rejection_sampling=False, speculative_token_map=None, speculative_attention_mode='prefill', speculative_draft_attention_backend=None, speculative_draft_window_size=None, speculative_moe_runner_backend='marlin', speculative_moe_a2a_backend=None, speculative_draft_model_quantization=None, speculative_skip_dp_mlp_sync=False, enable_multi_layer_eagle=False, speculative_adaptive=False, speculative_adaptive_config=None, decoupled_spec_bind_endpoint=None, decoupled_spec_connect_endpoints=None, decoupled_spec_rank=None, decoupled_spec_role='null', spec_trace_dir=None, speculative_ngram_min_bfs_breadth=1, speculative_ngram_max_bfs_breadth=10, speculative_ngram_match_type='BFS', speculative_ngram_max_trie_depth=18, speculative_ngram_capacity=10000000, speculative_ngram_external_corpus_path=None, speculative_ngram_external_sam_budget=0, speculative_ngram_external_corpus_max_tokens=10000000, ep_size=32, moe_a2a_backend='none', moe_runner_backend='marlin', flashinfer_mxfp4_moe_precision='default', deepep_mode='auto', fuseep_mode=2, deepep_dispatcher_output_dtype='auto', ep_num_redundant_experts=0, ep_dispatch_algorithm=None, init_expert_location='trivial', enable_eplb=False, eplb_algorithm='auto', eplb_rebalance_num_iterations=1000, eplb_rebalance_layers_per_chunk=None, eplb_min_rebalancing_utilization_threshold=1.0, expert_distribution_recorder_mode=None, expert_distribution_recorder_buffer_size=1000, enable_expert_distribution_metrics=False, deepep_config=None, moe_dense_tp_size=None, elastic_ep_backend=None, enable_elastic_expert_backup=False, mooncake_ib_device=None, enable_waterfill=False, ep_join_mode=None, ep_join_rank_offset=0, elastic_ep_initial_size=None, max_ep_size=None, elastic_ep_scale_timeout=600, elastic_ep_rejoin=False, disable_flashinfer_cutlass_moe_fp4_allgather=False, disable_shared_experts_fusion=False, enforce_shared_experts_fusion=False, max_mamba_cache_size=None, mamba_ssm_dtype=None, mamba_max_states_per_path=-1, enable_mamba_cache_stochastic_rounding=False, mamba_cache_philox_rounds=0, mamba_full_memory_ratio=0.21, mamba_radix_cache_strategy='extra_buffer_lazy', uses_mamba_radix_cache=False, mamba_track_interval=256, enable_int8_mamba_checkpoint=False, int8_mamba_ckpt_size=None, linear_attn_backend='triton', linear_attn_decode_backend=None, linear_attn_prefill_backend=None, linear_attn_verify_backend=None, enable_linear_replayssm=False, linear_replayssm_cache_len=16, enable_linear_replayssm_spec=False, enable_hierarchical_cache=False, hicache_ratio=2.0, hicache_size=0, hicache_write_policy='write_through', hicache_io_backend='kernel', hicache_mem_layout='page_first', hicache_storage_backend=None, hicache_storage_prefetch_policy='timeout', hicache_storage_backend_extra_config=None, enable_hisparse=False, hisparse_config=None, enable_broadcast_mm_inputs_process=False, enable_prefix_mm_cache=False, mm_enable_dp_encoder=False, mm_process_config={}, mm_processor_worker_num=0, mm_io_worker_num=0, limit_mm_data_per_request=None, enable_mm_global_cache=False, disable_fast_image_processor=False, mm_feature_transport='cpu', keep_mm_feature_on_device=False, enable_lora=None, enable_lora_overlap_loading=None, max_lora_rank=None, lora_target_modules=None, lora_paths=None, max_loaded_loras=None, max_loras_per_batch=8, lora_eviction_policy='lru', lora_backend='csgmv', max_lora_chunk_size=16, experts_shared_outer_loras=None, lora_use_virtual_experts=False, lora_strict_loading=False, lora_drain_wait_threshold=0.0, enable_two_batch_overlap=False, enable_single_batch_overlap=False, tbo_token_distribution_threshold=0.48, cpu_offload_gb=0, offload_group_size=-1, offload_num_in_group=1, offload_prefetch_step=1, offload_mode='cpu', enable_lmcache=False, lmcache_config_file=None, enable_flexkv=False, flexkv_config_file=None, kt_weight_path=None, kt_method='AMXINT4', kt_cpuinfer=None, kt_threadpool_count=2, kt_num_gpu_experts=None, kt_max_deferred_experts_per_token=None, dllm_algorithm=None, dllm_algorithm_config=None, dllm_fdfo=True, disaggregation_mode='decode', disaggregation_transfer_backend='mooncake', disaggregation_bootstrap_port=28800, disaggregation_ib_device='mlx5_0,mlx5_1,mlx5_2,mlx5_3', disaggregation_decode_enable_radix_cache=False, disaggregation_decode_enable_offload_kvcache=False, num_reserved_decode_tokens=512, disaggregation_decode_extra_slots=0, disaggregation_decode_polling_interval=1, optimistic_prefill_attempts=0, encoder_only=False, language_only=False, encoder_transfer_backend='zmq_to_tokenizer', encoder_urls=[], encoder_bootstrap_port=8997, encoder_register_urls=[], enable_adaptive_dispatch_to_encoder=False, enable_pdmux=False, pdmux_config_path=None, sm_group_num=8, custom_weight_loader=[], weight_loader_disable_mmap=False, weight_loader_prefetch_checkpoints=False, weight_loader_prefetch_num_threads=4, weight_loader_drop_cache_after_load=False, remote_instance_weight_loader_seed_instance_ip=None, remote_instance_weight_loader_seed_instance_service_port=None, remote_instance_weight_loader_send_weights_group_ports=None, remote_instance_weight_loader_backend='nccl', remote_instance_weight_loader_start_seed_via_transfer_engine=False, engine_info_bootstrap_port=6789, modelexpress_config=None, download_dir=None, model_checksum=None, delete_ckpt_after_loading=False, decrypted_config_file=None, decrypted_draft_config_file=None, checkpoint_engine_wait_weights_before_ready=False, enable_prefill_delayer=False, prefill_delayer_max_delay_passes=30, prefill_delayer_token_usage_low_watermark=None, prefill_delayer_forward_passes_buckets=None, prefill_delayer_wait_seconds_buckets=None, prefill_delayer_queue_min_ratio=None, prefill_delayer_max_delay_ms=None, min_free_slots_delay=None, enable_deterministic_inference=False, rl_on_policy_target=None, kv_canary='none', kv_canary_real_data='none', kv_canary_sweep_interval=0, enable_dynamic_batch_tokenizer=False, dynamic_batch_tokenizer_batch_size=32, dynamic_batch_tokenizer_batch_timeout=0.002, enable_tokenizer_batch_encode=False, disable_tokenizer_batch_decode=False, debug_tensor_dump_output_folder=None, debug_tensor_dump_layers=None, debug_tensor_dump_input_file=None, enable_memory_saver=False, enable_weights_cpu_backup=False, enable_draft_weights_cpu_backup=False, enable_custom_logit_processor=False, enable_return_hidden_states=False, enable_return_routed_experts=False, enable_return_indexer_topk=False, disable_outlines_disk_cache=False, enable_mis=False, weight_cache_mode='off', weight_cache_socket=None, weight_cache_timeout=1800, forward_hooks=None, msprobe_dump_config=None)
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:16 TP30 EP30] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:16 TP30 EP30] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:19 TP30 EP30] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:19.321749   241 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:19.321810   241 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:19.321820   241 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.8 port: 12001
+I0827 04:36:19.321856   241 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.8:16569
+I0827 04:36:19.321940   241 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:19.343153   241 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:19.347820   241 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:19.373265   241 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:19.380316   241 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:19.383023   241 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:19.383050   241 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:19.383101   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:19.390141   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:19.391583   241 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:08
+I0827 04:36:19.391615   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:19.417495   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:19.418977   241 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:08
+I0827 04:36:19.419003   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:19.445492   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:19.446861   241 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:08
+I0827 04:36:19.446884   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:19.473483   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:19.474817   241 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:08
+I0827 04:36:19.474922   241 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:19 TP30 EP30] Init torch distributed begin.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:20 TP27 EP27] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:20 TP27 EP27] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:21 TP25 EP25] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:21 TP25 EP25] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:21 TP26 EP26] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:21 TP26 EP26] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:21 TP28 EP28] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:21 TP28 EP28] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:21 TP24 EP24] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:21 TP24 EP24] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:21 TP29 EP29] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:21 TP29 EP29] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:21 TP31 EP31] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:21 TP31 EP31] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:22 TP27 EP27] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:22.734625   238 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:22.734687   238 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:22.734699   238 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.8 port: 12001
+I0827 04:36:22.734735   238 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.8:16566
+I0827 04:36:22.734848   238 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:22.760079   238 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:22.768187   238 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:22.777189   238 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:22.786154   238 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:22.789806   238 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:22.789830   238 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:22.789880   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:22.797986   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:22.799417   238 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:08
+I0827 04:36:22.799446   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:22.825555   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:22.826912   238 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:08
+I0827 04:36:22.826946   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:22.853536   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:22.854885   238 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:08
+I0827 04:36:22.854909   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:22.881553   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:22.882910   238 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:08
+I0827 04:36:22.883030   238 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:22 TP27 EP27] Init torch distributed begin.
+[2026-08-27 04:36:22 TP25 EP25] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:22 TP28 EP28] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:22 TP26 EP26] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:23 TP29 EP29] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:23 TP24 EP24] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:23.476856   236 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:23.476951   236 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:23.476967   236 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.8 port: 12001
+I0827 04:36:23.477013   236 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.8:15425
+I0827 04:36:23.477150   236 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:23.477173   239 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:23.477272   239 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:23.477288   239 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.8 port: 12001
+I0827 04:36:23.477344   239 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.8:15710
+I0827 04:36:23.477483   239 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:23.483197   236 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:23.483803   239 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:23.488940   236 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:23.489802   239 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:23.494866   236 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:23.495663   239 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:23.500973   236 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:23.501806   239 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:23.503927   236 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:23.503952   236 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:23.503998   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.504750   239 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:23.504770   239 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:23.504812   239 rdma_context.cpp:205] Using SIEVE endpoint store
+[2026-08-27 04:36:23 TP31 EP31] Multimodal data loading enabled with 16 worker threads (auto).
+I0827 04:36:23.510515   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:23.511322   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:23.511830   236 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:08
+I0827 04:36:23.511855   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.512616   239 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:08
+I0827 04:36:23.512640   239 rdma_context.cpp:205] Using SIEVE endpoint store
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:23.522899   237 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:23.523005   237 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:23.523020   237 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.8 port: 12001
+I0827 04:36:23.523065   237 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.8:16748
+I0827 04:36:23.523202   237 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:23.529469   237 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:23.535485   237 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:23.537460   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:23.537528   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:23.538903   236 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:08
+I0827 04:36:23.538913   239 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:08
+I0827 04:36:23.538928   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.538945   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.541347   237 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:23.546439   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:23.547672   237 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:23.547796   239 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:08
+I0827 04:36:23.547820   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.550041   237 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:23.550061   237 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:23.550102   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.555119   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:23.556490   239 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:08
+I0827 04:36:23.556598   239 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:23 TP28 EP28] Init torch distributed begin.
+I0827 04:36:23.557152   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:23.558562   237 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:08
+I0827 04:36:23.558589   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.565423   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:23.566766   236 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:08
+I0827 04:36:23.566788   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.589463   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:23.590811   237 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:08
+I0827 04:36:23.590835   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.593444   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:23.594800   236 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:08
+I0827 04:36:23.594898   236 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:23 TP25 EP25] Init torch distributed begin.
+I0827 04:36:23.598161   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:23.599563   237 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:08
+I0827 04:36:23.599589   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.625564   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:23.626922   237 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:08
+I0827 04:36:23.627038   237 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:23 TP26 EP26] Init torch distributed begin.
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:23.899124   240 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:23.899245   240 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:23.899261   240 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.8 port: 12001
+I0827 04:36:23.899313   240 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.8:15127
+I0827 04:36:23.899492   240 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:23.907753   235 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:23.907863   235 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:23.907878   235 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.8 port: 12001
+I0827 04:36:23.907927   235 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.8:15330
+I0827 04:36:23.908092   235 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:23.914011   235 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:23.919618   235 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:23.925108   235 topology.cpp:127] Device mlx5_2 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:23.925450   242 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:23.925558   242 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:23.925576   242 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.8 port: 12001
+I0827 04:36:23.925631   242 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.8:16235
+I0827 04:36:23.925693   240 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:23.925761   242 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:23.930419   242 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:23.933247   240 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:23.934639   242 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:23.939190   242 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:23.940605   240 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:23.943866   242 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:23.945531   242 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:23.945554   242 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:23.945595   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.947860   240 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:23.949263   235 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:23.951108   240 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:23.951148   240 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:23.951216   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.952417   235 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:23.952445   235 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:23.952502   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.958555   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:23.959836   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:23.960163   240 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:08
+I0827 04:36:23.960197   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.961452   235 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:08
+I0827 04:36:23.961483   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.969228   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:23.969322   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:23.970860   235 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:08
+I0827 04:36:23.970890   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.971050   242 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:08
+I0827 04:36:23.971084   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.979120   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:23.980706   242 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:08
+I0827 04:36:23.980736   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.989722   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:23.991313   240 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:08
+I0827 04:36:23.991346   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:24.001633   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:24.003211   235 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:08
+I0827 04:36:24.003237   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:24.005753   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:24.007426   242 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:08
+I0827 04:36:24.007457   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:24.021715   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:24.023293   240 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:08
+I0827 04:36:24.023322   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:24.034039   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:24.034071   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:24.035696   235 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:08
+I0827 04:36:24.035816   235 transfer_engine_impl.cpp:388] installTransport, type=rdma
+I0827 04:36:24.035794   242 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:08
+I0827 04:36:24.035987   242 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:24 TP24 EP24] Init torch distributed begin.
+[2026-08-27 04:36:24 TP31 EP31] Init torch distributed begin.
+I0827 04:36:24.049829   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:24.051388   240 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:08
+I0827 04:36:24.051532   240 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:24 TP29 EP29] Init torch distributed begin.
+[2026-08-27 04:36:48 TP31 EP31] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP30 EP30] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP28 EP28] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP29 EP29] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP27 EP27] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP26 EP26] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP25 EP25] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP24 EP24] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP31 EP31] Init torch distributed ends. elapsed=24.20 s, mem usage=0.40 GB
+[2026-08-27 04:36:48 TP30 EP30] Init torch distributed ends. elapsed=28.76 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP29 EP29] Init torch distributed ends. elapsed=24.18 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP28 EP28] Init torch distributed ends. elapsed=24.68 s, mem usage=0.40 GB
+[2026-08-27 04:36:48 TP27 EP27] Init torch distributed ends. elapsed=25.35 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP26 EP26] Init torch distributed ends. elapsed=24.61 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP25 EP25] Init torch distributed ends. elapsed=24.64 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP24 EP24] Init torch distributed ends. elapsed=24.20 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP28 EP28] Load weight begin. avail mem=82.19 GB
+[2026-08-27 04:36:48 TP27 EP27] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP30 EP30] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP24 EP24] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP29 EP29] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP26 EP26] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP25 EP25] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP28 EP28] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP30 EP30] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP27 EP27] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP29 EP29] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP24 EP24] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP26 EP26] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP25 EP25] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP31 EP31] Load weight begin. avail mem=82.19 GB
+[2026-08-27 04:36:48 TP31 EP31] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP27 EP27] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP28 EP28] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP24 EP24] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP27 EP27] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP30 EP30] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP29 EP29] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP26 EP26] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP25 EP25] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP28 EP28] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP24 EP24] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP29 EP29] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP31 EP31] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP30 EP30] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP26 EP26] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP25 EP25] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP31 EP31] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:50 TP29 EP29] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP28 EP28] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP30 EP30] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP26 EP26] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP25 EP25] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP31 EP31] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP27 EP27] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP24 EP24] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:38:54 TP26 EP26] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:54 TP27 EP27] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:54 TP24 EP24] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:54 TP26 EP26] Load weight end. elapsed=125.61 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:54 TP29 EP29] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:54 TP27 EP27] Load weight end. elapsed=125.61 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:54 TP24 EP24] Load weight end. elapsed=125.61 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:54 TP29 EP29] Load weight end. elapsed=125.61 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:54 TP25 EP25] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:54 TP25 EP25] Load weight end. elapsed=125.64 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:54 TP28 EP28] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:54 TP28 EP28] Load weight end. elapsed=125.69 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:54 TP30 EP30] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:54 TP30 EP30] Load weight end. elapsed=125.71 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:54 TP31 EP31] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:54 TP31 EP31] Load weight end. elapsed=125.77 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:39:01 TP26 EP26] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP26 EP26] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP30 EP30] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP30 EP30] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP28 EP28] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP28 EP28] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP27 EP27] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP27 EP27] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP25 EP25] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP25 EP25] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP31 EP31] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP31 EP31] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP24 EP24] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP24 EP24] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP29 EP29] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP29 EP29] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:02 TP29 EP29] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP28 EP28] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP26 EP26] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP27 EP27] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP30 EP30] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP25 EP25] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP31 EP31] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP24 EP24] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP29 EP29] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP30 EP30] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP28 EP28] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP26 EP26] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP27 EP27] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP25 EP25] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP24 EP24] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP31 EP31] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP29 EP29] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP29 EP29] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP28 EP28] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP28 EP28] Memory pool end. avail mem=9.18 GB
+[2026-08-27 04:39:02 TP27 EP27] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP24 EP24] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP26 EP26] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP27 EP27] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP31 EP31] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP24 EP24] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP26 EP26] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP31 EP31] Memory pool end. avail mem=9.18 GB
+[2026-08-27 04:39:02 TP30 EP30] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP25 EP25] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP30 EP30] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP25 EP25] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP29 EP29] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP29 EP29] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP24 EP24] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP24 EP24] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:39:02 TP28 EP28] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP28 EP28] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.17 GB
+[2026-08-27 04:39:02 TP25 EP25] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP31 EP31] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP25 EP25] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP31 EP31] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.17 GB
+[2026-08-27 04:39:02 TP26 EP26] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP26 EP26] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP27 EP27] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP27 EP27] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:39:02 TP30 EP30] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP30 EP30] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:42:56 TP30 EP30] Capture target decode CUDA graph end. elapsed=233.69 s, mem usage=0.56 GB, avail mem=8.59 GB.
+[2026-08-27 04:42:56 TP30 EP30] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP27 EP27] Capture target decode CUDA graph end. elapsed=233.73 s, mem usage=0.62 GB, avail mem=8.51 GB.
+[2026-08-27 04:42:56 TP27 EP27] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP31 EP31] Capture target decode CUDA graph end. elapsed=233.74 s, mem usage=0.62 GB, avail mem=8.55 GB.
+[2026-08-27 04:42:56 TP31 EP31] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP26 EP26] Capture target decode CUDA graph end. elapsed=233.76 s, mem usage=0.56 GB, avail mem=8.59 GB.
+[2026-08-27 04:42:56 TP26 EP26] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP29 EP29] Capture target decode CUDA graph end. elapsed=233.82 s, mem usage=0.58 GB, avail mem=8.58 GB.
+[2026-08-27 04:42:56 TP29 EP29] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP28 EP28] Capture target decode CUDA graph end. elapsed=233.79 s, mem usage=0.60 GB, avail mem=8.56 GB.
+[2026-08-27 04:42:56 TP28 EP28] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP24 EP24] Capture target decode CUDA graph end. elapsed=233.88 s, mem usage=0.60 GB, avail mem=8.52 GB.
+[2026-08-27 04:42:56 TP24 EP24] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP25 EP25] Capture target decode CUDA graph end. elapsed=233.89 s, mem usage=0.58 GB, avail mem=8.58 GB.
+[2026-08-27 04:42:56 TP25 EP25] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+W0827 04:42:57.132246  2431 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5913c7c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.132292  2432 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58be65c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.132460  2435 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5af7a600, len: 185344: Operation not permitted [1]
+W0827 04:42:57.132488  2433 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58ed5900, len: 23168: Operation not permitted [1]
+W0827 04:42:57.132501  2436 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5afa7a40, len: 185344: Operation not permitted [1]
+W0827 04:42:57.132512  2437 memory_location.cpp:73] Failed to get NUMA node, addr: 0x590583c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.132552  2434 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58c15080, len: 23168: Operation not permitted [1]
+W0827 04:42:57.132607  2439 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59046600, len: 23168: Operation not permitted [1]
+W0827 04:42:57.132640  2438 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5afd4e80, len: 46336: Operation not permitted [1]
+W0827 04:42:57.132716  2440 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58e5fc80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.276726  2580 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a339340, len: 23168: Operation not permitted [1]
+W0827 04:42:57.276731  2579 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59705880, len: 23168: Operation not permitted [1]
+W0827 04:42:57.276762  2581 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a32b080, len: 23168: Operation not permitted [1]
+W0827 04:42:57.276839  2582 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a331fc0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.276887  2583 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5bb81840, len: 185344: Operation not permitted [1]
+W0827 04:42:57.276939  2584 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5bbaec80, len: 185344: Operation not permitted [1]
+W0827 04:42:57.276983  2585 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59b7b6c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.277042  2587 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59953f00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.277055  2586 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59517380, len: 46336: Operation not permitted [1]
+W0827 04:42:57.277086  2588 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59b87640, len: 23168: Operation not permitted [1]
+W0827 04:42:57.448414  2727 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a4bd3c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.448436  2728 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ac9eb00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.448488  2729 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a2fe240, len: 23168: Operation not permitted [1]
+W0827 04:42:57.448798  2735 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a6135c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.448812  2734 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59ed25c0, len: 46336: Operation not permitted [1]
+W0827 04:42:57.448840  2733 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a3c3080, len: 23168: Operation not permitted [1]
+W0827 04:42:57.448849  2736 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a181e80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.451195  2730 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5acaf240, len: 23168: Operation not permitted [1]
+W0827 04:42:57.451217  2731 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5c61b840, len: 185344: Operation not permitted [1]
+W0827 04:42:57.455173  2732 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5c648c80, len: 185344: Operation not permitted [1]
+W0827 04:42:57.460536  2816 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a391f40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.460651  2817 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ae89700, len: 23168: Operation not permitted [1]
+W0827 04:42:57.460843  2819 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ae0fa00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.460901  2826 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a4e2c80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.460979  2830 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ae7a300, len: 23168: Operation not permitted [1]
+W0827 04:42:57.461052  2821 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5c6d6880, len: 185344: Operation not permitted [1]
+W0827 04:42:57.461104  2820 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a22ed40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.462260  2823 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5c703cc0, len: 185344: Operation not permitted [1]
+W0827 04:42:57.463069  2827 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4678c140, len: 46336: Operation not permitted [1]
+W0827 04:42:57.463085  2829 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a645e80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.466511  2955 memory_location.cpp:73] Failed to get NUMA node, addr: 0x597592c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.466544  2958 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59908b40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.466676  2960 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a2f9c80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.466849  2953 memory_location.cpp:73] Failed to get NUMA node, addr: 0x599f9a80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.467304  2964 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5b9e4d40, len: 185344: Operation not permitted [1]
+W0827 04:42:57.470024  2966 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5993fa80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.471112  2970 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59898900, len: 23168: Operation not permitted [1]
+W0827 04:42:57.471313  2962 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5b9b7900, len: 185344: Operation not permitted [1]
+W0827 04:42:57.475107  2968 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5991bac0, len: 46336: Operation not permitted [1]
+W0827 04:42:57.475121  2972 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59b5ea40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.493731  3089 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59564740, len: 23168: Operation not permitted [1]
+W0827 04:42:57.493901  3097 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5b355380, len: 185344: Operation not permitted [1]
+W0827 04:42:57.494069  3104 memory_location.cpp:73] Failed to get NUMA node, addr: 0x45dbac00, len: 46336: Operation not permitted [1]
+W0827 04:42:57.494072  3105 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5902f500, len: 23168: Operation not permitted [1]
+W0827 04:42:57.494150  3094 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59498600, len: 23168: Operation not permitted [1]
+W0827 04:42:57.494194  3099 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5b3827c0, len: 185344: Operation not permitted [1]
+W0827 04:42:57.499953  3107 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59486740, len: 23168: Operation not permitted [1]
+W0827 04:42:57.502707  3102 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59016c80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.504236  3091 memory_location.cpp:73] Failed to get NUMA node, addr: 0x594d2200, len: 23168: Operation not permitted [1]
+W0827 04:42:57.504359  3095 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59354e40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.514688  3200 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5885e940, len: 23168: Operation not permitted [1]
+W0827 04:42:57.514716  3202 memory_location.cpp:73] Failed to get NUMA node, addr: 0x587d55c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.514720  3201 memory_location.cpp:73] Failed to get NUMA node, addr: 0x586bd5c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.514801  3204 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5aa56c00, len: 185344: Operation not permitted [1]
+W0827 04:42:57.514810  3205 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5aa84040, len: 185344: Operation not permitted [1]
+W0827 04:42:57.514727  3203 memory_location.cpp:73] Failed to get NUMA node, addr: 0x44d09c40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.514963  3206 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58692200, len: 23168: Operation not permitted [1]
+W0827 04:42:57.515017  3209 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58865100, len: 23168: Operation not permitted [1]
+W0827 04:42:57.515022  3207 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58ae8bc0, len: 46336: Operation not permitted [1]
+W0827 04:42:57.515030  3208 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58683a40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.548924  3433 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5895be80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.548955  3434 memory_location.cpp:73] Failed to get NUMA node, addr: 0x593aa600, len: 23168: Operation not permitted [1]
+W0827 04:42:57.549010  3435 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59399e80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.549088  3436 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58a4fa40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.549110  3437 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ab36800, len: 185344: Operation not permitted [1]
+W0827 04:42:57.549167  3438 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ab63c40, len: 185344: Operation not permitted [1]
+W0827 04:42:57.551182  3439 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5936aac0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.551239  3441 memory_location.cpp:73] Failed to get NUMA node, addr: 0x455ad780, len: 23168: Operation not permitted [1]
+W0827 04:42:57.551265  3442 memory_location.cpp:73] Failed to get NUMA node, addr: 0x594169c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.551280  3440 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5853c980, len: 46336: Operation not permitted [1]
+[2026-08-27 04:42:59] Dummy health check server started in background thread at 0.0.0.0:30000
+I0827 04:48:13.997741  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:13.998107  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.014123  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.014460  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.045612  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.045893  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.063987  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.064270  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.072747  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.073004  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.082927  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.089962  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.090194  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.112046  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.112280  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.112514  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.115789  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.119459  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.119691  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.119886  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.120070  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.120432  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.120639  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.120857  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.127898  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.128136  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.153139  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.153350  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.170001  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.170248  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.170490  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.170712  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.171002  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.171211  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.171445  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.185374  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.185580  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.185820  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.186059  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.186275  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.218560  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.218878  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.219130  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.219380  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.239995  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.240248  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.240523  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.240849  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.241052  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.241318  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.241586  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.241847  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.242057  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.242271  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.242539  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.256910  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.258225  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.261878  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.262174  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.262421  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.262710  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.263056  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.263356  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.263574  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.265990  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.267298  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.281507  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.281759  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.281965  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.282989  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.289211  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.290711  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.291896  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.292208  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.292472  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.292713  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.292904  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.293118  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.293318  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.301455  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.301728  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.301918  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.302150  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.302408  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.302699  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.302891  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.303146  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.303368  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.303570  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.305574  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.306890  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.314224  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.315574  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.319671  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.319937  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.320156  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.326327  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.326542  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.330421  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.330546  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.330736  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.330998  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.331259  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.331496  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.331717  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.331938  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.332140  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.332340  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.332551  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.333663  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.338346  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.339445  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.339735  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.340694  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.345638  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.360560  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.361940  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.369402  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.369591  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.370901  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.377556  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.384063  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.384287  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.384516  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.385501  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.397218  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.397437  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.397625  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.398703  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.398864  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.399148  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.407058  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.407372  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.407644  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.407857  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.408042  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.408264  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.408543  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.408807  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.411617  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.411850  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.412068  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.414391  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.415768  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.423707  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.425092  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.425122  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.430542  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.430770  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.431028  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.431236  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.431470  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.434294  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.434525  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.435901  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.436133  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.437299  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.446563  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.446866  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.451089  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.453542  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.454408  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.455781  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.467486  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.467700  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.467938  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.468218  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.468482  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.468770  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.469058  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.472146  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.473528  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.480324  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.480634  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.481218  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.483843  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.484179  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.484373  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.485363  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.498565  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.498824  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.499142  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.502869  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.508553  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.511684  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.511938  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.513093  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.523473  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.523744  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.523972  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.524215  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.529819  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.531201  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.539283  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.540769  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.542526  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.542757  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.543026  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.543277  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.543532  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.543817  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.544056  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.544275  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.544507  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.544736  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.544970  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.545243  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.545476  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.545689  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.546013  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.551106  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.551363  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.551533  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.551725  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.551961  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.552152  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.552357  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.552618  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.555388  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.555505  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.558743  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.564610  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.564857  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.569089  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.570464  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.581254  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.581478  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.586069  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.586848  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.587337  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.589694  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.589882  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.590091  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.590329  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.590632  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.590873  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.591192  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.591454  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.591703  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.591930  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.592170  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.592396  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.592653  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.595007  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.596319  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.597950  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.599296  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.609437  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.609653  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.609938  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.610687  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.621923  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.622153  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.622318  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.623299  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.648794  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.649142  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.664134  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.664382  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.669473  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.669734  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.672606  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.672832  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.673055  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.673255  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.673466  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.673666  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.676374  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.676550  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.681711  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.689419  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.692066  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.692279  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.694999  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.695231  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.695472  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.698184  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.709776  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.717428  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.720180  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.720382  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.720589  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.720793  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.731559  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.731812  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.731972  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.742295  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.750133  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.750379  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.750625  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.750792  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.778556  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.778743  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.778970  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.779207  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.789633  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.789860  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.790076  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.795363  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.795607  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.795882  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.796195  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.796408  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.796597  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.796849  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.797055  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.797358  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.797609  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.800431  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.808163  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.808368  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.808629  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.808866  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.811561  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.811780  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.828982  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.830327  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.832255  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.835093  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.835287  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.835476  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.836141  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.840837  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.841059  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.841271  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.841498  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.841699  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.841900  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.842151  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.842352  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.842628  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.842826  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.843127  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.843389  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.856551  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.856762  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.856982  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.857259  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.857627  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.858045  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.863099  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.863319  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.882516  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.882750  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.890803  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.891052  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.891294  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.892642  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.892895  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.893175  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.893406  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.893621  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.894102  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.894142  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.896976  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.897236  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.897420  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.900352  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.902928  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.903223  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.903735  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.904199  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.909134  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.909372  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.913172  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.920277  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.920501  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.920786  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.920931  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.921003  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.921264  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.921265  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.921586  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.943581  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.943809  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.944059  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.944257  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.944458  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.944726  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.944954  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.945147  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.959767  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.960093  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.960309  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.960552  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.960808  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.961053  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.961282  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.961503  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.961782  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.962018  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.962278  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.962548  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.962779  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.963052  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.969125  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.969360  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.969631  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.969812  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.969980  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.970170  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.970402  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.970584  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.970847  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.976580  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.977794  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.981958  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.984225  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.984476  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.998291  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.998525  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.998750  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.999001  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.004917  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.005141  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.021842  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.023211  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.028899  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.028923  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.029189  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.029403  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.029655  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.029892  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.030141  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.030356  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.030586  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.030768  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.051952  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.052182  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.052386  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.052623  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.052874  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.053148  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.053403  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.054157  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.056090  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.056403  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.056617  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.056838  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.060328  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.068521  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.071317  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.071501  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.071699  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.077070  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.080605  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.080899  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.081156  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.081415  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.081660  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.081939  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.082155  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.082402  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.082623  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.082731  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.082814  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.083032  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.085912  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.086201  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.096980  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.098359  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.099546  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.102448  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.102720  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.102989  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.103223  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.103464  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.104277  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.108438  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.108711  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.108911  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.109165  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.109467  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.109733  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.110004  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.110239  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.110486  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.110714  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.110724  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.114183  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.122400  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.122675  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.122920  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.123195  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.123414  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.123673  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.123876  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.126001  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.126210  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.126441  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.126644  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.126863  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.127283  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.131070  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.131325  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.131569  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.131783  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.132396  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.134912  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.135169  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.135370  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.136338  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.140527  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.191358  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.191627  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.191875  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.192153  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.192342  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.197578  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.202698  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.202907  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.208132  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.208374  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.211109  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.211313  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.211566  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.214330  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.214540  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.214802  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.215395  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.216657  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.220089  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.220728  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.225381  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.227144  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.228049  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.228245  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.231040  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.231236  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.234068  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.240044  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.240254  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.240460  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.241421  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.244509  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.250010  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.252700  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.252844  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.252897  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.253087  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.253091  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.253288  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.253509  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.253962  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.263445  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.263675  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.269212  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.270474  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.273865  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.276260  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.279080  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.279291  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.279481  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.279681  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.282605  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.295027  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.295212  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.295395  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.298802  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.299988  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.300223  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.300441  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.300705  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.304690  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.307312  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.307541  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.308730  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.310832  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.313612  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.318708  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.318970  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.319183  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.319437  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.319680  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.319883  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.320147  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.320412  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.325534  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.325740  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.325922  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.326180  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.328954  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.329154  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.329488  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.341765  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.341984  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.347205  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.347421  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.347626  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.347848  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.348078  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.348287  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.348484  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.348680  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.350585  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.350865  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.356657  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.356881  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.357982  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.388767  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.389058  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.389303  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.399434  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.399714  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.409890  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.410177  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.410370  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.410568  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.415786  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.416013  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.416213  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.416402  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.421667  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.421880  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.434330  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.434568  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.444756  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.444969  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.445178  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.445397  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.445652  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.450814  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.451030  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.451257  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.451485  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.471215  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.471483  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.476500  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.476737  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.476964  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.477159  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.477352  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.477542  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.477756  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.477946  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.498270  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.498482  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.518587  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.518834  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.519048  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.519256  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.519551  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.519752  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.520037  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.520202  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.522962  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.523137  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.523340  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.523535  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.523741  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.523979  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.524163  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.524384  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.536993  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.542232  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.542461  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.542663  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.542858  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.543066  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.543337  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.543718  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.551376  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.551574  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.552415  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.552659  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.552764  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.552887  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.553094  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.595072  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.595356  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.595624  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.595870  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.596087  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.596268  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.596544  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.596803  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.597059  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.597254  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.614900  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.615161  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.615470  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.615666  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.615878  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.616151  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.616405  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.653784  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.654006  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.654214  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.654405  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.654635  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.654825  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.655045  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.655264  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.655457  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.655639  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.655900  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.656116  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.656322  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.656612  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.656821  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.704331  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.704610  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.704798  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.705024  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.705224  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.705487  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.705684  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.705888  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.706089  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.706321  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.706543  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.706796  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.707023  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.707245  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.707461  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.707670  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.707892  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.708086  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.708282  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.728194  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.728391  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.728615  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.728813  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.729079  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.729300  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.729538  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.729822  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.737394  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.737599  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.738767  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.451416  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.452901  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.467027  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.467362  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.468367  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.468600  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.468681  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.469802  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.473881  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.484158  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.484486  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.484820  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.485107  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.485355  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.485599  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.487003  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.490974  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.493090  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.493911  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.495500  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.501397  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.502763  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.503086  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.503175  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.503513  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.504180  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.504545  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.508555  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.515570  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.516867  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.518400  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.518653  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.518872  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.519848  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.521056  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.527827  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.528055  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.529295  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.544860  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.545089  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.545290  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.545504  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.546707  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.295120  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.296660  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.302729  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.316711  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.318199  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.323024  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.333889  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.334213  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.334910  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.335601  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.336284  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.340574  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.340840  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.342151  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.347932  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.348773  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.349017  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.350518  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.357188  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.358728  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.360765  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.362203  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.363308  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.366835  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.371485  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.371713  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.373103  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.377363  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.377615  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.377725  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.378958  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.379127  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.382095  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.382319  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.383715  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.383874  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.384758  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.388401  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.392554  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.394655  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.396057  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.398231  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.398520  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.399797  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.400022  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.402158  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.402491  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.402809  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.403522  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.404407  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.409164  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.413456  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.414511  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.416069  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.420145  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.421628  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.421854  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.422119  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.423108  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.427431  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.440521  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.440709  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.440948  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.441156  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.441833  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.447186  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.448485  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.392441  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.393929  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.404150  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.404464  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.405719  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.603718  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.605327  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.611670  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.613114  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.138854  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.139194  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.140391  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.148223  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.149956  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.155593  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.157107  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.663831  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.665465  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.672602  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.675256  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.675565  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.676985  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.683609  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.684000  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.685106  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.687062  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.688648  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.689873  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.693089  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.697635  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.697898  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.699429  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.700985  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.703440  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.703691  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.704947  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.707213  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.710541  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.712060  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.718225  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.719748  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.721971  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.723291  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.725860  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.727556  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.461587  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.463192  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.469041  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.485112  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.485399  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.486702  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.491591  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.499243  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.500805  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.507115  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.510190  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.510407  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.511715  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.517388  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.183045  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.184537  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.190621  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:47.592581  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:47.604241  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:48.578392  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:48.587992  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:48.597157  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:48.605240  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:49.361904  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:49.370812  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:49.376947  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:49.384886  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.134733  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.145203  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.152640  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.159878  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.185983  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.195700  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.204260  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.212427  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.024849  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.026432  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.036523  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.047621  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.048969  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.057524  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.825331  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.834313  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.843958  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.852244  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:54.591287  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:54.605118  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:55.477183  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:57.617564  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:57.677883  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:58.316752  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:58.327472  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:59.071226  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:06.744064  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:07.467173  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:09.220778  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:09.244503  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:10.161363  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:10.162961  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:10.231134  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:15.442806  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:18.016840  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:18.924312  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:28.097031  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:33.444605  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:35.220340  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:51.309719  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:51.309983  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:51.311292  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:51.317121  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:52.153962  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:59.389873  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:59.512159  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:23:07.393066  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:23:08.411403  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:23:09.137790  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:23:09.151702  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:23:10.920717  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:44:34.277145  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-c8-20260827-1340/logs/manual_mc_master.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-c8-20260827-1340/logs/manual_mc_master.log
new file mode 100644
index 0000000..f3322fe
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-c8-20260827-1340/logs/manual_mc_master.log
@@ -0,0 +1,476 @@
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+I0827 04:28:16.835302    99 master.cpp:1366] Master service started on port 50051, max_threads=16, enable_metric_reporting=1, metrics_port=9003, default_kv_lease_ttl=10000, default_kv_soft_pin_ttl=1800000, allow_evict_soft_pinned_objects=1, eviction_ratio=0.05, eviction_high_watermark_ratio=0.9, enable_ha=0, enable_offload=0, enable_kv_events=0, kv_events_bind_endpoint=, kv_events_backend_id=, offload_on_evict=0, offload_force_evict=0, offloading_queue_limit=50000, offload_cap_ratio=0.5, ha_backend_type=etcd, ha_backend_connstring=, etcd_endpoints=, client_ttl=10, rpc_thread_num=16, rpc_port=50051, rpc_address=0.0.0.0, rpc_interface=, rpc_conn_timeout_seconds=0, rpc_enable_tcp_no_delay=1, rpc protocol=tcp, cluster_id=mooncake_cluster, root_fs_dir=, global_file_segment_size=9223372036854775807, memory_allocator=offset, enable_http_metadata_server=0, http_metadata_server_port=8080, http_metadata_server_host=0.0.0.0, enable_metadata_cleanup_on_timeout=0, put_start_discard_timeout_sec=30, put_start_release_timeout_sec=600, max_total_finished_tasks=10000, max_total_pending_tasks=10000, max_total_processing_tasks=10000, pending_task_timeout_sec=300, processing_task_timeout_sec=300, enable_snapshot=0, enable_snapshot_restore=0, snapshot_interval_seconds=600, snapshot_backup_dir=, snapshot_object_store_type=, snapshot_catalog_store_type=, snapshot_retention_count=2, max_retry_attempts=10, enable_cxl=0, cxl_path=/dev/dax0.0, cxl_size=8589934592
+I0827 04:28:16.866606   108 master_service.cpp:1520] Task cleanup thread started
+I0827 04:28:16.867491    99 master_admin_service.cpp:311] Master admin server started on port 9003
+I0827 04:28:16.869802   111 master_admin_service.cpp:302] Master Admin Metrics: role=standby, state=starting, service_ready=false, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:28:26.870203   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:28:36.870568   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:28:46.870873   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:28:56.871232   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:29:06.871615   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:29:16.871969   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:29:26.872316   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:29:36.872687   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:29:46.873068   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:29:56.873467   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:30:06.873914   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:30:16.874214   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:30:26.874567   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:30:36.874953   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:30:46.875332   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:30:56.875741   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:31:06.876119   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:31:16.876497   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:31:26.876866   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:31:36.877233   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:31:46.877605   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:31:56.877948   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:32:06.878437   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:32:16.879210   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:32:26.879711   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:32:36.880080   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:32:46.880565   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:32:56.881004   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:33:06.881398   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:33:16.881786   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:33:26.882170   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:33:36.882550   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:33:46.882939   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:33:56.883315   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:34:06.883746   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:34:16.884127   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:34:26.884568   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:34:36.884953   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:34:46.885334   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:34:56.885706   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:35:06.886085   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:35:16.886464   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:35:26.886835   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:35:36.887218   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:35:46.887596   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:35:56.887964   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:36:06.888356   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:36:16.888742   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:36:26.889113   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:36:36.889495   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:36:46.889882   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:36:56.890261   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:37:06.890656   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:37:16.891040   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:37:26.891430   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:37:36.891836   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:37:46.892215   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:37:56.892597   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:38:06.892982   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:38:16.893360   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:38:26.893747   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:38:36.894131   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:38:46.894515   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:38:56.894920   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:39:06.895303   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:39:16.895701   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:39:26.896085   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:39:36.896471   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:39:46.896852   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:39:56.897240   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:40:06.897629   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:40:16.898005   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:40:26.898384   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:40:36.898761   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:40:46.899144   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:40:56.899530   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:41:06.899937   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:41:16.900316   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:41:26.900695   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:41:36.901089   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:41:46.901479   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:41:56.901856   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:42:06.902242   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:42:16.902666   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:42:26.903064   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:42:36.903537   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:42:46.903987   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:42:56.904397   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:43:06.904863   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:43:16.905256   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:43:26.905628   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:43:36.905993   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:43:46.906375   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:43:56.906759   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:44:06.907138   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:44:16.907518   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:44:26.907922   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:44:36.908319   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:44:46.908928   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:44:56.909319   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:45:06.909730   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:45:16.910125   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:45:26.910491   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:45:36.910866   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:45:46.911247   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:45:56.911623   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:46:06.911993   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:46:16.912426   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:46:26.912810   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:46:36.913184   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:46:46.913556   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:46:56.913941   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:47:06.914324   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:47:16.914712   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:47:26.915107   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:47:36.915493   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:47:46.915882   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:47:56.916265   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:48:06.916646   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:48:16.917042   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:48:26.917429   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:48:36.917809   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:48:46.918201   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:48:56.918605   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:49:06.918992   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:49:16.919385   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:49:26.919793   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:49:36.920181   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:49:46.920554   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:49:56.920938   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:50:06.921316   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:50:16.921710   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:50:26.922091   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:50:36.922484   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:50:46.922891   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:50:56.923311   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:51:06.923827   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:51:16.924309   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:51:26.924793   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:51:36.925184   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:51:46.925541   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:51:56.925873   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:52:06.926222   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:52:16.926549   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:52:26.926885   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:52:36.927227   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:52:46.927629   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:52:56.927992   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:53:06.928339   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:53:16.928666   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:53:26.928995   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:53:36.929322   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:53:46.929646   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:53:56.930056   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:54:06.930403   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:54:16.930779   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:54:26.931145   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:54:36.931521   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:54:46.931916   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:54:56.932343   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:55:06.932718   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:55:16.933064   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:55:26.933395   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:55:36.933718   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:55:46.934052   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:55:56.934386   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:56:06.934712   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:56:16.935036   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:56:26.935360   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:56:36.935707   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:56:46.936054   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:56:56.936385   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:57:06.936734   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:57:16.937136   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:57:26.937498   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:57:36.937844   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:57:46.938345   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:57:56.938614   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:58:06.938869   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:58:16.939126   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:58:26.939358   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:58:36.939590   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:58:46.939940   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:58:56.940317   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:59:06.940711   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:59:16.941143   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:59:26.941520   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:59:36.941883   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:59:46.942258   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:59:56.942601   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:00:06.942938   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:00:16.943261   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:00:26.943506   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:00:36.943737   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:00:46.943951   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:00:56.944247   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:01:06.944607   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:01:16.944896   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:01:26.945246   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:01:36.945606   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:01:46.945940   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:01:56.946230   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:02:06.946504   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:02:16.946818   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:02:26.947096   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:02:36.947449   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:02:46.947767   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:02:56.948025   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:03:06.948321   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:03:16.948577   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:03:26.948939   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:03:36.949285   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:03:46.949551   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:03:56.949826   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:04:06.950160   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:04:16.950445   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:04:26.950799   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:04:36.951154   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:04:46.951506   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:04:56.951869   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:05:06.952240   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:05:16.952581   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:05:26.952893   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:05:36.953207   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:05:46.953542   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:05:56.953862   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:06:06.954190   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:06:16.954527   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:06:26.954862   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:06:36.955159   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:06:46.955430   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:06:56.955727   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:07:06.956014   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:07:16.956292   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:07:26.956629   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:07:36.956969   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:07:46.957286   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:07:56.957538   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:08:06.957857   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:08:16.958175   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:08:26.958505   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:08:36.958873   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:08:46.959213   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:08:56.959580   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:09:06.959961   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:09:16.960343   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:09:26.960726   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:09:36.961073   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:09:46.961421   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:09:56.961745   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:10:06.962085   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:10:16.962410   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:10:26.962738   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:10:36.963074   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:10:46.963395   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:10:56.963728   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:11:06.964046   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:11:16.964359   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:11:26.964668   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:11:36.964977   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:11:46.965345   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:11:56.965719   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:12:06.966044   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:12:16.966377   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:12:26.966706   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:12:36.967024   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:12:46.967339   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:12:56.967662   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:13:06.967989   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:13:16.968293   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:13:26.968617   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:13:36.968954   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:13:46.969303   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:13:56.969642   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:14:06.969978   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:14:16.970424   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:14:26.970767   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:14:36.971103   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:14:46.971450   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:14:56.971846   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:15:06.972218   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:15:16.972595   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:15:26.972944   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:15:36.973273   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:15:46.973601   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:15:56.973927   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:16:06.974232   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:16:16.974531   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:16:26.974860   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:16:36.975173   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:16:46.975457   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:16:56.975767   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:17:06.976097   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:17:16.976425   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:17:26.976751   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:17:36.977074   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:17:46.977340   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:17:56.977588   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:18:06.977923   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:18:16.978257   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:18:26.978587   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:18:36.978917   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:18:46.979271   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:18:56.979615   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:19:06.979946   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:19:16.980322   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:19:26.980680   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:19:36.981024   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:19:46.981348   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:19:56.981689   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:20:06.982029   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:20:16.982355   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:20:26.982690   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:20:36.983021   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:20:46.983355   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:20:56.983701   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:21:06.984002   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:21:16.984310   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:21:26.984650   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:21:36.985025   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:21:46.985387   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:21:56.985735   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:22:06.986088   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:22:16.986429   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:22:26.986770   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:22:36.987112   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:22:46.987457   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:22:56.987808   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:23:06.988150   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:23:16.988442   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:23:26.988807   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:23:36.989144   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:23:46.989451   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:23:56.989902   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:24:06.990252   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:24:16.990571   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:24:26.990865   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:24:36.991190   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:24:46.991520   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:24:56.991888   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:25:06.992215   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:25:16.992558   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:25:26.993088   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:25:36.993425   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:25:46.993746   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:25:56.994050   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:26:06.994376   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:26:16.994699   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:26:26.995038   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:26:36.995332   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:26:46.995611   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:26:56.995985   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:27:06.996469   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:27:16.996830   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:27:26.997224   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:27:36.997588   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:27:46.997939   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:27:56.998252   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:28:06.998566   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:28:16.998903   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:28:26.999221   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:28:36.999548   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:28:46.999909   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:28:57.000319   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:29:07.000710   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:29:17.001116   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:29:27.001483   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:29:37.001886   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:29:47.002224   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:29:57.002625   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:30:07.002960   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:30:17.003293   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:30:27.003625   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:30:37.003929   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:30:47.004259   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:30:57.004590   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:31:07.004930   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:31:17.005257   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:31:27.005599   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:31:37.005923   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:31:47.006310   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:31:57.006702   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:32:07.007108   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:32:17.007465   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:32:27.007848   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:32:37.008191   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:32:47.008533   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:32:57.008893   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:33:07.009240   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:33:17.009582   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:33:27.009918   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:33:37.010253   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:33:47.010578   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:33:57.010911   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:34:07.011226   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:34:17.011554   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:34:27.011901   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:34:37.012240   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:34:47.012563   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:34:57.012892   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:35:07.013228   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:35:17.013548   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:35:27.013886   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:35:37.014219   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:35:47.014546   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:35:57.014894   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:36:07.015219   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:36:17.015552   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:36:27.015900   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:36:37.016237   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:36:47.016562   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:36:57.016880   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:37:07.017205   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:37:17.017522   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:37:27.017875   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:37:37.018208   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:37:47.018537   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:37:57.018862   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:38:07.019249   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:38:17.019589   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:38:27.019936   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:38:37.020287   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:38:47.020581   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:38:57.020905   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:39:07.021240   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:39:17.021579   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:39:27.021906   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:39:37.022253   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:39:47.022643   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:39:57.023057   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:40:07.023453   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:40:17.023824   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:40:27.024180   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:40:37.024562   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:40:47.024909   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:40:57.025257   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:41:07.025607   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:41:17.025949   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:41:27.026325   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:41:37.026783   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:41:47.027132   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:41:57.027501   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:42:07.027843   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:42:17.028157   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:42:27.028415   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:42:37.028738   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:42:47.029076   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:42:57.029417   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:43:07.029729   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:43:17.030092   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:43:27.030454   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:43:37.030807   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:43:47.031152   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:43:57.031495   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:44:07.031854   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:44:17.032115   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:44:27.032442   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:44:37.032795   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:44:47.033133   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:44:57.033485   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:45:07.033839   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:45:17.034091   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:45:27.034423   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:45:37.034760   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:45:47.035167   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:45:57.035507   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:46:07.035853   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:46:17.036199   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:46:27.036579   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:46:37.036921   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:46:47.037261   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-c8-20260827-1340/logs/manual_p_node0.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-c8-20260827-1340/logs/manual_p_node0.log
new file mode 100644
index 0000000..55fe45a
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-c8-20260827-1340/logs/manual_p_node0.log
@@ -0,0 +1,1602 @@
+/usr/local/lib/python3.12/dist-packages/torchao/quantization/quant_api.py:1731: SyntaxWarning: invalid escape sequence '\.'
+  """Configuration class for applying different quantization configs to modules or parameters based on their fully qualified names (FQNs).
+/sgl-workspace/sglang/python/sglang/launch_server.py:56: UserWarning: 'python -m sglang.launch_server' is still supported, but 'sglang serve' is the recommended entrypoint.
+  Example: sglang serve --model-path <model> [options]
+  warnings.warn(
+[2026-08-27 04:28:54] Breakable CUDA graph is incompatible with MLA attention; disabling prefill CUDA graph.
+[2026-08-27 04:28:55] Attention backend not specified. Use triton backend by default.
+[2026-08-27 04:28:55] Pipeline parallelism is incompatible with overlap schedule.
+[2026-08-27 04:28:58] server_args=ServerArgs(model_path='/data/hf_models/Kimi-K3', tokenizer_path='/data/hf_models/Kimi-K3', tokenizer_mode='auto', tokenizer_backend='huggingface', tokenizer_worker_num=1, detokenizer_worker_num=1, skip_tokenizer_init=False, load_format='auto', model_loader_extra_config='{}', trust_remote_code=True, context_length=None, is_embedding=False, enable_multimodal=None, revision=None, model_impl='auto', model_config_parser='auto', json_model_override_args='{}', dtype='auto', quantization=None, quantization_param_path=None, kv_cache_dtype='auto', enable_fp32_lm_head=False, modelopt_quant=None, modelopt_checkpoint_restore_path=None, modelopt_checkpoint_save_path=None, modelopt_export_path=None, quantize_and_serve=False, rl_quant_profile=None, enable_tf32_matmul=False, mem_fraction_static=0.88, max_running_requests=None, max_queued_requests=None, max_total_tokens=None, chunked_prefill_size=8192, enable_dynamic_chunking=False, max_prefill_tokens=16384, prefill_max_requests=None, schedule_policy='fcfs', enable_priority_scheduling=False, disable_priority_preemption=False, default_priority_value=None, abort_on_priority_when_disabled=False, schedule_low_priority_values_first=False, priority_scheduling_preemption_threshold=10, retraction_policy='length', schedule_conservativeness=1.0, page_size=64, swa_full_tokens_ratio=0.8, disable_hybrid_swa_memory=False, radix_eviction_policy='lru', prefill_only_disable_kv_cache=False, disable_radix_cache=True, enable_page_major_kv_layout=False, enable_unified_memory=False, disable_chunked_prefix_cache=False, disable_overlap_schedule=True, num_continuous_decode_steps=1, scheduler_recv_interval=1, enable_mixed_chunk=False, nccl_port=None, dist_timeout=3600, dist_init_addr='174.1.60.1:20000', nnodes=4, node_rank=0, tp_size=4, dcp_size=1, pp_size=8, pp_max_micro_batch_size=None, pp_async_batch_depth=0, dp_size=1, load_balance_method='follow_bootstrap_room', attn_cp_size=1, moe_dp_size=1, dcp_comm_backend='ag_rs', dcp_replicate_q_proj=None, dwdp_size=1, enable_prefill_cp=False, cp_strategy=None, enable_dsa_cache_layer_split=False, enable_dsa_prefill_context_parallel=False, dsa_prefill_cp_mode='round-robin-split', enable_prefill_context_parallel=False, prefill_cp_mode='in-seq-split', enable_cp_decode_attn_tp=False, enable_dp_attention=False, enable_dp_attention_local_control_broadcast=False, enable_dp_lm_head=False, enable_attn_tp_input_scattered=False, disable_attn_tp_gather=False, enable_p2p_check=False, device='cuda', base_gpu_id=0, gpu_id_step=1, random_seed=984607766, watchdog_timeout=300, soft_watchdog_timeout=None, sleep_on_idle=False, use_ray=False, custom_sigquit_handler=None, numa_node=None, gc_threshold=None, host='0.0.0.0', port=30000, fastapi_root_path='', smg_grpc_mode=False, grpc_mode=False, grpc_port=None, sidecar=None, sidecar_args=None, skip_server_warmup=False, warmups=None, enable_http2=False, ssl_keyfile=None, ssl_certfile=None, ssl_ca_certs=None, ssl_keyfile_password=None, enable_ssl_refresh=False, api_key=None, admin_api_key=None, served_model_name='kimi-k3', weight_version='default', chat_template=None, hf_chat_template_name=None, completion_template=None, file_storage_path='sglang_storage', enable_cache_report=False, reasoning_parser=None, default_chat_template_kwargs=None, strip_thinking_cache=False, enable_strict_thinking=False, tool_call_parser=None, tool_server=None, sampling_defaults='model', asr_max_buffer_seconds=60, asr_max_concurrent_sessions=32, preferred_sampling_params=None, allow_auto_truncate=False, stream_interval=1, batch_notify_size=16, stream_response_default_include_usage=False, incremental_streaming_output=False, enable_streaming_session=False, enable_session_radix_cache=False, log_level='info', log_level_http=None, log_requests=False, log_requests_level=2, log_requests_format='text', log_requests_target=None, uvicorn_access_log_exclude_prefixes=[], crash_dump_folder=None, show_time_cost=False, enable_metrics=False, smg_http_sidecar_port=None, enable_mfu_metrics=False, enable_metrics_for_all_schedulers=False, load_snapshot_publish_interval=15, tokenizer_metrics_custom_labels_header='x-custom-labels', tokenizer_metrics_allowed_custom_labels=None, extra_metric_labels=None, bucket_time_to_first_token=None, bucket_inter_token_latency=None, bucket_e2e_request_latency=None, prompt_tokens_buckets=None, generation_tokens_buckets=None, gc_warning_threshold_secs=0.0, decode_log_interval=40, enable_request_time_stats_logging=False, kv_events_config=None, enable_forward_pass_metrics=False, forward_pass_metrics_worker_id='', forward_pass_metrics_ipc_name=None, enable_trace=False, trace_modules='request', otlp_traces_endpoint='localhost:4317', export_metrics_to_file=False, export_metrics_to_file_dir=None, stat_loggers=None, constrained_json_whitespace_pattern=None, constrained_json_disable_any_whitespace=False, attention_backend='triton', decode_attention_backend=None, prefill_attention_backend=None, sampling_backend='flashinfer', grammar_backend='xgrammar', radix_cache_backend=None, mm_attention_backend=None, fp8_gemm_runner_backend='auto', fp4_gemm_runner_backend='auto', bf16_gemm_backend='auto', dsa_prefill_backend=None, dsa_decode_backend=None, dsa_paged_mqa_logits_backend='auto', dsa_topk_backend='sgl-kernel', disable_flashinfer_autotune=False, flashinfer_autotune_skip_ops=None, mamba_backend='triton', cuda_graph_config=CudaGraphConfig(decode=PhaseConfig(backend='disabled', max_bs=16, bs=[1, 2, 4, 8, 12, 16], tc_compiler='eager', full_prefill_max_req=None), prefill=PhaseConfig(backend='disabled', max_bs=2048, bs=[4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], tc_compiler='eager', full_prefill_max_req=None)), cuda_graph_backend_decode=None, cuda_graph_backend_prefill=None, cuda_graph_max_bs_decode=16, cuda_graph_max_bs_prefill=None, cuda_graph_bs_decode=None, cuda_graph_bs_prefill=None, cuda_graph_tc_compiler=None, disable_prefill_cuda_graph=False, disable_decode_cuda_graph=False, disable_cuda_graph=False, disable_cuda_graph_padding=False, enable_profile_cuda_graph=False, enable_cudagraph_gc=False, debug_cuda_graph=False, enable_layerwise_nvtx_marker=False, enable_nccl_nvls=False, enable_symm_mem=False, triton_attention_reduce_in_fp32=False, triton_attention_num_kv_splits=8, triton_attention_split_tile_size=None, flashinfer_mla_disable_ragged=False, enable_fused_qk_norm_rope=False, enable_precise_embedding_interpolation=False, enable_fused_moe_sum_all_reduce=False, enable_deepseek_v4_fp4_indexer=False, disable_custom_all_reduce=False, enable_mscclpp=False, enable_torch_symm_mem=False, enable_scattered_sconv=False, pre_warm_nccl=False, enable_quant_communications=False, enable_flashinfer_allreduce_fusion=False, enforce_disable_flashinfer_allreduce_fusion=False, flashinfer_allreduce_fusion_backend=None, enable_aiter_allreduce_fusion=False, enable_torch_compile=False, enable_torch_compile_debug_mode=False, torch_compile_max_bs=32, torchao_config='', speculative_algorithm=None, speculative_draft_model_path=None, speculative_draft_model_revision=None, speculative_draft_load_format=None, speculative_num_steps=None, speculative_eagle_topk=None, speculative_num_draft_tokens=None, speculative_dflash_block_size=None, speculative_dspark_block_size=None, speculative_dspark_sps_table_path=None, speculative_dspark_confidence_sts_path=None, speculative_dspark_align_verify_tokens_to_graph_tier=False, speculative_accept_threshold_single=1.0, speculative_accept_threshold_acc=1.0, speculative_use_rejection_sampling=False, speculative_token_map=None, speculative_attention_mode='prefill', speculative_draft_attention_backend=None, speculative_draft_window_size=None, speculative_moe_runner_backend='flashinfer_mxfp4', speculative_moe_a2a_backend=None, speculative_draft_model_quantization=None, speculative_skip_dp_mlp_sync=False, enable_multi_layer_eagle=False, speculative_adaptive=False, speculative_adaptive_config=None, decoupled_spec_bind_endpoint=None, decoupled_spec_connect_endpoints=None, decoupled_spec_rank=None, decoupled_spec_role='null', spec_trace_dir=None, speculative_ngram_min_bfs_breadth=1, speculative_ngram_max_bfs_breadth=10, speculative_ngram_match_type='BFS', speculative_ngram_max_trie_depth=18, speculative_ngram_capacity=10000000, speculative_ngram_external_corpus_path=None, speculative_ngram_external_sam_budget=0, speculative_ngram_external_corpus_max_tokens=10000000, ep_size=4, moe_a2a_backend='none', moe_runner_backend='flashinfer_mxfp4', flashinfer_mxfp4_moe_precision='default', deepep_mode='auto', fuseep_mode=2, deepep_dispatcher_output_dtype='auto', ep_num_redundant_experts=0, ep_dispatch_algorithm=None, init_expert_location='trivial', enable_eplb=False, eplb_algorithm='auto', eplb_rebalance_num_iterations=1000, eplb_rebalance_layers_per_chunk=None, eplb_min_rebalancing_utilization_threshold=1.0, expert_distribution_recorder_mode=None, expert_distribution_recorder_buffer_size=1000, enable_expert_distribution_metrics=False, deepep_config=None, moe_dense_tp_size=None, elastic_ep_backend=None, enable_elastic_expert_backup=False, mooncake_ib_device=None, enable_waterfill=False, ep_join_mode=None, ep_join_rank_offset=0, elastic_ep_initial_size=None, max_ep_size=None, elastic_ep_scale_timeout=600, elastic_ep_rejoin=False, disable_flashinfer_cutlass_moe_fp4_allgather=False, disable_shared_experts_fusion=False, enforce_shared_experts_fusion=False, max_mamba_cache_size=None, mamba_ssm_dtype=None, mamba_max_states_per_path=-1, enable_mamba_cache_stochastic_rounding=False, mamba_cache_philox_rounds=0, mamba_full_memory_ratio=0.36, mamba_radix_cache_strategy='extra_buffer_lazy', uses_mamba_radix_cache=False, mamba_track_interval=256, enable_int8_mamba_checkpoint=False, int8_mamba_ckpt_size=None, linear_attn_backend='triton', linear_attn_decode_backend=None, linear_attn_prefill_backend=None, linear_attn_verify_backend=None, enable_linear_replayssm=False, linear_replayssm_cache_len=16, enable_linear_replayssm_spec=False, enable_hierarchical_cache=False, hicache_ratio=2.0, hicache_size=0, hicache_write_policy='write_through', hicache_io_backend='kernel', hicache_mem_layout='page_first', hicache_storage_backend=None, hicache_storage_prefetch_policy='timeout', hicache_storage_backend_extra_config=None, enable_hisparse=False, hisparse_config=None, enable_broadcast_mm_inputs_process=False, enable_prefix_mm_cache=False, mm_enable_dp_encoder=False, mm_process_config={}, mm_processor_worker_num=0, mm_io_worker_num=0, limit_mm_data_per_request=None, enable_mm_global_cache=False, disable_fast_image_processor=False, mm_feature_transport='cpu', keep_mm_feature_on_device=False, enable_lora=None, enable_lora_overlap_loading=None, max_lora_rank=None, lora_target_modules=None, lora_paths=None, max_loaded_loras=None, max_loras_per_batch=8, lora_eviction_policy='lru', lora_backend='csgmv', max_lora_chunk_size=16, experts_shared_outer_loras=None, lora_use_virtual_experts=False, lora_strict_loading=False, lora_drain_wait_threshold=0.0, enable_two_batch_overlap=False, enable_single_batch_overlap=False, tbo_token_distribution_threshold=0.48, cpu_offload_gb=0, offload_group_size=-1, offload_num_in_group=1, offload_prefetch_step=1, offload_mode='cpu', enable_lmcache=False, lmcache_config_file=None, enable_flexkv=False, flexkv_config_file=None, kt_weight_path=None, kt_method='AMXINT4', kt_cpuinfer=None, kt_threadpool_count=2, kt_num_gpu_experts=None, kt_max_deferred_experts_per_token=None, dllm_algorithm=None, dllm_algorithm_config=None, dllm_fdfo=True, disaggregation_mode='prefill', disaggregation_transfer_backend='mooncake', disaggregation_bootstrap_port=28800, disaggregation_ib_device='mlx5_0,mlx5_1,mlx5_2,mlx5_3', disaggregation_decode_enable_radix_cache=False, disaggregation_decode_enable_offload_kvcache=False, num_reserved_decode_tokens=512, disaggregation_decode_extra_slots=None, disaggregation_decode_polling_interval=1, optimistic_prefill_attempts=0, encoder_only=False, language_only=False, encoder_transfer_backend='zmq_to_tokenizer', encoder_urls=[], encoder_bootstrap_port=8997, encoder_register_urls=[], enable_adaptive_dispatch_to_encoder=False, enable_pdmux=False, pdmux_config_path=None, sm_group_num=8, custom_weight_loader=[], weight_loader_disable_mmap=False, weight_loader_prefetch_checkpoints=False, weight_loader_prefetch_num_threads=4, weight_loader_drop_cache_after_load=False, remote_instance_weight_loader_seed_instance_ip=None, remote_instance_weight_loader_seed_instance_service_port=None, remote_instance_weight_loader_send_weights_group_ports=None, remote_instance_weight_loader_backend='nccl', remote_instance_weight_loader_start_seed_via_transfer_engine=False, engine_info_bootstrap_port=6789, modelexpress_config=None, download_dir=None, model_checksum=None, delete_ckpt_after_loading=False, decrypted_config_file=None, decrypted_draft_config_file=None, checkpoint_engine_wait_weights_before_ready=False, enable_prefill_delayer=False, prefill_delayer_max_delay_passes=30, prefill_delayer_token_usage_low_watermark=None, prefill_delayer_forward_passes_buckets=None, prefill_delayer_wait_seconds_buckets=None, prefill_delayer_queue_min_ratio=None, prefill_delayer_max_delay_ms=None, min_free_slots_delay=None, enable_deterministic_inference=False, rl_on_policy_target=None, kv_canary='none', kv_canary_real_data='none', kv_canary_sweep_interval=0, enable_dynamic_batch_tokenizer=False, dynamic_batch_tokenizer_batch_size=32, dynamic_batch_tokenizer_batch_timeout=0.002, enable_tokenizer_batch_encode=False, disable_tokenizer_batch_decode=False, debug_tensor_dump_output_folder=None, debug_tensor_dump_layers=None, debug_tensor_dump_input_file=None, enable_memory_saver=False, enable_weights_cpu_backup=False, enable_draft_weights_cpu_backup=False, enable_custom_logit_processor=False, enable_return_hidden_states=False, enable_return_routed_experts=False, enable_return_indexer_topk=False, disable_outlines_disk_cache=False, enable_mis=False, weight_cache_mode='off', weight_cache_socket=None, weight_cache_timeout=1800, forward_hooks=None, msprobe_dump_config=None)
+[2026-08-27 04:28:58] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:58] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:58] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:58] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:58] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:58] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:58] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:58] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:01] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:01] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:01] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:03] Multimodal processor concurrency enabled with 2 isolated worker threads (auto).
+[2026-08-27 04:29:03] CommonKVBootstrapServer started successfully on 0.0.0.0:28800
+[2026-08-27 04:29:03] No HuggingFace chat template found
+[2026-08-27 04:29:03] No chat template found, defaulting to 'string' content format
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:15 PP0 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:15 PP0 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:16] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:16] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:16 PP1 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:16 PP1 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+/sgl-workspace/sglang/python/sglang/srt/utils/hf_transformers/tokenizer.py:423: UserWarning: Using a slow tokenizer. This might cause a significant slowdown. Consider using a fast tokenizer instead.
+  warnings.warn(
+[2026-08-27 04:29:16] Applying special tokens cache patch for Kimi tokenizer: <class 'transformers_modules.Kimi_hyphen_K3.caf40b2773f9029e.tokenization_kimi.TikTokenTokenizer'>
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:16 PP1 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:16 PP1 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:16 PP0 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:16 PP0 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:16 PP1 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:16 PP1 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:17 PP0 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:17 PP0 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:17 PP1 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:17 PP1 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:17 PP0 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:17 PP0 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:17 PP0 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:17.587481   237 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:17.587555   237 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:17.587566   237 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.1 port: 12001
+I0827 04:29:17.587602   237 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.1:16830
+I0827 04:29:17.587723   237 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:17.613691   237 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:17.620473   237 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:17.627439   237 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:17.653980   237 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:17.656663   237 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:17.656692   237 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:17.656746   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:17.663833   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:17.665359   237 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:01
+I0827 04:29:17.665387   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:17.690173   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:17.691654   237 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:01
+I0827 04:29:17.691689   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:17.718215   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:17.719650   237 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:01
+I0827 04:29:17.719686   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:17.746297   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:17.747779   237 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:01
+I0827 04:29:17.747891   237 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:17 PP0 TP1 EP1] Init torch distributed begin.
+[2026-08-27 04:29:17 PP1 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:18.219898   241 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:18.219964   241 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:18.219973   241 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.1 port: 12001
+I0827 04:29:18.220007   241 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.1:15425
+I0827 04:29:18.220109   241 transfer_engine_impl.cpp:258] Auto-discovering topology...
+[2026-08-27 04:29:18 PP1 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+I0827 04:29:18.244532   241 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:18.252967   241 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:18.261787   241 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:18.289944   241 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:18.293610   241 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:18.293633   241 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:18.293681   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.301677   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:18.303126   241 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:01
+I0827 04:29:18.303155   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.330181   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:18.331590   241 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:01
+I0827 04:29:18.331614   241 rdma_context.cpp:205] Using SIEVE endpoint store
+[2026-08-27 04:29:18 PP0 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+I0827 04:29:18.358213   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:18.359581   241 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:01
+I0827 04:29:18.359604   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.386204   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:18.387562   241 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:01
+I0827 04:29:18.387681   241 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:18 PP1 TP1 EP1] Init torch distributed begin.
+[2026-08-27 04:29:18 PP1 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:18.588883   242 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:18.588970   242 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:18.588984   242 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.1 port: 12001
+I0827 04:29:18.589022   242 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.1:15650
+I0827 04:29:18.589143   242 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:18.594419   242 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:18.599439   242 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:18.604553   242 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:18.609825   242 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:18.612036   242 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:18.612056   242 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:18.612097   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.618021   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:18.619308   242 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:01
+I0827 04:29:18.619333   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.646080   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:18.647465   242 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:01
+I0827 04:29:18.647495   242 rdma_context.cpp:205] Using SIEVE endpoint store
+[2026-08-27 04:29:18 PP0 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+I0827 04:29:18.674199   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:18.675598   242 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:01
+I0827 04:29:18.675621   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.682866   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:18.684276   242 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:01
+I0827 04:29:18.684391   242 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:18 PP1 TP2 EP2] Init torch distributed begin.
+[2026-08-27 04:29:18 PP1 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:18.695660   238 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:18.695755   238 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:18.695767   238 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.1 port: 12001
+I0827 04:29:18.695806   238 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.1:15123
+I0827 04:29:18.695909   238 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:18.701093   238 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:18.706161   238 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:18.711323   238 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:18.737910   238 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:18.740556   238 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:18.740602   238 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:18.740651   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.747717   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:18.749193   238 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:01
+I0827 04:29:18.749220   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.774145   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:18.775534   238 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:01
+I0827 04:29:18.775558   238 rdma_context.cpp:205] Using SIEVE endpoint store
+[2026-08-27 04:29:18 PP0 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+I0827 04:29:18.802196   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:18.803623   238 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:01
+I0827 04:29:18.803649   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.810976   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:18.812341   238 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:01
+I0827 04:29:18.812443   238 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:18 PP0 TP2 EP2] Init torch distributed begin.
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:19.084213   243 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:19.084332   243 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:19.084348   243 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.1 port: 12001
+I0827 04:29:19.084400   243 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.1:15411
+I0827 04:29:19.084602   243 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:19.114387   243 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:19.121773   243 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:19.146405   243 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:19.153759   243 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:19.156515   243 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:19.156548   243 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:19.156603   243 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.163982   243 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:19.165694   243 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:01
+I0827 04:29:19.165726   243 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.173307   243 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:19.174904   243 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:01
+I0827 04:29:19.174932   243 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.206454   243 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:19.208142   243 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:01
+I0827 04:29:19.208168   243 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.215909   243 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:19.217583   243 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:01
+I0827 04:29:19.217706   243 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:19 PP1 TP3 EP3] Init torch distributed begin.
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:19.239588   236 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:19.239699   236 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:19.239715   236 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.1 port: 12001
+I0827 04:29:19.239763   236 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.1:15605
+I0827 04:29:19.239933   236 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:19.240546   240 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:19.240644   240 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:19.240658   240 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.1 port: 12001
+I0827 04:29:19.240702   240 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.1:15487
+I0827 04:29:19.240862   240 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:19.242301   239 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:19.242408   239 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:19.242422   239 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.1 port: 12001
+I0827 04:29:19.242470   239 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.1:15552
+I0827 04:29:19.242636   239 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:19.246201   240 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:19.248451   239 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:19.251204   240 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:19.254226   239 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:19.256254   240 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:19.259852   239 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:19.261372   240 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:19.263411   240 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:19.263432   240 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:19.263473   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.265838   239 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:19.268059   239 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:19.268082   239 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:19.268126   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.269106   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:19.270298   236 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:19.270401   240 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:01
+I0827 04:29:19.270423   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.274480   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:19.275991   239 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:01
+I0827 04:29:19.276019   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.276877   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:19.278193   240 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:01
+I0827 04:29:19.278220   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.297820   236 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:19.302485   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:19.302623   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:19.303958   239 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:01
+I0827 04:29:19.303983   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.304121   240 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:01
+I0827 04:29:19.304147   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.311621   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:19.313341   240 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:01
+I0827 04:29:19.313478   240 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:19 PP1 TP0 EP0] Init torch distributed begin.
+I0827 04:29:19.322126   236 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:19.330281   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:19.331759   239 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:01
+I0827 04:29:19.331784   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.339274   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:19.340766   239 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:01
+I0827 04:29:19.340884   239 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:19 PP0 TP3 EP3] Init torch distributed begin.
+I0827 04:29:19.345958   236 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:19.348632   236 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:19.348659   236 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:19.348712   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.355962   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:19.417273   236 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:01
+I0827 04:29:19.417309   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.424894   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:19.426437   236 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:01
+I0827 04:29:19.426491   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.434231   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:19.435868   236 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:01
+I0827 04:29:19.435896   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.443578   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:19.445148   236 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:01
+I0827 04:29:19.445264   236 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:19 PP0 TP0 EP0] Init torch distributed begin.
+[2026-08-27 04:29:26 PP0 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:26 PP1 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:27 PP1 TP2 EP2] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP1 TP0 EP0] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP1 TP3 EP3] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP1 TP1 EP1] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP0 TP3 EP3] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP0 TP0 EP0] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP0 TP1 EP1] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP0 TP2 EP2] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP0 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:27 PP0 TP3 EP3] sglang is using nccl==2.28.9
+[2026-08-27 04:29:27 PP0 TP1 EP1] sglang is using nccl==2.28.9
+[2026-08-27 04:29:27 PP0 TP2 EP2] sglang is using nccl==2.28.9
+[2026-08-27 04:29:27 PP0 TP0 EP0] Init torch distributed ends. elapsed=8.22 s, mem usage=0.46 GB
+[2026-08-27 04:29:27 PP1 TP3 EP3] Init torch distributed ends. elapsed=8.45 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP1 TP1 EP1] Init torch distributed ends. elapsed=9.28 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP1 TP2 EP2] Init torch distributed ends. elapsed=8.98 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP1 TP0 EP0] Init torch distributed ends. elapsed=8.35 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP0 TP2 EP2] Init torch distributed ends. elapsed=8.85 s, mem usage=0.46 GB
+[2026-08-27 04:29:27 PP0 TP3 EP3] Init torch distributed ends. elapsed=8.32 s, mem usage=0.46 GB
+[2026-08-27 04:29:27 PP0 TP1 EP1] Init torch distributed ends. elapsed=9.92 s, mem usage=0.46 GB
+[2026-08-27 04:29:27 PP1 TP0 EP0] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP1 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP0 TP0 EP0] Load weight begin. avail mem=82.12 GB
+[2026-08-27 04:29:27 PP1 TP3 EP3] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP1 TP2 EP2] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP0 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP1 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP1 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP1 TP1 EP1] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP0 TP2 EP2] Load weight begin. avail mem=82.12 GB
+[2026-08-27 04:29:27 PP1 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP0 TP3 EP3] Load weight begin. avail mem=82.12 GB
+[2026-08-27 04:29:27 PP0 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP0 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP1 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:27 PP1 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:27 PP0 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:27 PP1 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:27 PP1 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:27 PP1 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:27 PP0 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:27 PP1 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:27 PP1 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:27 PP1 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:27 PP0 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:27 PP0 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:27 PP0 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:27 PP0 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:27 PP0 TP1 EP1] Load weight begin. avail mem=82.12 GB
+[2026-08-27 04:29:27 PP0 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP0 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP0 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+
Multi-thread loading shards:   0% Completed | 0/96 [00:00<?, ?it/s]
Multi-thread loading shards:   1% Completed | 1/96 [00:00<01:01,  1.55it/s]
Multi-thread loading shards:   2% Completed | 2/96 [00:09<08:49,  5.63s/it]
Multi-thread loading shards:   3% Completed | 3/96 [00:12<06:35,  4.25s/it]
Multi-thread loading shards:   4% Completed | 4/96 [00:14<05:26,  3.55s/it]
Multi-thread loading shards:   5% Completed | 5/96 [00:17<04:43,  3.11s/it]
Multi-thread loading shards:   6% Completed | 6/96 [00:19<04:17,  2.86s/it]
Multi-thread loading shards:   7% Completed | 7/96 [00:21<04:00,  2.71s/it]
Multi-thread loading shards:   8% Completed | 8/96 [00:24<03:49,  2.61s/it]
Multi-thread loading shards:   9% Completed | 9/96 [00:26<03:44,  2.58s/it]
Multi-thread loading shards:  10% Completed | 10/96 [00:29<03:40,  2.56s/it]
Multi-thread loading shards:  11% Completed | 11/96 [00:31<03:36,  2.54s/it]
Multi-thread loading shards:  12% Completed | 12/96 [00:32<02:43,  1.95s/it]
Multi-thread loading shards:  14% Completed | 13/96 [00:32<02:05,  1.52s/it]
Multi-thread loading shards:  15% Completed | 14/96 [00:33<01:33,  1.14s/it]
Multi-thread loading shards:  16% Completed | 15/96 [00:33<01:21,  1.01s/it]
Multi-thread loading shards:  17% Completed | 16/96 [00:34<01:19,  1.00it/s]
Multi-thread loading shards:  18% Completed | 17/96 [00:35<01:02,  1.26it/s]
Multi-thread loading shards:  19% Completed | 18/96 [00:35<00:51,  1.53it/s]
Multi-thread loading shards:  20% Completed | 19/96 [00:35<00:42,  1.80it/s]
Multi-thread loading shards:  21% Completed | 20/96 [00:36<00:37,  2.01it/s]
Multi-thread loading shards:  22% Completed | 21/96 [00:36<00:33,  2.23it/s]
Multi-thread loading shards:  23% Completed | 22/96 [00:36<00:30,  2.43it/s]
Multi-thread loading shards:  24% Completed | 23/96 [00:37<00:27,  2.68it/s]
Multi-thread loading shards:  25% Completed | 24/96 [00:37<00:25,  2.80it/s]
Multi-thread loading shards:  26% Completed | 25/96 [00:37<00:24,  2.89it/s]
Multi-thread loading shards:  27% Completed | 26/96 [00:38<00:23,  2.99it/s]
Multi-thread loading shards:  28% Completed | 27/96 [00:38<00:22,  3.01it/s]
Multi-thread loading shards:  29% Completed | 28/96 [00:38<00:22,  2.99it/s]
Multi-thread loading shards:  30% Completed | 29/96 [00:39<00:22,  2.98it/s]
Multi-thread loading shards:  31% Completed | 30/96 [00:39<00:22,  2.98it/s]
Multi-thread loading shards:  32% Completed | 31/96 [00:39<00:21,  3.01it/s]
Multi-thread loading shards:  33% Completed | 32/96 [00:40<00:21,  3.00it/s]
Multi-thread loading shards:  34% Completed | 33/96 [00:41<00:35,  1.79it/s]
Multi-thread loading shards:  35% Completed | 34/96 [00:41<00:30,  2.02it/s]
Multi-thread loading shards:  36% Completed | 35/96 [00:41<00:27,  2.23it/s]
Multi-thread loading shards:  38% Completed | 36/96 [00:42<00:24,  2.41it/s]
Multi-thread loading shards:  39% Completed | 37/96 [00:42<00:22,  2.57it/s]
Multi-thread loading shards:  40% Completed | 38/96 [00:42<00:21,  2.64it/s]
Multi-thread loading shards:  41% Completed | 39/96 [00:43<00:20,  2.73it/s]
Multi-thread loading shards:  42% Completed | 40/96 [00:43<00:20,  2.80it/s]
Multi-thread loading shards:  43% Completed | 41/96 [00:43<00:19,  2.79it/s]
Multi-thread loading shards:  44% Completed | 42/96 [00:44<00:19,  2.80it/s]
Multi-thread loading shards:  45% Completed | 43/96 [00:44<00:22,  2.39it/s]
Multi-thread loading shards:  46% Completed | 44/96 [00:45<00:20,  2.51it/s]
Multi-thread loading shards:  47% Completed | 45/96 [00:45<00:19,  2.64it/s]
Multi-thread loading shards:  48% Completed | 46/96 [00:45<00:18,  2.67it/s]
Multi-thread loading shards:  49% Completed | 47/96 [00:46<00:18,  2.61it/s]
Multi-thread loading shards:  50% Completed | 48/96 [00:46<00:17,  2.72it/s]
Multi-thread loading shards:  51% Completed | 49/96 [00:47<00:16,  2.84it/s]
Multi-thread loading shards:  52% Completed | 50/96 [00:47<00:16,  2.81it/s]
Multi-thread loading shards:  53% Completed | 51/96 [00:47<00:15,  2.93it/s]
Multi-thread loading shards:  54% Completed | 52/96 [00:48<00:21,  2.05it/s]
Multi-thread loading shards:  55% Completed | 53/96 [00:49<00:33,  1.30it/s]
Multi-thread loading shards:  56% Completed | 54/96 [00:50<00:27,  1.55it/s]
Multi-thread loading shards:  57% Completed | 55/96 [00:50<00:22,  1.82it/s]
Multi-thread loading shards:  58% Completed | 56/96 [00:50<00:19,  2.06it/s]
Multi-thread loading shards:  59% Completed | 57/96 [00:51<00:17,  2.25it/s]
Multi-thread loading shards:  60% Completed | 58/96 [00:51<00:15,  2.40it/s]
Multi-thread loading shards:  61% Completed | 59/96 [00:51<00:14,  2.56it/s]
Multi-thread loading shards:  62% Completed | 60/96 [00:52<00:13,  2.67it/s]
Multi-thread loading shards:  64% Completed | 61/96 [00:52<00:12,  2.77it/s]
Multi-thread loading shards:  65% Completed | 62/96 [00:52<00:11,  2.84it/s]
Multi-thread loading shards:  66% Completed | 63/96 [00:53<00:11,  2.89it/s]
Multi-thread loading shards:  67% Completed | 64/96 [00:53<00:11,  2.87it/s]
Multi-thread loading shards:  68% Completed | 65/96 [00:54<00:10,  2.89it/s]
Multi-thread loading shards:  69% Completed | 66/96 [00:54<00:10,  2.91it/s]
Multi-thread loading shards:  70% Completed | 67/96 [00:54<00:10,  2.90it/s]
Multi-thread loading shards:  71% Completed | 68/96 [00:55<00:09,  2.90it/s]
Multi-thread loading shards:  72% Completed | 69/96 [00:55<00:09,  2.89it/s]
Multi-thread loading shards:  73% Completed | 70/96 [00:56<00:14,  1.85it/s]
Multi-thread loading shards:  74% Completed | 71/96 [00:56<00:11,  2.10it/s]
Multi-thread loading shards:  75% Completed | 72/96 [00:57<00:10,  2.23it/s]
Multi-thread loading shards:  76% Completed | 73/96 [00:57<00:09,  2.39it/s]
Multi-thread loading shards:  77% Completed | 74/96 [00:57<00:08,  2.48it/s]
Multi-thread loading shards:  78% Completed | 75/96 [00:58<00:09,  2.15it/s]
Multi-thread loading shards:  79% Completed | 76/96 [00:59<00:10,  1.89it/s]
Multi-thread loading shards:  80% Completed | 77/96 [00:59<00:09,  2.05it/s]
Multi-thread loading shards:  81% Completed | 78/96 [00:59<00:08,  2.24it/s]
Multi-thread loading shards:  82% Completed | 79/96 [01:00<00:07,  2.39it/s]
Multi-thread loading shards:  83% Completed | 80/96 [01:00<00:06,  2.51it/s]
Multi-thread loading shards:  84% Completed | 81/96 [01:00<00:05,  2.65it/s]
Multi-thread loading shards:  85% Completed | 82/96 [01:01<00:04,  2.84it/s]
Multi-thread loading shards:  86% Completed | 83/96 [01:02<00:06,  1.99it/s]
Multi-thread loading shards:  88% Completed | 84/96 [01:02<00:06,  1.99it/s]
Multi-thread loading shards:  90% Completed | 86/96 [01:02<00:03,  3.27it/s]
Multi-thread loading shards:  92% Completed | 88/96 [01:02<00:01,  4.72it/s]
Multi-thread loading shards:  94% Completed | 90/96 [01:02<00:00,  6.49it/s]
Multi-thread loading shards:  96% Completed | 92/96 [01:03<00:00,  8.37it/s]
Multi-thread loading shards:  98% Completed | 94/96 [01:03<00:00,  8.91it/s]
Multi-thread loading shards: 100% Completed | 96/96 [01:03<00:00,  9.35it/s]
Multi-thread loading shards: 100% Completed | 96/96 [01:03<00:00,  1.51it/s]
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:33 PP0 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:33 PP0 TP2 EP2] Load weight end. elapsed=65.63 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=39.03 GB, mem usage=43.09 GB.
+[2026-08-27 04:30:33 PP0 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:33 PP0 TP0 EP0] Load weight end. elapsed=65.81 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=39.03 GB, mem usage=43.09 GB.
+[2026-08-27 04:30:35 PP0 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:35 PP0 TP3 EP3] Load weight end. elapsed=67.37 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=39.03 GB, mem usage=43.09 GB.
+[2026-08-27 04:30:35 PP0 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:35 PP0 TP1 EP1] Load weight end. elapsed=67.46 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=39.03 GB, mem usage=43.09 GB.
+[2026-08-27 04:30:35 PP0 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:35 PP0 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:35 PP0 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:35 PP0 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:35 PP0 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:35 PP0 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:35 PP0 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:35 PP0 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:36 PP1 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:36 PP1 TP3 EP3] Load weight end. elapsed=68.32 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=36.12 GB, mem usage=45.92 GB.
+[2026-08-27 04:30:36 PP1 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP1 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP1 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP1 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP1 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP1 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP1 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP1 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:37 PP1 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:37 PP1 TP1 EP1] Load weight end. elapsed=69.55 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=36.12 GB, mem usage=45.92 GB.
+[2026-08-27 04:30:37 PP1 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:37 PP1 TP0 EP0] Load weight end. elapsed=69.66 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=36.12 GB, mem usage=45.92 GB.
+[2026-08-27 04:30:39 PP1 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:39 PP1 TP2 EP2] Load weight end. elapsed=71.33 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=36.12 GB, mem usage=45.92 GB.
+[2026-08-27 04:30:39 PP1 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:39 PP1 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:39 PP1 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:39 PP1 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:39 PP1 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:39 PP1 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:39 PP1 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:39 PP1 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:53 PP0 TP0 EP0] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:53 PP1 TP3 EP3] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:53 PP1 TP2 EP2] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:53 PP1 TP1 EP1] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:53 PP1 TP0 EP0] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:53 PP0 TP2 EP2] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:53 PP0 TP3 EP3] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:53 PP0 TP1 EP1] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:53 PP0 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:53 PP1 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:53 PP1 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:53 PP1 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:53 PP0 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:53 PP1 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:53 PP0 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:53 PP0 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:53 PP0 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 8.16 GB
+[2026-08-27 04:30:53 PP0 TP0 EP0] Memory pool end. avail mem=30.14 GB
+[2026-08-27 04:30:53 PP0 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 8.16 GB
+[2026-08-27 04:30:53 PP0 TP2 EP2] Memory pool end. avail mem=30.14 GB
+[2026-08-27 04:30:53 PP1 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:53 PP1 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:53 PP1 TP0 EP0] Memory pool end. avail mem=23.24 GB
+[2026-08-27 04:30:53 PP1 TP2 EP2] Memory pool end. avail mem=23.24 GB
+[2026-08-27 04:30:53 PP0 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 8.16 GB
+[2026-08-27 04:30:53 PP1 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:53 PP0 TP3 EP3] Memory pool end. avail mem=30.14 GB
+[2026-08-27 04:30:53 PP1 TP1 EP1] Memory pool end. avail mem=23.24 GB
+[2026-08-27 04:30:53 PP1 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:53 PP0 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 8.16 GB
+[2026-08-27 04:30:53 PP1 TP3 EP3] Memory pool end. avail mem=23.24 GB
+[2026-08-27 04:30:53 PP0 TP1 EP1] Memory pool end. avail mem=30.14 GB
+[2026-08-27 04:30:54 PP0 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:30:54 PP0 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:30:54 PP0 TP0 EP0] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp0_pp0_dp0.json
+[2026-08-27 04:30:54 PP1 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:30:54 PP1 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:30:54 PP0 TP2 EP2] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp2_pp0_dp0.json
+[2026-08-27 04:30:54 PP1 TP2 EP2] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp2_pp1_dp0.json
+[2026-08-27 04:30:54 PP1 TP0 EP0] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp0_pp1_dp0.json
+[2026-08-27 04:30:54 PP1 TP1 EP1] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp1_pp1_dp0.json
+[2026-08-27 04:30:54 PP1 TP3 EP3] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp3_pp1_dp0.json
+[2026-08-27 04:30:54 PP0 TP1 EP1] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp1_pp0_dp0.json
+[2026-08-27 04:30:54 PP0 TP3 EP3] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp3_pp0_dp0.json
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.46profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.32profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.51profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  6.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.09profile/s]
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.07profile/s]
+[TensorRT-LLM][INFO] Set logger level to INFO
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.48profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.11profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.29profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.37profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.16profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.82profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.18profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.58profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:01<00:02,  1.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.66profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.03profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.49profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:01<00:01,  2.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.25profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:01,  2.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:02<00:00,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:02<00:00,  3.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.20profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.01profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  9.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  9.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  9.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 12.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 12.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 11.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 11.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.78profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.16profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  3.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.75profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.50profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.92profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.68profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  3.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.97profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.87profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 10.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.51profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.74profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 11.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.30profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.70profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.72profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.63profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.53profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.29profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.58profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  8.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 10.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.71profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.62profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.43profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.18profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.29profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.37profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.53profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.37profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.35profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.00profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.71profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.60profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.78profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.79profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  8.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.86profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  8.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.16profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.78profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  3.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.15profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.94profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.93profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.25profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.23profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.97profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.79profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.96profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.64profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.12profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.75profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.12profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.25profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.91profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.05profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.66profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.90profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  9.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.73profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.55profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.49profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.13profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.74profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.17profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  3.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.72profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.57profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.59profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.64profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.83profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 13.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.49profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 12.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  9.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  7.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.32profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.93profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.29profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.72profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.75profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.46profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.51profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  3.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.90profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.16profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.72profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.63profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.83profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.70profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.59profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.55profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  6.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.93profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.33profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.75profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.30profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  6.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  7.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.79profile/s]
+[2026-08-27 04:31:21 PP0 TP1 EP1] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.53profile/s][2026-08-27 04:31:21 PP0 TP0 EP0] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.97profile/s]
+[2026-08-27 04:31:21 PP0 TP3 EP3] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.43profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.06profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s][2026-08-27 04:31:21 PP0 TP2 EP2] FlashInfer autotune completed.
+[2026-08-27 04:31:21 PP0 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:21 PP0 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:21 PP0 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:21 PP0 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:21 PP0 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:21 PP0 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:21 PP0 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:21 PP0 TP0 EP0] max_total_num_tokens=3801408, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=54, context_len=1048576, available_gpu_mem=27.35 GB
+[2026-08-27 04:31:21 PP0 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.86profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.58profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:01<00:07,  1.18s/profile]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  2.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.27profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.55profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  3.96profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.86profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  6.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  7.01profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.23profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.05profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  6.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  7.06profile/s]
+[2026-08-27 04:31:26 PP1 TP1 EP1] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.40profile/s]
+[2026-08-27 04:31:26 PP1 TP2 EP2] FlashInfer autotune completed.
+[2026-08-27 04:31:26 PP1 TP0 EP0] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.81profile/s]
+[2026-08-27 04:31:26 PP1 TP3 EP3] FlashInfer autotune completed.
+[2026-08-27 04:31:26 PP1 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:26 PP1 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:26 PP1 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:26 PP1 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:26 PP1 TP0 EP0] max_total_num_tokens=3801408, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=54, context_len=1048576, available_gpu_mem=20.45 GB
+[2026-08-27 04:31:26 PP1 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:26 PP1 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:26 PP1 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:26 PP1 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+W0827 04:31:30.805795  2607 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fe48940, len: 6912: Operation not permitted [1]
+W0827 04:31:30.805797  2608 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4f684400, len: 6912: Operation not permitted [1]
+W0827 04:31:30.805945  2609 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4f962700, len: 6912: Operation not permitted [1]
+W0827 04:31:30.805979  2610 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4f9e8a00, len: 6912: Operation not permitted [1]
+W0827 04:31:30.806130  2612 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fe5f640, len: 55296: Operation not permitted [1]
+W0827 04:31:30.806133  2611 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4f83f000, len: 55296: Operation not permitted [1]
+W0827 04:31:30.806264  2613 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fe328c0, len: 6912: Operation not permitted [1]
+W0827 04:31:30.806334  2614 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fa75cc0, len: 13824: Operation not permitted [1]
+W0827 04:31:30.806449  2615 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fa87f80, len: 6912: Operation not permitted [1]
+W0827 04:31:30.806493  2616 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fa29a80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.203712  2635 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50c65e80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.203859  2636 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50c6bb40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.203975  2637 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50d1ffc0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.204025  2638 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50a0c300, len: 6912: Operation not permitted [1]
+W0827 04:31:31.204075  2640 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50de1940, len: 55296: Operation not permitted [1]
+W0827 04:31:31.204082  2639 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50dc93c0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.204438  2641 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50c87940, len: 6912: Operation not permitted [1]
+W0827 04:31:31.204561  2642 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50cffb00, len: 13824: Operation not permitted [1]
+W0827 04:31:31.204641  2644 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50dc7380, len: 6912: Operation not permitted [1]
+W0827 04:31:31.204722  2643 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50c243c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.323861  2661 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50b13c40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.323863  2662 memory_location.cpp:73] Failed to get NUMA node, addr: 0x511191c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.323912  2663 memory_location.cpp:73] Failed to get NUMA node, addr: 0x511b37c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.324069  2664 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50b2da80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.324110  2666 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5115c980, len: 55296: Operation not permitted [1]
+W0827 04:31:31.324155  2665 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5116ac40, len: 55296: Operation not permitted [1]
+W0827 04:31:31.324479  2667 memory_location.cpp:73] Failed to get NUMA node, addr: 0x511a7e80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.324587  2668 memory_location.cpp:73] Failed to get NUMA node, addr: 0x511b8cc0, len: 13824: Operation not permitted [1]
+W0827 04:31:31.324621  2669 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50fc6800, len: 6912: Operation not permitted [1]
+W0827 04:31:31.324682  2670 memory_location.cpp:73] Failed to get NUMA node, addr: 0x511bc780, len: 6912: Operation not permitted [1]
+W0827 04:31:31.543082  2689 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4902e440, len: 6912: Operation not permitted [1]
+W0827 04:31:31.543203  2690 memory_location.cpp:73] Failed to get NUMA node, addr: 0x49484200, len: 6912: Operation not permitted [1]
+W0827 04:31:31.543314  2691 memory_location.cpp:73] Failed to get NUMA node, addr: 0x49119600, len: 6912: Operation not permitted [1]
+W0827 04:31:31.543375  2692 memory_location.cpp:73] Failed to get NUMA node, addr: 0x493dc780, len: 6912: Operation not permitted [1]
+W0827 04:31:31.543467  2693 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51ac5fc0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.543601  2694 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51a9e780, len: 55296: Operation not permitted [1]
+W0827 04:31:31.543751  2695 memory_location.cpp:73] Failed to get NUMA node, addr: 0x49149a40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.543912  2696 memory_location.cpp:73] Failed to get NUMA node, addr: 0x518c0840, len: 13824: Operation not permitted [1]
+W0827 04:31:31.543947  2697 memory_location.cpp:73] Failed to get NUMA node, addr: 0x49303540, len: 6912: Operation not permitted [1]
+W0827 04:31:31.543990  2698 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51951a80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681525  2717 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50376e40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681525  2718 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5058ee40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681525  2716 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5008b980, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681548  2715 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fe0e840, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681727  2719 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5040eb80, len: 55296: Operation not permitted [1]
+W0827 04:31:31.681779  2724 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50209ac0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681811  2720 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5041c3c0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.681828  2722 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5048cc40, len: 13824: Operation not permitted [1]
+W0827 04:31:31.681844  2723 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fde7a00, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681851  2721 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50386580, len: 6912: Operation not permitted [1]
+W0827 04:31:31.730942  2745 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fce4240, len: 6912: Operation not permitted [1]
+W0827 04:31:31.730918  2744 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fed7d40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.730950  2743 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fdcaf80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.731007  2748 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5074c400, len: 55296: Operation not permitted [1]
+W0827 04:31:31.731073  2747 memory_location.cpp:73] Failed to get NUMA node, addr: 0x507bd9c0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.731086  2746 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4f931480, len: 6912: Operation not permitted [1]
+W0827 04:31:31.731294  2749 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd958c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.731361  2751 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd31d80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.731361  2750 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4ffabe00, len: 13824: Operation not permitted [1]
+W0827 04:31:31.731566  2752 memory_location.cpp:73] Failed to get NUMA node, addr: 0x508527c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.808167  2778 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd3ce40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.808190  2777 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd20740, len: 6912: Operation not permitted [1]
+W0827 04:31:31.808192  2776 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4ff86c80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.808637  2782 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fe88780, len: 6912: Operation not permitted [1]
+W0827 04:31:31.808724  2780 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fe48c80, len: 55296: Operation not permitted [1]
+W0827 04:31:31.808760  2781 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fe564c0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.809056  2785 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4ffa7e40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.809049  2779 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4feaa740, len: 6912: Operation not permitted [1]
+W0827 04:31:31.809075  2784 memory_location.cpp:73] Failed to get NUMA node, addr: 0x502fabc0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.809239  2783 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4ffa1f80, len: 13824: Operation not permitted [1]
+W0827 04:31:31.851475  2807 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4ff6fcc0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.851596  2808 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fc0a280, len: 6912: Operation not permitted [1]
+W0827 04:31:31.851615  2809 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd7d700, len: 6912: Operation not permitted [1]
+W0827 04:31:31.851732  2810 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd761c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.851951  2811 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fde6cc0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.851970  2812 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4feeb000, len: 55296: Operation not permitted [1]
+W0827 04:31:31.851962  2813 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fbe2680, len: 6912: Operation not permitted [1]
+W0827 04:31:31.851987  2816 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fbf07c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.852048  2815 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4ffa9940, len: 6912: Operation not permitted [1]
+W0827 04:31:31.852046  2814 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd235c0, len: 13824: Operation not permitted [1]
+[2026-08-27 04:31:32] INFO:     Started server process [1]
+[2026-08-27 04:31:32] INFO:     Waiting for application startup.
+[2026-08-27 04:31:32] Calling super().encode with {'add_special_tokens': False}
+[2026-08-27 04:31:32] Calling super().encode with {'add_special_tokens': False}
+[2026-08-27 04:31:32] INFO:     Application startup complete.
+[2026-08-27 04:31:32] INFO:     Uvicorn running on http://0.0.0.0:30000 (Press CTRL+C to quit)
+[2026-08-27 04:31:33] INFO:     127.0.0.1:60348 - "GET /model_info HTTP/1.1" 200 OK
+[2026-08-27 04:31:33] Start of pd disaggregation warmup ...
+[2026-08-27 04:31:37] INFO:     174.1.60.1:33088 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:38] INFO:     174.1.60.1:33092 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:39] INFO:     174.1.60.1:33106 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:40] INFO:     174.1.60.1:33110 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:41] INFO:     174.1.60.1:42028 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:42] INFO:     174.1.60.1:42030 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:43] INFO:     174.1.60.1:42046 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:44] INFO:     174.1.60.1:42050 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:45] INFO:     174.1.60.1:42060 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:46] INFO:     174.1.60.1:42072 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:47] INFO:     174.1.60.1:42086 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:48] INFO:     174.1.60.1:42092 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:49] INFO:     174.1.60.1:42098 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:50] INFO:     174.1.60.1:42100 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:51] INFO:     174.1.60.1:46852 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:52] INFO:     174.1.60.1:46856 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:53] INFO:     174.1.60.1:46872 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:54] INFO:     174.1.60.1:46876 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:55] INFO:     174.1.60.1:46886 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:56] INFO:     174.1.60.1:46890 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:57] INFO:     174.1.60.1:46898 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:58] INFO:     174.1.60.1:46912 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:59] INFO:     174.1.60.1:46928 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:00] INFO:     174.1.60.1:46936 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:01] INFO:     174.1.60.1:57672 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:02] INFO:     174.1.60.1:57676 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:03] INFO:     174.1.60.1:57686 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:04] INFO:     174.1.60.1:57688 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:05] INFO:     174.1.60.1:57696 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:06] INFO:     174.1.60.1:57708 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:12] INFO:     174.1.60.1:39714 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:13] INFO:     174.1.60.1:39728 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:14] INFO:     174.1.60.1:39736 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:15] INFO:     174.1.60.1:39746 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:16] INFO:     174.1.60.1:39760 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:17] INFO:     174.1.60.1:39772 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:18] INFO:     174.1.60.1:39782 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:19] INFO:     174.1.60.1:39786 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:20] INFO:     174.1.60.1:39800 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:21] INFO:     174.1.60.1:41036 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:22] INFO:     174.1.60.1:41052 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:23] INFO:     174.1.60.1:41058 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:24] INFO:     174.1.60.1:41068 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:25] INFO:     174.1.60.1:41074 - "GET /health HTTP/1.1" 503 Service Unavailable
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank1]:[W827 04:32:26.019157362 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank0]:[W827 04:32:26.019963648 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank3]:[W827 04:32:26.020028119 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank2]:[W827 04:32:26.020057305 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank6]:[W827 04:32:26.045976535 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank7]:[W827 04:32:26.046176399 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank5]:[W827 04:32:26.048033943 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank4]:[W827 04:32:26.048331224 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[2026-08-27 04:32:26] INFO:     174.1.60.1:41078 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:27] INFO:     174.1.60.1:41090 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:28] INFO:     174.1.60.1:41102 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:29] INFO:     174.1.60.1:41104 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:30] INFO:     174.1.60.1:41120 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:31] INFO:     174.1.60.1:57002 - "GET /health HTTP/1.1" 503 Service Unavailable
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank6]:[W827 04:32:31.709756552 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank7]:[W827 04:32:31.709839299 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank5]:[W827 04:32:31.709991082 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank4]:[W827 04:32:31.710123265 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[2026-08-27 04:32:32] INFO:     174.1.60.1:57016 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:33] INFO:     174.1.60.1:57030 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:34] INFO:     174.1.60.1:57046 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:35] INFO:     174.1.60.1:57056 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:36] INFO:     174.1.60.1:57066 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:37] INFO:     174.1.60.1:57070 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:38] INFO:     174.1.60.1:57072 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:39] INFO:     174.1.60.1:57082 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:40] INFO:     174.1.60.1:57098 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:41] INFO:     174.1.60.1:56606 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:50] INFO:     174.1.60.1:56620 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:51] INFO:     174.1.60.1:45658 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:52] INFO:     174.1.60.1:45672 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:53] INFO:     174.1.60.1:45674 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:54] INFO:     174.1.60.1:45690 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:55] INFO:     174.1.60.1:45704 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:56] INFO:     174.1.60.1:45708 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:57] INFO:     174.1.60.1:45714 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:58] INFO:     174.1.60.1:45722 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:00] INFO:     174.1.60.1:45736 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:01] INFO:     174.1.60.1:45746 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:02] INFO:     174.1.60.1:33310 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:03] INFO:     174.1.60.1:33320 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:04] INFO:     174.1.60.1:33324 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:05] INFO:     174.1.60.1:33328 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:06] INFO:     174.1.60.1:33336 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:07] INFO:     174.1.60.1:33350 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:08] INFO:     174.1.60.1:33354 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:09] INFO:     174.1.60.1:33356 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:10] INFO:     174.1.60.1:33368 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:11] INFO:     174.1.60.1:39318 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:12] INFO:     174.1.60.1:39328 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:13] INFO:     174.1.60.1:39336 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:14] INFO:     174.1.60.1:39344 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:15] INFO:     174.1.60.1:39346 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:16] INFO:     174.1.60.1:39352 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:17] INFO:     174.1.60.1:39364 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:18] INFO:     174.1.60.1:39378 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:19] INFO:     174.1.60.1:39390 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:20] INFO:     174.1.60.1:39402 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:25] INFO:     174.1.60.1:37792 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:26] INFO:     174.1.60.1:37804 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:27] INFO:     174.1.60.1:37812 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:28] INFO:     174.1.60.1:37826 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:29] INFO:     174.1.60.1:37830 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:30] INFO:     174.1.60.1:37844 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:31] INFO:     174.1.60.1:44208 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:32] INFO:     174.1.60.1:44220 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:33] INFO:     174.1.60.1:44226 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:34] INFO:     174.1.60.1:44228 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:35] INFO:     174.1.60.1:44238 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:37] INFO:     174.1.60.1:44248 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:38] INFO:     174.1.60.1:44258 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:39] INFO:     174.1.60.1:44272 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:40] INFO:     174.1.60.1:44276 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:41] INFO:     174.1.60.1:44194 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:42] INFO:     174.1.60.1:44198 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:43] INFO:     174.1.60.1:44212 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:44] INFO:     174.1.60.1:44216 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:45] INFO:     174.1.60.1:44222 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:46] INFO:     174.1.60.1:44236 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:47] INFO:     174.1.60.1:44246 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:48] INFO:     174.1.60.1:44252 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:49] INFO:     174.1.60.1:44264 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:50] INFO:     174.1.60.1:44278 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:51] INFO:     174.1.60.1:35728 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:52] INFO:     174.1.60.1:35736 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:53] INFO:     174.1.60.1:35752 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:54] INFO:     174.1.60.1:35754 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:55] INFO:     174.1.60.1:35758 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:04] INFO:     174.1.60.1:44514 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:05] INFO:     174.1.60.1:44518 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:06] INFO:     174.1.60.1:44520 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:07] INFO:     174.1.60.1:44526 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:08] INFO:     174.1.60.1:44530 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:09] INFO:     174.1.60.1:44546 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:10] INFO:     174.1.60.1:44550 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:11] INFO:     174.1.60.1:45098 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:12] INFO:     127.0.0.1:59722 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:12] INFO:     174.1.60.1:45106 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:13] INFO:     174.1.60.1:45122 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:14] INFO:     174.1.60.1:45124 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:15] INFO:     174.1.60.1:45136 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:16] INFO:     174.1.60.1:45146 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:17] INFO:     174.1.60.1:45152 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:19] INFO:     174.1.60.1:45158 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:20] INFO:     174.1.60.1:45170 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:21] INFO:     174.1.60.1:45182 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:22] INFO:     174.1.60.1:39946 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:23] INFO:     174.1.60.1:39958 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:24] INFO:     174.1.60.1:39962 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:25] INFO:     174.1.60.1:39978 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:26] INFO:     174.1.60.1:39990 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:27] INFO:     174.1.60.1:39994 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:28] INFO:     174.1.60.1:39996 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:29] INFO:     174.1.60.1:40002 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:30] INFO:     174.1.60.1:40012 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:31] INFO:     174.1.60.1:46978 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:32] INFO:     174.1.60.1:46988 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:33] INFO:     174.1.60.1:46996 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:34] INFO:     174.1.60.1:47004 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:39] INFO:     174.1.60.1:47020 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:40] INFO:     174.1.60.1:47022 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:41] INFO:     174.1.60.1:59158 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:42] INFO:     174.1.60.1:59170 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:43] INFO:     174.1.60.1:59172 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:44] INFO:     174.1.60.1:59180 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:45] INFO:     174.1.60.1:59184 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:46] INFO:     174.1.60.1:59186 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:47] INFO:     174.1.60.1:59200 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:48] INFO:     174.1.60.1:59202 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:49] INFO:     174.1.60.1:59214 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:50] INFO:     174.1.60.1:59220 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:52] INFO:     174.1.60.1:40566 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:53] INFO:     174.1.60.1:40576 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:54] INFO:     174.1.60.1:40588 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:55] INFO:     174.1.60.1:40596 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:56] INFO:     174.1.60.1:40598 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:57] INFO:     174.1.60.1:40604 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:58] INFO:     174.1.60.1:40608 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:59] INFO:     174.1.60.1:40610 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:00] INFO:     174.1.60.1:40622 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:01] INFO:     174.1.60.1:41102 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:02] INFO:     174.1.60.1:41108 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:03] INFO:     174.1.60.1:41122 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:04] INFO:     174.1.60.1:41124 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:05] INFO:     174.1.60.1:41134 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:06] INFO:     174.1.60.1:41140 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:07] INFO:     174.1.60.1:41142 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:08] INFO:     174.1.60.1:41158 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:09] INFO:     174.1.60.1:41160 - "GET /health HTTP/1.1" 503 Service Unavailable
+[rank3]:[W827 04:35:10.359210278 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank2]:[W827 04:35:10.359272236 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank0]:[W827 04:35:10.360047094 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank1]:[W827 04:35:10.370610409 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[2026-08-27 04:35:10 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.28
+[2026-08-27 04:35:10 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.29
+[2026-08-27 04:35:10] INFO:     127.0.0.1:60364 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 04:35:10] Disaggregation warmup requests completed for all 1 DP ranks
+[2026-08-27 04:35:10] End of disaggregation warmup
+[2026-08-27 04:35:10] The server is fired up and ready to roll!
+[2026-08-27 04:35:29 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.39
+[2026-08-27 04:35:29 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.39
+[2026-08-27 04:35:31 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25.83
+[2026-08-27 04:35:31 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25.84
+[2026-08-27 04:35:32] INFO:     174.1.60.1:49902 - "GET /health HTTP/1.1" 200 OK
+[2026-08-27 04:43:22 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.14
+[2026-08-27 04:43:22 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.14
+[2026-08-27 04:43:23] INFO:     174.1.60.1:33240 - "GET /health HTTP/1.1" 200 OK
+/sgl-workspace/sglang/python/sglang/srt/entrypoints/http_server.py:450: FastAPIDeprecationWarning: ORJSONResponse is deprecated, FastAPI now serializes data directly to JSON bytes via Pydantic when a return type or response model is set, which is faster and doesn't need a custom response class. Read more in the FastAPI docs: https://fastapi.tiangolo.com/advanced/custom-response/#orjson-or-response-model and https://fastapi.tiangolo.com/tutorial/response-model/
+  return await original_handler(ORJSONRequest(request.scope, request.receive))
+[2026-08-27 04:45:43] INFO:     174.1.60.5:35536 - "GET /v1/models HTTP/1.1" 200 OK
+[2026-08-27 04:45:43] INFO:     174.1.60.5:35542 - "GET /v1/models HTTP/1.1" 200 OK
+[2026-08-27 04:48:11] INFO:     174.1.60.5:47258 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 04:48:13 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.22
+[2026-08-27 04:48:13 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.22
+[2026-08-27 04:48:13 PP0 TP1 EP1] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:13 PP0 TP2 EP2] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:13 PP0 TP0 EP0] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:13 PP0 TP3 EP3] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP1 TP1 EP1] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP1 TP2 EP2] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP1 TP0 EP0] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP1 TP3 EP3] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:16] INFO:     174.1.60.5:50348 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 04:48:16] INFO:     174.1.60.5:50364 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 04:54:46] INFO:     174.1.60.5:51618 - "GET /v1/models HTTP/1.1" 200 OK
+[2026-08-27 04:54:46] INFO:     174.1.60.5:51626 - "GET /v1/models HTTP/1.1" 200 OK
+[2026-08-27 04:57:13] INFO:     174.1.60.5:48050 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 04:57:14 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.12
+[2026-08-27 04:57:14 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.12
+[2026-08-27 04:57:14] INFO:     174.1.60.5:48060 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 04:57:14] INFO:     174.1.60.5:48074 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 04:57:40] INFO:     174.1.60.5:51792 - "GET /v1/models HTTP/1.1" 200 OK
+[2026-08-27 04:57:40] INFO:     174.1.60.5:51798 - "GET /v1/models HTTP/1.1" 200 OK
+[2026-08-27 05:00:07] INFO:     174.1.60.5:48432 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:04:21 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1088, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2.55
+[2026-08-27 05:04:21 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1088, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2.55
+[2026-08-27 05:04:22] INFO:     174.1.60.5:36022 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:04:22] INFO:     174.1.60.5:36028 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:04:41 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.19
+[2026-08-27 05:04:41 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.19
+[2026-08-27 05:04:41] INFO:     174.1.60.5:33310 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:04:49] INFO:     174.1.60.5:39866 - "GET /v1/models HTTP/1.1" 200 OK
+[2026-08-27 05:04:49] INFO:     174.1.60.5:39874 - "GET /v1/models HTTP/1.1" 200 OK
+[2026-08-27 05:05:38] INFO:     174.1.60.5:57632 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:05:39 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1.10
+[2026-08-27 05:05:39 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1.10
+[2026-08-27 05:07:16] INFO:     174.1.60.5:42880 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:07:30 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 73.88
+[2026-08-27 05:07:32 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 72.71
+[2026-08-27 05:07:32 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3187.73
+[2026-08-27 05:07:33 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9987.21
+[2026-08-27 05:07:33 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 5011.85
+[2026-08-27 05:07:33 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7713.95
+[2026-08-27 05:07:33] INFO:     174.1.60.5:49324 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:07:38 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8673, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1473.19
+[2026-08-27 05:07:39 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8673, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1235.01
+[2026-08-27 05:07:40 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 481, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4257.97
+[2026-08-27 05:07:40 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 481, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9114.58
+[2026-08-27 05:07:40 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4643.44
+[2026-08-27 05:07:40 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7350.46
+[2026-08-27 05:07:40] INFO:     174.1.60.5:49326 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:07:46 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8656, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1470.23
+[2026-08-27 05:07:47 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8656, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1236.74
+[2026-08-27 05:07:48 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 464, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4291.82
+[2026-08-27 05:07:48 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 464, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9091.63
+[2026-08-27 05:07:48 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4126.73
+[2026-08-27 05:07:48 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6652.98
+[2026-08-27 05:07:48] INFO:     174.1.60.5:38928 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:07:53 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8663, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1477.97
+[2026-08-27 05:07:54 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8663, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1240.34
+[2026-08-27 05:07:55 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 471, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4270.28
+[2026-08-27 05:07:55 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 471, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9111.26
+[2026-08-27 05:07:55 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4676.01
+[2026-08-27 05:07:55 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7851.58
+[2026-08-27 05:07:56] INFO:     174.1.60.5:34214 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:08:01 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8659, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1474.57
+[2026-08-27 05:08:02 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8659, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1237.36
+[2026-08-27 05:08:03 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 467, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4273.37
+[2026-08-27 05:08:03 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 467, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9096.53
+[2026-08-27 05:08:03 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4082.13
+[2026-08-27 05:08:03 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6360.56
+[2026-08-27 05:08:03] INFO:     174.1.60.5:38204 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:08:08 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8692, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1476.50
+[2026-08-27 05:08:10 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8692, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1239.29
+[2026-08-27 05:08:10 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 500, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4267.60
+[2026-08-27 05:08:10 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 500, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8989.39
+[2026-08-27 05:08:11 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4419.86
+[2026-08-27 05:08:11 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8656.56
+[2026-08-27 05:08:11] INFO:     174.1.60.5:47402 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:08:16 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8557, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1477.68
+[2026-08-27 05:08:17 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8557, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1239.64
+[2026-08-27 05:08:18 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 365, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4273.32
+[2026-08-27 05:08:18 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 365, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9138.56
+[2026-08-27 05:08:18 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4206.66
+[2026-08-27 05:08:18 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8364.41
+[2026-08-27 05:08:18] INFO:     174.1.60.5:47404 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:08:24 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8683, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1475.28
+[2026-08-27 05:08:25 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8683, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1237.96
+[2026-08-27 05:08:26 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 491, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4263.38
+[2026-08-27 05:08:26 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 491, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9074.22
+[2026-08-27 05:08:26 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4537.22
+[2026-08-27 05:08:26 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7174.96
+[2026-08-27 05:08:26] INFO:     174.1.60.5:48174 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:08:26] INFO:     174.1.60.5:48178 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:09:17] INFO:     174.1.60.5:41772 - "GET /v1/models HTTP/1.1" 200 OK
+[2026-08-27 05:09:17] INFO:     174.1.60.5:41776 - "GET /v1/models HTTP/1.1" 200 OK
+[2026-08-27 05:11:44] INFO:     174.1.60.5:45272 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:11:45 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.32
+[2026-08-27 05:11:45 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.32
+E0827 05:11:45.751729  1489 worker_pool.cpp:1217] CQ completion timeout diagnostic: context=mlx5_3, outstanding=2, outstanding_age_ms=0, poll_gap_ms=199552, last_poll_interval_ms=0, max_poll_interval_ms=539649, submitted=34398, processed=34396
+[2026-08-27 05:11:46] INFO:     174.1.60.5:45274 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:11:46] INFO:     174.1.60.5:45276 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:12:14] INFO:     174.1.60.5:34478 - "GET /v1/models HTTP/1.1" 200 OK
+[2026-08-27 05:12:14] INFO:     174.1.60.5:34486 - "GET /v1/models HTTP/1.1" 200 OK
+[2026-08-27 05:14:42 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1088, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6.16
+[2026-08-27 05:14:42 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1088, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6.16
+[2026-08-27 05:14:42] INFO:     174.1.60.5:46142 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:14:43] INFO:     174.1.60.5:46156 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:14:43] INFO:     174.1.60.5:46162 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:15:12] INFO:     174.1.60.5:60686 - "GET /v1/models HTTP/1.1" 200 OK
+[2026-08-27 05:15:12] INFO:     174.1.60.5:60696 - "GET /v1/models HTTP/1.1" 200 OK
+[2026-08-27 05:17:40] INFO:     174.1.60.5:49298 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:17:45 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 44.69
+[2026-08-27 05:17:46 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 44.43
+[2026-08-27 05:17:47 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4291.89
+[2026-08-27 05:17:47 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9142.33
+[2026-08-27 05:17:47 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4792.42
+[2026-08-27 05:17:47 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8482.06
+[2026-08-27 05:17:47] INFO:     174.1.60.5:45358 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:17:53 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8673, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1475.80
+[2026-08-27 05:17:54 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8673, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1239.62
+[2026-08-27 05:17:55 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 481, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4273.03
+[2026-08-27 05:17:55 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 481, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9047.83
+[2026-08-27 05:17:55 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4673.57
+[2026-08-27 05:17:55 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8029.35
+[2026-08-27 05:17:55] INFO:     174.1.60.5:37922 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:18:00 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8656, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1479.36
+[2026-08-27 05:18:01 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8656, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1239.75
+[2026-08-27 05:18:02 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 464, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4268.49
+[2026-08-27 05:18:02 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 464, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9151.27
+[2026-08-27 05:18:02 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4773.15
+[2026-08-27 05:18:02 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8723.97
+[2026-08-27 05:18:02] INFO:     174.1.60.5:59842 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:18:08 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8663, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1480.27
+[2026-08-27 05:18:09 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8663, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1240.59
+[2026-08-27 05:18:10 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 471, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4269.06
+[2026-08-27 05:18:10 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 471, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9108.64
+[2026-08-27 05:18:10 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4695.76
+[2026-08-27 05:18:10 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8458.08
+[2026-08-27 05:18:10] INFO:     174.1.60.5:59848 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:18:15 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8659, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1476.05
+[2026-08-27 05:18:17 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8659, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1238.88
+[2026-08-27 05:18:17 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 467, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4269.01
+[2026-08-27 05:18:17 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 467, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9060.08
+[2026-08-27 05:18:17 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4685.85
+[2026-08-27 05:18:17 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8293.92
+[2026-08-27 05:18:18] INFO:     174.1.60.5:46946 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:18:23 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8692, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1478.44
+[2026-08-27 05:18:24 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8692, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1238.63
+[2026-08-27 05:18:25 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 500, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4261.28
+[2026-08-27 05:18:25 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 500, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9140.66
+[2026-08-27 05:18:25 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4667.18
+[2026-08-27 05:18:25 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8159.39
+[2026-08-27 05:18:25] INFO:     174.1.60.5:42086 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:18:31 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8557, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1480.63
+[2026-08-27 05:18:32 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8557, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1240.79
+[2026-08-27 05:18:32 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 365, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4272.82
+[2026-08-27 05:18:33 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 365, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9138.60
+[2026-08-27 05:18:33 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4248.66
+[2026-08-27 05:18:33 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8746.98
+[2026-08-27 05:18:33] INFO:     174.1.60.5:36136 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:18:38 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8683, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1480.90
+[2026-08-27 05:18:39 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8683, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1240.92
+[2026-08-27 05:18:40 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 491, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4261.25
+[2026-08-27 05:18:40 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 491, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9102.59
+[2026-08-27 05:18:40 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4532.61
+[2026-08-27 05:18:40 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8049.42
+[2026-08-27 05:18:40] INFO:     174.1.60.5:36144 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:18:40] INFO:     174.1.60.5:36150 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:19:09] INFO:     174.1.60.5:57946 - "GET /v1/models HTTP/1.1" 200 OK
+[2026-08-27 05:19:09] INFO:     174.1.60.5:57958 - "GET /v1/models HTTP/1.1" 200 OK
+[2026-08-27 05:21:37] INFO:     174.1.60.5:50774 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:37] INFO:     174.1.60.5:50788 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:37] INFO:     174.1.60.5:50802 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:37] INFO:     174.1.60.5:50814 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:37] INFO:     174.1.60.5:50826 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:37] INFO:     174.1.60.5:50840 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:37] INFO:     174.1.60.5:50844 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:37] INFO:     174.1.60.5:50846 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:43 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 2, #pending-token: 8652, #bootstrap-req: 2, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 44.89
+[2026-08-27 05:21:44 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 2, #pending-token: 8652, #bootstrap-req: 2, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 44.68
+[2026-08-27 05:21:44 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 3, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 5768.82
+[2026-08-27 05:21:45 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 3, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 5827.47
+[2026-08-27 05:21:45 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 3, #pending-token: 9185, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9671.83
+[2026-08-27 05:21:45 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 3, #pending-token: 9185, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 19750.66
+[2026-08-27 05:21:46 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 2, #pending-token: 993, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6916.66
+[2026-08-27 05:21:46 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 2, #pending-token: 993, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8672.27
+[2026-08-27 05:21:47 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 2, #pending-token: 9680, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10637.56
+[2026-08-27 05:21:47 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 2, #pending-token: 9680, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10354.39
+[2026-08-27 05:21:48 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1488, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8222.75
+[2026-08-27 05:21:48 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1488, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8344.22
+[2026-08-27 05:21:49 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 10199, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10480.00
+[2026-08-27 05:21:49 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 10199, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10387.32
+[2026-08-27 05:21:50 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 18858, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8578.91
+[2026-08-27 05:21:50 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 18858, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8578.44
+[2026-08-27 05:21:50 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 27591, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9988.39
+[2026-08-27 05:21:51 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 27591, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9972.71
+[2026-08-27 05:21:51 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 53023, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8837.11
+[2026-08-27 05:21:51] INFO:     174.1.60.5:43264 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:52 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 53023, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9171.43
+[2026-08-27 05:21:52 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 44876, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9896.54
+[2026-08-27 05:21:52 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 44876, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10032.97
+[2026-08-27 05:21:53 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 36684, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9174.07
+[2026-08-27 05:21:53] INFO:     174.1.60.5:43266 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:53 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 36684, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9172.94
+[2026-08-27 05:21:54 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 28504, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9586.45
+[2026-08-27 05:21:54 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 28504, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9826.60
+[2026-08-27 05:21:55 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 20312, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9357.29
+[2026-08-27 05:21:55] INFO:     174.1.60.5:43280 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:55 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 20312, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9123.76
+[2026-08-27 05:21:56 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 12139, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9447.89
+[2026-08-27 05:21:56 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 12139, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9658.04
+[2026-08-27 05:21:57 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 3947, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9443.97
+[2026-08-27 05:21:57] INFO:     174.1.60.5:43282 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:57 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 3947, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9638.68
+[2026-08-27 05:21:57 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 3968, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7366.63
+[2026-08-27 05:21:57 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 3968, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8902.23
+[2026-08-27 05:21:57] INFO:     174.1.60.5:43298 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:57] INFO:     174.1.60.5:43310 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:59] INFO:     174.1.60.5:43312 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:00] INFO:     174.1.60.5:43328 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:02 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 0, #pending-token: 8608, #bootstrap-req: 4, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1515.73
+[2026-08-27 05:22:03 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 0, #pending-token: 8608, #bootstrap-req: 4, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1303.59
+[2026-08-27 05:22:04 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 416, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4709.52
+[2026-08-27 05:22:04 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 416, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8416.60
+[2026-08-27 05:22:05 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 9149, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10991.56
+[2026-08-27 05:22:05 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 9149, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9867.38
+[2026-08-27 05:22:06 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 957, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7879.22
+[2026-08-27 05:22:06 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 957, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7915.28
+[2026-08-27 05:22:07 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 9638, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10712.86
+[2026-08-27 05:22:07 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 9638, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11079.28
+[2026-08-27 05:22:08 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1446, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8251.19
+[2026-08-27 05:22:08 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1446, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8053.96
+[2026-08-27 05:22:09 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 10033, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10494.65
+[2026-08-27 05:22:09 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 10033, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11035.63
+[2026-08-27 05:22:09 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1841, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8664.83
+[2026-08-27 05:22:10 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1841, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8259.51
+[2026-08-27 05:22:10 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1856, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6733.51
+[2026-08-27 05:22:10 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1856, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2578.77
+[2026-08-27 05:22:11 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8540, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7967.20
+[2026-08-27 05:22:11] INFO:     174.1.60.5:53576 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:11 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8540, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9138.35
+[2026-08-27 05:22:12 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 17066, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6059.77
+[2026-08-27 05:22:12 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 17066, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8568.79
+[2026-08-27 05:22:13 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8910, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10873.13
+[2026-08-27 05:22:13] INFO:     174.1.60.5:53584 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:13 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8910, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9603.70
+[2026-08-27 05:22:14 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 17466, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7794.12
+[2026-08-27 05:22:14 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 17466, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8622.72
+[2026-08-27 05:22:15 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9324, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10734.99
+[2026-08-27 05:22:15] INFO:     174.1.60.5:53590 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:15 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9324, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9808.25
+[2026-08-27 05:22:16 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 17751, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8115.61
+[2026-08-27 05:22:16 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 17751, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8814.22
+[2026-08-27 05:22:16 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9579, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10679.21
+[2026-08-27 05:22:17] INFO:     174.1.60.5:53596 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:17 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9579, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9729.78
+[2026-08-27 05:22:17 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1387, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8283.85
+[2026-08-27 05:22:18 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1387, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9429.57
+[2026-08-27 05:22:18 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1408, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6343.40
+[2026-08-27 05:22:18 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1408, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8479.29
+[2026-08-27 05:22:18] INFO:     174.1.60.5:53608 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:18] INFO:     174.1.60.5:53620 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:20] INFO:     174.1.60.5:53636 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:21] INFO:     174.1.60.5:44034 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:23 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8547, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1526.94
+[2026-08-27 05:22:24 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8547, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1315.46
+[2026-08-27 05:22:25 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 355, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4760.70
+[2026-08-27 05:22:25 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 355, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8428.33
+[2026-08-27 05:22:26 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8932, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10829.34
+[2026-08-27 05:22:26 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8932, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9556.73
+[2026-08-27 05:22:27 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 740, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7781.14
+[2026-08-27 05:22:27 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 740, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7932.50
+[2026-08-27 05:22:27 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9259, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10727.88
+[2026-08-27 05:22:28 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9259, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10866.67
+[2026-08-27 05:22:28 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1067, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8113.36
+[2026-08-27 05:22:29 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1067, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8138.49
+[2026-08-27 05:22:29 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9745, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10655.24
+[2026-08-27 05:22:29 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9745, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10900.49
+[2026-08-27 05:22:30 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1553, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8338.75
+[2026-08-27 05:22:30 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1553, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8227.97
+[2026-08-27 05:22:30 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1600, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6665.39
+[2026-08-27 05:22:31 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1600, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2199.04
+[2026-08-27 05:22:31 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8602, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7710.31
+[2026-08-27 05:22:31] INFO:     174.1.60.5:37214 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:32 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8602, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9261.97
+[2026-08-27 05:22:33 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 410, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6121.64
+[2026-08-27 05:22:33 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 410, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8576.46
+[2026-08-27 05:22:34 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8696, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10967.17
+[2026-08-27 05:22:34] INFO:     174.1.60.5:37230 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:34 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8696, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9567.88
+[2026-08-27 05:22:35 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 504, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7845.65
+[2026-08-27 05:22:35 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 504, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8659.79
+[2026-08-27 05:22:35 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9164, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10983.83
+[2026-08-27 05:22:35] INFO:     174.1.60.5:37246 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:36 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9164, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9874.30
+[2026-08-27 05:22:36 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 972, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7934.46
+[2026-08-27 05:22:36 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 972, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8762.34
+[2026-08-27 05:22:37 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9322, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10758.34
+[2026-08-27 05:22:37] INFO:     174.1.60.5:37262 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:37 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9322, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9763.35
+[2026-08-27 05:22:38 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1130, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8224.94
+[2026-08-27 05:22:38 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1130, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9463.33
+[2026-08-27 05:22:38 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1152, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6381.46
+[2026-08-27 05:22:38 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1152, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8785.04
+[2026-08-27 05:22:38] INFO:     174.1.60.5:37274 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:39] INFO:     174.1.60.5:37280 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:40] INFO:     174.1.60.5:37284 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:42] INFO:     174.1.60.5:57706 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:44 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8602, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1531.42
+[2026-08-27 05:22:44 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8602, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1318.33
+[2026-08-27 05:22:45 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 410, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4775.67
+[2026-08-27 05:22:45 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 410, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8534.85
+[2026-08-27 05:22:46 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 9105, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10895.13
+[2026-08-27 05:22:46 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 9105, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9621.85
+[2026-08-27 05:22:47 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 913, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7859.49
+[2026-08-27 05:22:47 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 913, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8012.84
+[2026-08-27 05:22:48 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9375, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10691.27
+[2026-08-27 05:22:48 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9375, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10846.99
+[2026-08-27 05:22:49 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1183, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8180.68
+[2026-08-27 05:22:49 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1183, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8142.25
+[2026-08-27 05:22:50 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9868, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10633.74
+[2026-08-27 05:22:50 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9868, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10953.99
+[2026-08-27 05:22:51 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1676, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8427.87
+[2026-08-27 05:22:51 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1676, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8243.07
+[2026-08-27 05:22:51 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6603.73
+[2026-08-27 05:22:52 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2380.37
+[2026-08-27 05:22:52 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8467, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7941.05
+[2026-08-27 05:22:52] INFO:     174.1.60.5:59718 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:52 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8467, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9276.71
+[2026-08-27 05:22:53 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 275, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6098.49
+[2026-08-27 05:22:53 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 275, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8603.25
+[2026-08-27 05:22:54 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8988, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11035.64
+[2026-08-27 05:22:54] INFO:     174.1.60.5:59730 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:54 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8988, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9672.42
+[2026-08-27 05:22:55 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 796, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7825.33
+[2026-08-27 05:22:55 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 796, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8635.70
+[2026-08-27 05:22:56 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9363, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10703.37
+[2026-08-27 05:22:56] INFO:     174.1.60.5:59738 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:56 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9363, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9819.19
+[2026-08-27 05:22:57 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 1171, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8175.11
+[2026-08-27 05:22:57 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 1171, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8823.65
+[2026-08-27 05:22:58 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9880, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10599.05
+[2026-08-27 05:22:58] INFO:     174.1.60.5:59744 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:58 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9880, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9922.77
+[2026-08-27 05:22:59 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1688, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8423.21
+[2026-08-27 05:22:59 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1688, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9407.16
+[2026-08-27 05:22:59 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6506.01
+[2026-08-27 05:22:59 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8452.04
+[2026-08-27 05:22:59] INFO:     174.1.60.5:59746 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:23:00] INFO:     174.1.60.5:59748 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:23:01] INFO:     174.1.60.5:59088 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:23:03] INFO:     174.1.60.5:59094 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:23:04 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8601, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1533.96
+[2026-08-27 05:23:05 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8601, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1320.52
+[2026-08-27 05:23:06 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 409, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4797.37
+[2026-08-27 05:23:06 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 409, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8455.98
+[2026-08-27 05:23:07 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 9104, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10736.15
+[2026-08-27 05:23:07 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 9104, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9859.62
+[2026-08-27 05:23:08 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 912, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8011.36
+[2026-08-27 05:23:08 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 912, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8059.42
+[2026-08-27 05:23:08 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9485, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10534.48
+[2026-08-27 05:23:09 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9485, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10717.32
+[2026-08-27 05:23:09 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1293, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8282.08
+[2026-08-27 05:23:10 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1293, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8188.84
+[2026-08-27 05:23:10 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9900, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10556.34
+[2026-08-27 05:23:10 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9900, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10839.28
+[2026-08-27 05:23:11 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1708, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8457.93
+[2026-08-27 05:23:11 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1708, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8302.17
+[2026-08-27 05:23:11 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6651.64
+[2026-08-27 05:23:12 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2387.61
+[2026-08-27 05:23:12 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8525, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7923.97
+[2026-08-27 05:23:13 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8525, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9210.87
+[2026-08-27 05:23:14 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 333, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6091.62
+[2026-08-27 05:23:14 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 333, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8517.86
+[2026-08-27 05:23:15 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8909, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10878.98
+[2026-08-27 05:23:15 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8909, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9736.23
+[2026-08-27 05:23:16 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 717, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7902.78
+[2026-08-27 05:23:16 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 717, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8689.24
+[2026-08-27 05:23:16 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9452, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10764.37
+[2026-08-27 05:23:17 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9452, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9778.82
+[2026-08-27 05:23:17 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 1260, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8087.43
+[2026-08-27 05:23:18 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 1260, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8848.19
+[2026-08-27 05:23:18 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9830, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10642.74
+[2026-08-27 05:23:18 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9830, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9957.02
+[2026-08-27 05:23:19 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1638, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8506.01
+[2026-08-27 05:23:19 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1638, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9493.47
+[2026-08-27 05:23:19 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1664, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6758.94
+[2026-08-27 05:23:19 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1664, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9006.24
+[2026-08-27 05:23:20] INFO:     174.1.60.5:56558 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:23:20] INFO:     174.1.60.5:56572 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:32:52 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.11
+[2026-08-27 05:32:52 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.11
+[2026-08-27 05:32:53] INFO:     174.1.60.1:38326 - "GET /health HTTP/1.1" 200 OK
+[2026-08-27 05:33:19] INFO:     174.1.60.5:50438 - "GET /v1/models HTTP/1.1" 200 OK
+[2026-08-27 05:33:19] INFO:     174.1.60.5:50442 - "GET /v1/models HTTP/1.1" 200 OK
+[2026-08-27 05:35:46] INFO:     174.1.60.5:53276 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:35:52 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 45.65
+[2026-08-27 05:35:53 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 45.39
+[2026-08-27 05:35:54 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4281.94
+[2026-08-27 05:35:54 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9046.45
+[2026-08-27 05:35:54 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4655.95
+[2026-08-27 05:35:54 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8048.57
+[2026-08-27 05:36:14] INFO:     174.1.60.5:41824 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:36:14] INFO:     174.1.60.5:41836 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:36:34 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1.59
+[2026-08-27 05:36:34 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1.59
+[2026-08-27 05:36:35] INFO:     174.1.60.1:37570 - "GET /health HTTP/1.1" 200 OK
+[2026-08-27 05:36:36] INFO:     174.1.60.5:52850 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:39:56] INFO:     174.1.60.5:59122 - "GET /v1/models HTTP/1.1" 200 OK
+[2026-08-27 05:39:56] INFO:     174.1.60.5:59138 - "GET /v1/models HTTP/1.1" 200 OK
+[2026-08-27 05:42:24] INFO:     174.1.60.5:45934 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:42:24] INFO:     174.1.60.5:45940 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:42:24] INFO:     174.1.60.5:45948 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:42:24] INFO:     174.1.60.5:45962 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:42:24] INFO:     174.1.60.5:45972 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:42:24] INFO:     174.1.60.5:45986 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:42:24] INFO:     174.1.60.5:45990 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:42:24] INFO:     174.1.60.5:46004 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:42:30 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 8652, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 23.05
+[2026-08-27 05:42:30 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 8652, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 22.99
+[2026-08-27 05:42:31 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 3, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4672.22
+[2026-08-27 05:42:31 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 3, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8067.88
+[2026-08-27 05:42:32 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 3, #pending-token: 9185, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10975.93
+[2026-08-27 05:42:32 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 3, #pending-token: 9185, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10328.06
+[2026-08-27 05:42:33 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 2, #pending-token: 993, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7813.04
+[2026-08-27 05:42:33 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 2, #pending-token: 993, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8088.91
+[2026-08-27 05:42:34 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 2, #pending-token: 9680, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10652.10
+[2026-08-27 05:42:34 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 2, #pending-token: 9680, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10495.19
+[2026-08-27 05:42:35 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1488, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8192.12
+[2026-08-27 05:42:35 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1488, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8276.22
+[2026-08-27 05:42:36 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 10199, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10481.84
+[2026-08-27 05:42:36 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 10199, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10379.38
+[2026-08-27 05:42:37 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 18858, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8557.89
+[2026-08-27 05:42:37 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 18858, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8557.27
+[2026-08-27 05:42:37 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 10707, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10088.36
+[2026-08-27 05:42:38 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 10707, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10114.21
+[2026-08-27 05:42:38 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 53023, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8782.27
+[2026-08-27 05:42:38 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 53023, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9104.10
+[2026-08-27 05:42:39 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 44876, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9774.95
+[2026-08-27 05:42:39 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 44876, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10169.80
+[2026-08-27 05:42:40 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 36684, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9184.92
+[2026-08-27 05:42:40 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 36684, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9020.55
+[2026-08-27 05:42:41 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 28504, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9562.85
+[2026-08-27 05:42:41 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 28504, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9893.19
+[2026-08-27 05:42:42 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 20312, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9374.35
+[2026-08-27 05:42:42 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 20312, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9155.04
+[2026-08-27 05:42:43 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 12139, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9495.23
+[2026-08-27 05:42:43 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 12139, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9663.71
+[2026-08-27 05:42:44 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 3947, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9440.41
+[2026-08-27 05:42:44 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 3947, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9682.42
+[2026-08-27 05:42:44 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 3968, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7442.65
+[2026-08-27 05:42:44 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 3968, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8773.51
+[2026-08-27 05:43:11] INFO:     174.1.60.5:47410 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:43:14] INFO:     174.1.60.5:36402 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:43:16] INFO:     174.1.60.5:36418 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:43:16 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8608, #bootstrap-req: 1, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 255.62
+[2026-08-27 05:43:17 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8608, #bootstrap-req: 1, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 247.50
+[2026-08-27 05:43:18] INFO:     174.1.60.5:36434 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:43:18 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 416, #bootstrap-req: 2, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4294.81
+[2026-08-27 05:43:18 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 416, #bootstrap-req: 2, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8707.36
+[2026-08-27 05:43:18 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 448, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2765.42
+[2026-08-27 05:43:18 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 448, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 5334.39
+[2026-08-27 05:43:18] INFO:     174.1.60.5:36438 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:43:18] INFO:     174.1.60.5:36454 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:43:19] INFO:     174.1.60.5:36470 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:43:19] INFO:     174.1.60.5:36472 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:43:23 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.06, #running-req: 0, #queue-req: 0, #pending-token: 8701, #bootstrap-req: 4, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1545.56
+[2026-08-27 05:43:25 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.06, #running-req: 0, #queue-req: 0, #pending-token: 8701, #bootstrap-req: 4, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1288.88
+[2026-08-27 05:43:25 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.06, #running-req: 0, #queue-req: 0, #pending-token: 509, #bootstrap-req: 4, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4260.69
+[2026-08-27 05:43:25 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.06, #running-req: 0, #queue-req: 0, #pending-token: 509, #bootstrap-req: 4, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9013.24
+[2026-08-27 05:43:26 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.01, mamba usage: 0.06, #running-req: 0, #queue-req: 4, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4485.44
+[2026-08-27 05:43:26 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.01, mamba usage: 0.06, #running-req: 0, #queue-req: 4, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 781.73
+[2026-08-27 05:43:27 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 3, #pending-token: 8678, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7207.05
+[2026-08-27 05:43:27 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 3, #pending-token: 8678, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9450.64
+[2026-08-27 05:43:28 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 3, #pending-token: 17239, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6955.26
+[2026-08-27 05:43:28 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 3, #pending-token: 17239, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7759.19
+[2026-08-27 05:43:29 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 2, #pending-token: 9073, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10863.40
+[2026-08-27 05:43:29 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 2, #pending-token: 9073, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10829.36
+[2026-08-27 05:43:30 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 2, #pending-token: 881, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7872.99
+[2026-08-27 05:43:30 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 2, #pending-token: 881, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7937.70
+[2026-08-27 05:43:30 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 896, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4392.16
+[2026-08-27 05:43:31 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 896, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1260.87
+[2026-08-27 05:43:32 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 58625, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3614.28
+[2026-08-27 05:43:33 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 58625, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3397.58
+[2026-08-27 05:43:34 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 50433, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4895.64
+[2026-08-27 05:43:34 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 50433, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8625.84
+[2026-08-27 05:43:35 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 42277, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10812.45
+[2026-08-27 05:43:35 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 42277, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9481.59
+[2026-08-27 05:43:36 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 34085, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7652.72
+[2026-08-27 05:43:36 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 34085, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8542.14
+[2026-08-27 05:43:36 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 25943, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10736.52
+[2026-08-27 05:43:37 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 25943, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9586.39
+[2026-08-27 05:43:37 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 17751, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7891.97
+[2026-08-27 05:43:38 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 17751, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8744.84
+[2026-08-27 05:43:38 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 9579, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10693.58
+[2026-08-27 05:43:38 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 9579, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9896.76
+[2026-08-27 05:43:39 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 0, #pending-token: 1387, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8322.70
+[2026-08-27 05:43:39 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 0, #pending-token: 1387, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9330.84
+[2026-08-27 05:43:39 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1408, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6351.50
+[2026-08-27 05:43:39 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1408, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8763.59
+[2026-08-27 05:43:42] INFO:     174.1.60.5:51864 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:43:48 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8547, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 993.97
+[2026-08-27 05:43:49 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8547, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 882.45
+[2026-08-27 05:43:50 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 355, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4317.25
+[2026-08-27 05:43:50 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 355, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9035.61
+[2026-08-27 05:43:50 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4191.25
+[2026-08-27 05:43:50 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9101.70
+[2026-08-27 05:44:01] INFO:     174.1.60.5:43536 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:44:06] INFO:     174.1.60.5:43552 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:44:06 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8548, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 487.70
+[2026-08-27 05:44:07 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8548, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 459.02
+[2026-08-27 05:44:08 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 356, #bootstrap-req: 1, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4298.64
+[2026-08-27 05:44:08 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 356, #bootstrap-req: 1, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9065.57
+[2026-08-27 05:44:08 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4370.68
+[2026-08-27 05:44:08 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9155.58
+[2026-08-27 05:44:08] INFO:     174.1.60.5:43562 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:44:13] INFO:     174.1.60.5:36440 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:44:13] INFO:     174.1.60.5:36446 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:44:13] INFO:     174.1.60.5:36450 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:44:13] INFO:     174.1.60.5:36466 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:44:14 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8491, #bootstrap-req: 1, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1543.75
+[2026-08-27 05:44:15 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8491, #bootstrap-req: 1, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1283.29
+[2026-08-27 05:44:16 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 299, #bootstrap-req: 5, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4239.25
+[2026-08-27 05:44:16 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 299, #bootstrap-req: 5, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9055.93
+[2026-08-27 05:44:16 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 320, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 1, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3547.91
+[2026-08-27 05:44:16 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 320, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 1, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2307.31
+[2026-08-27 05:44:19] INFO:     174.1.60.5:36474 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:44:21 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.06, #running-req: 0, #queue-req: 3, #pending-token: 8657, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1563.75
+[2026-08-27 05:44:22 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.06, #running-req: 0, #queue-req: 3, #pending-token: 8657, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1324.49
+[2026-08-27 05:44:23 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 2, #pending-token: 465, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4301.20
+[2026-08-27 05:44:23 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 2, #pending-token: 465, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8092.76
+[2026-08-27 05:44:23 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.01, mamba usage: 0.06, #running-req: 0, #queue-req: 2, #pending-token: 0, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3103.43
+[2026-08-27 05:44:24 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.01, mamba usage: 0.06, #running-req: 0, #queue-req: 2, #pending-token: 0, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 701.92
+[2026-08-27 05:44:26 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 58376, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2735.75
+[2026-08-27 05:44:27 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 58376, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2622.52
+[2026-08-27 05:44:27 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 50184, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 5891.42
+[2026-08-27 05:44:28 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 50184, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10056.43
+[2026-08-27 05:44:28 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 42030, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10930.59
+[2026-08-27 05:44:29 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 42030, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9394.30
+[2026-08-27 05:44:29 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 33838, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6627.45
+[2026-08-27 05:44:30 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 33838, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7897.54
+[2026-08-27 05:44:30 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 25654, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10675.51
+[2026-08-27 05:44:30 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 25654, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10417.25
+[2026-08-27 05:44:31 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 17462, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7789.44
+[2026-08-27 05:44:31 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 17462, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8682.67
+[2026-08-27 05:44:32] INFO:     174.1.60.5:39784 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:44:32 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 9322, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10702.04
+[2026-08-27 05:44:32 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 9322, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9816.51
+[2026-08-27 05:44:33 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1130, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8256.57
+[2026-08-27 05:44:33 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1130, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9343.53
+[2026-08-27 05:44:33 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1152, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6176.78
+[2026-08-27 05:44:34 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1152, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1657.02
+[2026-08-27 05:44:34 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8602, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7347.77
+[2026-08-27 05:44:35 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8602, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7743.44
+[2026-08-27 05:44:36 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 410, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6019.53
+[2026-08-27 05:44:36 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 410, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9056.78
+[2026-08-27 05:44:36 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 448, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4496.50
+[2026-08-27 05:44:36 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 448, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9070.83
+[2026-08-27 05:44:41 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8657, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1559.66
+[2026-08-27 05:44:42 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8657, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1300.97
+[2026-08-27 05:44:43 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 465, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4316.14
+[2026-08-27 05:44:43 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 465, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9115.80
+[2026-08-27 05:44:43 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4614.15
+[2026-08-27 05:44:43 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8167.22
+[2026-08-27 05:44:45] INFO:     174.1.60.5:44142 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:44:50 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8415, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1170.41
+[2026-08-27 05:44:51 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8415, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1014.43
+[2026-08-27 05:44:52 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 223, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4304.48
+[2026-08-27 05:44:52 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 223, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9325.89
+[2026-08-27 05:44:52 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 256, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3066.33
+[2026-08-27 05:44:52 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 256, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7687.56
+[2026-08-27 05:45:00] INFO:     174.1.60.5:55924 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:45:05 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 615.58
+[2026-08-27 05:45:06 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 570.90
+[2026-08-27 05:45:07] INFO:     174.1.60.5:35974 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:45:07 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4303.66
+[2026-08-27 05:45:07 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9022.36
+[2026-08-27 05:45:07 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4615.06
+[2026-08-27 05:45:07 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7574.98
+[2026-08-27 05:45:09] INFO:     174.1.60.5:35984 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:45:09] INFO:     174.1.60.5:35998 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:45:09] INFO:     174.1.60.5:36006 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:45:11] INFO:     174.1.60.5:38932 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:45:13 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8467, #bootstrap-req: 4, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1531.05
+[2026-08-27 05:45:14 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8467, #bootstrap-req: 4, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1280.94
+[2026-08-27 05:45:15] INFO:     174.1.60.5:38944 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:45:15 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 275, #bootstrap-req: 4, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4297.37
+[2026-08-27 05:45:15 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 275, #bootstrap-req: 4, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8967.07
+[2026-08-27 05:45:15 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 320, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3333.43
+[2026-08-27 05:45:15 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 320, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6518.76
+[2026-08-27 05:45:20 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 42247, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1506.63
+[2026-08-27 05:45:21] INFO:     174.1.60.5:48332 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:45:21 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 42247, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1298.25
+[2026-08-27 05:45:22 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 34055, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4702.97
+[2026-08-27 05:45:22 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 34055, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8062.04
+[2026-08-27 05:45:23 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 42692, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10918.74
+[2026-08-27 05:45:23 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 42692, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10280.81
+[2026-08-27 05:45:24 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 34500, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7853.47
+[2026-08-27 05:45:24 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 34500, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8376.29
+[2026-08-27 05:45:24 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 26353, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10775.59
+[2026-08-27 05:45:25 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 26353, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10204.17
+[2026-08-27 05:45:25 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 18161, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8109.19
+[2026-08-27 05:45:26 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 18161, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8744.57
+[2026-08-27 05:45:26 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 26857, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10538.09
+[2026-08-27 05:45:26 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 26857, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9974.50
+[2026-08-27 05:45:27 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 18665, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8509.43
+[2026-08-27 05:45:27 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 18665, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8822.26
+[2026-08-27 05:45:28 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 10512, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10269.78
+[2026-08-27 05:45:28 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 10512, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9857.87
+[2026-08-27 05:45:29 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 2320, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8661.84
+[2026-08-27 05:45:29 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 2320, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9228.23
+[2026-08-27 05:45:29 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 2368, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6475.66
+[2026-08-27 05:45:30 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 2368, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4004.99
+[2026-08-27 05:45:31] INFO:     174.1.60.5:40632 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:45:34 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.06, #running-req: 0, #queue-req: 0, #pending-token: 8525, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1622.55
+[2026-08-27 05:45:35 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.06, #running-req: 0, #queue-req: 0, #pending-token: 8525, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1467.18
+[2026-08-27 05:45:36 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.04, #running-req: 0, #queue-req: 0, #pending-token: 333, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4802.48
+[2026-08-27 05:45:36 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.04, #running-req: 0, #queue-req: 0, #pending-token: 333, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8988.80
+[2026-08-27 05:45:36 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.01, mamba usage: 0.04, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3927.39
+[2026-08-27 05:45:36 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.01, mamba usage: 0.04, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9031.88
+[2026-08-27 05:45:41 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8556, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1568.18
+[2026-08-27 05:45:42 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8556, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1300.74
+[2026-08-27 05:45:43 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 364, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4309.98
+[2026-08-27 05:45:43 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 364, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8995.10
+[2026-08-27 05:45:43 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3577.86
+[2026-08-27 05:45:43 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 18968.23
+[2026-08-27 05:45:44] INFO:     174.1.60.5:35706 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:45:50 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8525, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1296.76
+[2026-08-27 05:45:51 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8525, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1105.60
+[2026-08-27 05:45:52 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 333, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4298.14
+[2026-08-27 05:45:52 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 333, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8818.76
+[2026-08-27 05:45:52 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2853.01
+[2026-08-27 05:45:52 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 20042.91
+[2026-08-27 05:46:01] INFO:     174.1.60.5:35758 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:46:02] INFO:     174.1.60.5:35774 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:46:04] INFO:     174.1.60.5:35784 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:46:07 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8525, #bootstrap-req: 2, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 548.95
+[2026-08-27 05:46:08 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8525, #bootstrap-req: 2, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 512.63
+[2026-08-27 05:46:09 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 333, #bootstrap-req: 2, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4297.52
+[2026-08-27 05:46:09 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 333, #bootstrap-req: 2, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9087.53
+[2026-08-27 05:46:09 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4230.34
+[2026-08-27 05:46:09 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9386.08
+[2026-08-27 05:46:14 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.06, #running-req: 0, #queue-req: 0, #pending-token: 8684, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1546.24
+[2026-08-27 05:46:15 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.06, #running-req: 0, #queue-req: 0, #pending-token: 8684, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1303.58
+[2026-08-27 05:46:16 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.04, #running-req: 0, #queue-req: 0, #pending-token: 492, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4427.06
+[2026-08-27 05:46:16 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.04, #running-req: 0, #queue-req: 0, #pending-token: 492, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8440.44
+[2026-08-27 05:46:17 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.04, #running-req: 0, #queue-req: 0, #pending-token: 9062, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10912.79
+[2026-08-27 05:46:17 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.04, #running-req: 0, #queue-req: 0, #pending-token: 9062, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9716.37
+[2026-08-27 05:46:18 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.04, #running-req: 0, #queue-req: 0, #pending-token: 870, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7876.07
+[2026-08-27 05:46:18 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.04, #running-req: 0, #queue-req: 0, #pending-token: 870, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9169.32
+[2026-08-27 05:46:18 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 896, #cached-token: 0, full token usage: 0.01, mamba usage: 0.04, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 5731.60
+[2026-08-27 05:46:18 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 896, #cached-token: 0, full token usage: 0.01, mamba usage: 0.04, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8837.14
+[2026-08-27 05:46:41] INFO:     174.1.60.5:43344 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:46:41] INFO:     174.1.60.5:43350 - "GET /server_info HTTP/1.1" 200 OK
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-c8-20260827-1340/logs/manual_p_node1.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-c8-20260827-1340/logs/manual_p_node1.log
new file mode 100644
index 0000000..c140674
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-c8-20260827-1340/logs/manual_p_node1.log
@@ -0,0 +1,1255 @@
+/usr/local/lib/python3.12/dist-packages/torchao/quantization/quant_api.py:1731: SyntaxWarning: invalid escape sequence '\.'
+  """Configuration class for applying different quantization configs to modules or parameters based on their fully qualified names (FQNs).
+/sgl-workspace/sglang/python/sglang/launch_server.py:56: UserWarning: 'python -m sglang.launch_server' is still supported, but 'sglang serve' is the recommended entrypoint.
+  Example: sglang serve --model-path <model> [options]
+  warnings.warn(
+[2026-08-27 04:28:43] Breakable CUDA graph is incompatible with MLA attention; disabling prefill CUDA graph.
+[2026-08-27 04:28:43] Attention backend not specified. Use triton backend by default.
+[2026-08-27 04:28:43] Pipeline parallelism is incompatible with overlap schedule.
+[2026-08-27 04:28:46] server_args=ServerArgs(model_path='/data/hf_models/Kimi-K3', tokenizer_path='/data/hf_models/Kimi-K3', tokenizer_mode='auto', tokenizer_backend='huggingface', tokenizer_worker_num=1, detokenizer_worker_num=1, skip_tokenizer_init=False, load_format='auto', model_loader_extra_config='{}', trust_remote_code=True, context_length=None, is_embedding=False, enable_multimodal=None, revision=None, model_impl='auto', model_config_parser='auto', json_model_override_args='{}', dtype='auto', quantization=None, quantization_param_path=None, kv_cache_dtype='auto', enable_fp32_lm_head=False, modelopt_quant=None, modelopt_checkpoint_restore_path=None, modelopt_checkpoint_save_path=None, modelopt_export_path=None, quantize_and_serve=False, rl_quant_profile=None, enable_tf32_matmul=False, mem_fraction_static=0.88, max_running_requests=None, max_queued_requests=None, max_total_tokens=None, chunked_prefill_size=8192, enable_dynamic_chunking=False, max_prefill_tokens=16384, prefill_max_requests=None, schedule_policy='fcfs', enable_priority_scheduling=False, disable_priority_preemption=False, default_priority_value=None, abort_on_priority_when_disabled=False, schedule_low_priority_values_first=False, priority_scheduling_preemption_threshold=10, retraction_policy='length', schedule_conservativeness=1.0, page_size=64, swa_full_tokens_ratio=0.8, disable_hybrid_swa_memory=False, radix_eviction_policy='lru', prefill_only_disable_kv_cache=False, disable_radix_cache=True, enable_page_major_kv_layout=False, enable_unified_memory=False, disable_chunked_prefix_cache=False, disable_overlap_schedule=True, num_continuous_decode_steps=1, scheduler_recv_interval=1, enable_mixed_chunk=False, nccl_port=None, dist_timeout=3600, dist_init_addr='174.1.60.1:20000', nnodes=4, node_rank=1, tp_size=4, dcp_size=1, pp_size=8, pp_max_micro_batch_size=None, pp_async_batch_depth=0, dp_size=1, load_balance_method='follow_bootstrap_room', attn_cp_size=1, moe_dp_size=1, dcp_comm_backend='ag_rs', dcp_replicate_q_proj=None, dwdp_size=1, enable_prefill_cp=False, cp_strategy=None, enable_dsa_cache_layer_split=False, enable_dsa_prefill_context_parallel=False, dsa_prefill_cp_mode='round-robin-split', enable_prefill_context_parallel=False, prefill_cp_mode='in-seq-split', enable_cp_decode_attn_tp=False, enable_dp_attention=False, enable_dp_attention_local_control_broadcast=False, enable_dp_lm_head=False, enable_attn_tp_input_scattered=False, disable_attn_tp_gather=False, enable_p2p_check=False, device='cuda', base_gpu_id=0, gpu_id_step=1, random_seed=577755302, watchdog_timeout=300, soft_watchdog_timeout=None, sleep_on_idle=False, use_ray=False, custom_sigquit_handler=None, numa_node=None, gc_threshold=None, host='0.0.0.0', port=30000, fastapi_root_path='', smg_grpc_mode=False, grpc_mode=False, grpc_port=None, sidecar=None, sidecar_args=None, skip_server_warmup=False, warmups=None, enable_http2=False, ssl_keyfile=None, ssl_certfile=None, ssl_ca_certs=None, ssl_keyfile_password=None, enable_ssl_refresh=False, api_key=None, admin_api_key=None, served_model_name='kimi-k3', weight_version='default', chat_template=None, hf_chat_template_name=None, completion_template=None, file_storage_path='sglang_storage', enable_cache_report=False, reasoning_parser=None, default_chat_template_kwargs=None, strip_thinking_cache=False, enable_strict_thinking=False, tool_call_parser=None, tool_server=None, sampling_defaults='model', asr_max_buffer_seconds=60, asr_max_concurrent_sessions=32, preferred_sampling_params=None, allow_auto_truncate=False, stream_interval=1, batch_notify_size=16, stream_response_default_include_usage=False, incremental_streaming_output=False, enable_streaming_session=False, enable_session_radix_cache=False, log_level='info', log_level_http=None, log_requests=False, log_requests_level=2, log_requests_format='text', log_requests_target=None, uvicorn_access_log_exclude_prefixes=[], crash_dump_folder=None, show_time_cost=False, enable_metrics=False, smg_http_sidecar_port=None, enable_mfu_metrics=False, enable_metrics_for_all_schedulers=False, load_snapshot_publish_interval=15, tokenizer_metrics_custom_labels_header='x-custom-labels', tokenizer_metrics_allowed_custom_labels=None, extra_metric_labels=None, bucket_time_to_first_token=None, bucket_inter_token_latency=None, bucket_e2e_request_latency=None, prompt_tokens_buckets=None, generation_tokens_buckets=None, gc_warning_threshold_secs=0.0, decode_log_interval=40, enable_request_time_stats_logging=False, kv_events_config=None, enable_forward_pass_metrics=False, forward_pass_metrics_worker_id='', forward_pass_metrics_ipc_name=None, enable_trace=False, trace_modules='request', otlp_traces_endpoint='localhost:4317', export_metrics_to_file=False, export_metrics_to_file_dir=None, stat_loggers=None, constrained_json_whitespace_pattern=None, constrained_json_disable_any_whitespace=False, attention_backend='triton', decode_attention_backend=None, prefill_attention_backend=None, sampling_backend='flashinfer', grammar_backend='xgrammar', radix_cache_backend=None, mm_attention_backend=None, fp8_gemm_runner_backend='auto', fp4_gemm_runner_backend='auto', bf16_gemm_backend='auto', dsa_prefill_backend=None, dsa_decode_backend=None, dsa_paged_mqa_logits_backend='auto', dsa_topk_backend='sgl-kernel', disable_flashinfer_autotune=False, flashinfer_autotune_skip_ops=None, mamba_backend='triton', cuda_graph_config=CudaGraphConfig(decode=PhaseConfig(backend='disabled', max_bs=16, bs=[1, 2, 4, 8, 12, 16], tc_compiler='eager', full_prefill_max_req=None), prefill=PhaseConfig(backend='disabled', max_bs=2048, bs=[4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], tc_compiler='eager', full_prefill_max_req=None)), cuda_graph_backend_decode=None, cuda_graph_backend_prefill=None, cuda_graph_max_bs_decode=16, cuda_graph_max_bs_prefill=None, cuda_graph_bs_decode=None, cuda_graph_bs_prefill=None, cuda_graph_tc_compiler=None, disable_prefill_cuda_graph=False, disable_decode_cuda_graph=False, disable_cuda_graph=False, disable_cuda_graph_padding=False, enable_profile_cuda_graph=False, enable_cudagraph_gc=False, debug_cuda_graph=False, enable_layerwise_nvtx_marker=False, enable_nccl_nvls=False, enable_symm_mem=False, triton_attention_reduce_in_fp32=False, triton_attention_num_kv_splits=8, triton_attention_split_tile_size=None, flashinfer_mla_disable_ragged=False, enable_fused_qk_norm_rope=False, enable_precise_embedding_interpolation=False, enable_fused_moe_sum_all_reduce=False, enable_deepseek_v4_fp4_indexer=False, disable_custom_all_reduce=False, enable_mscclpp=False, enable_torch_symm_mem=False, enable_scattered_sconv=False, pre_warm_nccl=False, enable_quant_communications=False, enable_flashinfer_allreduce_fusion=False, enforce_disable_flashinfer_allreduce_fusion=False, flashinfer_allreduce_fusion_backend=None, enable_aiter_allreduce_fusion=False, enable_torch_compile=False, enable_torch_compile_debug_mode=False, torch_compile_max_bs=32, torchao_config='', speculative_algorithm=None, speculative_draft_model_path=None, speculative_draft_model_revision=None, speculative_draft_load_format=None, speculative_num_steps=None, speculative_eagle_topk=None, speculative_num_draft_tokens=None, speculative_dflash_block_size=None, speculative_dspark_block_size=None, speculative_dspark_sps_table_path=None, speculative_dspark_confidence_sts_path=None, speculative_dspark_align_verify_tokens_to_graph_tier=False, speculative_accept_threshold_single=1.0, speculative_accept_threshold_acc=1.0, speculative_use_rejection_sampling=False, speculative_token_map=None, speculative_attention_mode='prefill', speculative_draft_attention_backend=None, speculative_draft_window_size=None, speculative_moe_runner_backend='flashinfer_mxfp4', speculative_moe_a2a_backend=None, speculative_draft_model_quantization=None, speculative_skip_dp_mlp_sync=False, enable_multi_layer_eagle=False, speculative_adaptive=False, speculative_adaptive_config=None, decoupled_spec_bind_endpoint=None, decoupled_spec_connect_endpoints=None, decoupled_spec_rank=None, decoupled_spec_role='null', spec_trace_dir=None, speculative_ngram_min_bfs_breadth=1, speculative_ngram_max_bfs_breadth=10, speculative_ngram_match_type='BFS', speculative_ngram_max_trie_depth=18, speculative_ngram_capacity=10000000, speculative_ngram_external_corpus_path=None, speculative_ngram_external_sam_budget=0, speculative_ngram_external_corpus_max_tokens=10000000, ep_size=4, moe_a2a_backend='none', moe_runner_backend='flashinfer_mxfp4', flashinfer_mxfp4_moe_precision='default', deepep_mode='auto', fuseep_mode=2, deepep_dispatcher_output_dtype='auto', ep_num_redundant_experts=0, ep_dispatch_algorithm=None, init_expert_location='trivial', enable_eplb=False, eplb_algorithm='auto', eplb_rebalance_num_iterations=1000, eplb_rebalance_layers_per_chunk=None, eplb_min_rebalancing_utilization_threshold=1.0, expert_distribution_recorder_mode=None, expert_distribution_recorder_buffer_size=1000, enable_expert_distribution_metrics=False, deepep_config=None, moe_dense_tp_size=None, elastic_ep_backend=None, enable_elastic_expert_backup=False, mooncake_ib_device=None, enable_waterfill=False, ep_join_mode=None, ep_join_rank_offset=0, elastic_ep_initial_size=None, max_ep_size=None, elastic_ep_scale_timeout=600, elastic_ep_rejoin=False, disable_flashinfer_cutlass_moe_fp4_allgather=False, disable_shared_experts_fusion=False, enforce_shared_experts_fusion=False, max_mamba_cache_size=None, mamba_ssm_dtype=None, mamba_max_states_per_path=-1, enable_mamba_cache_stochastic_rounding=False, mamba_cache_philox_rounds=0, mamba_full_memory_ratio=0.36, mamba_radix_cache_strategy='extra_buffer_lazy', uses_mamba_radix_cache=False, mamba_track_interval=256, enable_int8_mamba_checkpoint=False, int8_mamba_ckpt_size=None, linear_attn_backend='triton', linear_attn_decode_backend=None, linear_attn_prefill_backend=None, linear_attn_verify_backend=None, enable_linear_replayssm=False, linear_replayssm_cache_len=16, enable_linear_replayssm_spec=False, enable_hierarchical_cache=False, hicache_ratio=2.0, hicache_size=0, hicache_write_policy='write_through', hicache_io_backend='kernel', hicache_mem_layout='page_first', hicache_storage_backend=None, hicache_storage_prefetch_policy='timeout', hicache_storage_backend_extra_config=None, enable_hisparse=False, hisparse_config=None, enable_broadcast_mm_inputs_process=False, enable_prefix_mm_cache=False, mm_enable_dp_encoder=False, mm_process_config={}, mm_processor_worker_num=0, mm_io_worker_num=0, limit_mm_data_per_request=None, enable_mm_global_cache=False, disable_fast_image_processor=False, mm_feature_transport='cpu', keep_mm_feature_on_device=False, enable_lora=None, enable_lora_overlap_loading=None, max_lora_rank=None, lora_target_modules=None, lora_paths=None, max_loaded_loras=None, max_loras_per_batch=8, lora_eviction_policy='lru', lora_backend='csgmv', max_lora_chunk_size=16, experts_shared_outer_loras=None, lora_use_virtual_experts=False, lora_strict_loading=False, lora_drain_wait_threshold=0.0, enable_two_batch_overlap=False, enable_single_batch_overlap=False, tbo_token_distribution_threshold=0.48, cpu_offload_gb=0, offload_group_size=-1, offload_num_in_group=1, offload_prefetch_step=1, offload_mode='cpu', enable_lmcache=False, lmcache_config_file=None, enable_flexkv=False, flexkv_config_file=None, kt_weight_path=None, kt_method='AMXINT4', kt_cpuinfer=None, kt_threadpool_count=2, kt_num_gpu_experts=None, kt_max_deferred_experts_per_token=None, dllm_algorithm=None, dllm_algorithm_config=None, dllm_fdfo=True, disaggregation_mode='prefill', disaggregation_transfer_backend='mooncake', disaggregation_bootstrap_port=28800, disaggregation_ib_device='mlx5_0,mlx5_1,mlx5_2,mlx5_3', disaggregation_decode_enable_radix_cache=False, disaggregation_decode_enable_offload_kvcache=False, num_reserved_decode_tokens=512, disaggregation_decode_extra_slots=None, disaggregation_decode_polling_interval=1, optimistic_prefill_attempts=0, encoder_only=False, language_only=False, encoder_transfer_backend='zmq_to_tokenizer', encoder_urls=[], encoder_bootstrap_port=8997, encoder_register_urls=[], enable_adaptive_dispatch_to_encoder=False, enable_pdmux=False, pdmux_config_path=None, sm_group_num=8, custom_weight_loader=[], weight_loader_disable_mmap=False, weight_loader_prefetch_checkpoints=False, weight_loader_prefetch_num_threads=4, weight_loader_drop_cache_after_load=False, remote_instance_weight_loader_seed_instance_ip=None, remote_instance_weight_loader_seed_instance_service_port=None, remote_instance_weight_loader_send_weights_group_ports=None, remote_instance_weight_loader_backend='nccl', remote_instance_weight_loader_start_seed_via_transfer_engine=False, engine_info_bootstrap_port=6789, modelexpress_config=None, download_dir=None, model_checksum=None, delete_ckpt_after_loading=False, decrypted_config_file=None, decrypted_draft_config_file=None, checkpoint_engine_wait_weights_before_ready=False, enable_prefill_delayer=False, prefill_delayer_max_delay_passes=30, prefill_delayer_token_usage_low_watermark=None, prefill_delayer_forward_passes_buckets=None, prefill_delayer_wait_seconds_buckets=None, prefill_delayer_queue_min_ratio=None, prefill_delayer_max_delay_ms=None, min_free_slots_delay=None, enable_deterministic_inference=False, rl_on_policy_target=None, kv_canary='none', kv_canary_real_data='none', kv_canary_sweep_interval=0, enable_dynamic_batch_tokenizer=False, dynamic_batch_tokenizer_batch_size=32, dynamic_batch_tokenizer_batch_timeout=0.002, enable_tokenizer_batch_encode=False, disable_tokenizer_batch_decode=False, debug_tensor_dump_output_folder=None, debug_tensor_dump_layers=None, debug_tensor_dump_input_file=None, enable_memory_saver=False, enable_weights_cpu_backup=False, enable_draft_weights_cpu_backup=False, enable_custom_logit_processor=False, enable_return_hidden_states=False, enable_return_routed_experts=False, enable_return_indexer_topk=False, disable_outlines_disk_cache=False, enable_mis=False, weight_cache_mode='off', weight_cache_socket=None, weight_cache_timeout=1800, forward_hooks=None, msprobe_dump_config=None)
+[2026-08-27 04:28:46] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:46] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:46] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:46] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:46] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:46] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:46] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:46] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:02 PP3 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:02 PP3 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:03 PP3 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:03 PP3 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:03 PP3 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:03 PP3 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:03 PP2 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:03 PP2 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:03 PP2 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:03 PP2 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:04 PP2 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:04 PP2 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:04 PP2 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:04 PP2 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:04 PP3 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:04 PP3 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:05 PP3 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:05 PP2 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:05 PP3 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:05 PP3 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:05 PP2 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:06.044592   241 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:06.044694   241 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:06.044710   241 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.2 port: 12001
+I0827 04:29:06.044759   241 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.2:16343
+I0827 04:29:06.044883   241 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:06.053642   242 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:06.053738   242 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:06.053754   242 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.2 port: 12001
+I0827 04:29:06.053797   242 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.2:15170
+I0827 04:29:06.053925   242 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:06.058039   237 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:06.058135   237 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:06.058149   237 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.2 port: 12001
+I0827 04:29:06.058192   237 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.2:15486
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:06.058215   236 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:06.058310   236 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:06.058316   237 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:06.058326   236 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.2 port: 12001
+I0827 04:29:06.058367   236 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.2:16140
+I0827 04:29:06.058492   236 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:06.059734   239 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:06.059828   239 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:06.059844   239 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.2 port: 12001
+I0827 04:29:06.059890   239 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.2:15755
+I0827 04:29:06.060027   239 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:06.064433   237 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:06.064513   236 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:06.065588   239 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:06.069945   237 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:06.070852   236 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:06.071113   239 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:06.071978   241 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:06.079348   236 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:06.079967   242 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:06.080070   241 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:06.087044   241 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:06.087074   242 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:06.094024   242 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:06.095021   241 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:06.095822   237 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:06.097810   241 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:06.097839   241 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:06.097889   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.099939   239 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:06.100984   242 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:06.103683   242 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:06.103708   242 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:06.103756   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.103785   236 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:06.103833   237 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:06.106014   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:06.106982   236 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:06.107007   236 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:06.107053   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.107446   241 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:02
+I0827 04:29:06.107475   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.107936   239 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:06.107971   237 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:06.107997   237 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:06.108048   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.110760   239 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:06.110790   239 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:06.110847   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.111824   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:06.113214   242 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:02
+I0827 04:29:06.113242   242 rdma_context.cpp:205] Using SIEVE endpoint store
+[2026-08-27 04:29:06 PP2 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+I0827 04:29:06.114821   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:06.115196   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+[2026-08-27 04:29:06 PP2 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+I0827 04:29:06.116189   241 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:02
+I0827 04:29:06.116214   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.116621   237 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:02
+I0827 04:29:06.116650   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.118325   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:06.119874   239 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:02
+I0827 04:29:06.119906   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.121601   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:06.122952   242 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:02
+I0827 04:29:06.122977   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.127393   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:06.128841   239 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:02
+I0827 04:29:06.128867   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.132040   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:06.133428   236 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:02
+I0827 04:29:06.133458   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.136312   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:06.137817   239 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:02
+I0827 04:29:06.137846   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.144285   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:06.144373   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:06.145742   237 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:02
+I0827 04:29:06.145776   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.145833   241 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:02
+I0827 04:29:06.145866   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.152084   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:06.153820   242 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:02
+I0827 04:29:06.153860   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.154352   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:06.155728   241 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:02
+I0827 04:29:06.155835   241 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:06 PP3 TP2 EP2] Init torch distributed begin.
+I0827 04:29:06.163965   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:06.164443   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:06.164605   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:06.165319   236 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:02
+I0827 04:29:06.165345   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.165951   239 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:02
+I0827 04:29:06.166070   239 transfer_engine_impl.cpp:388] installTransport, type=rdma
+I0827 04:29:06.166186   242 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:02
+I0827 04:29:06.166311   242 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:06 PP3 TP0 EP0] Init torch distributed begin.
+[2026-08-27 04:29:06 PP3 TP3 EP3] Init torch distributed begin.
+I0827 04:29:06.172065   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:06.173699   237 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:02
+I0827 04:29:06.173743   237 rdma_context.cpp:205] Using SIEVE endpoint store
+[2026-08-27 04:29:06 PP3 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+I0827 04:29:06.192251   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:06.193763   236 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:02
+I0827 04:29:06.193797   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.200259   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:06.201810   237 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:02
+I0827 04:29:06.201937   237 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:06 PP2 TP2 EP2] Init torch distributed begin.
+I0827 04:29:06.220070   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:06.221505   236 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:02
+I0827 04:29:06.221653   236 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:06 PP2 TP1 EP1] Init torch distributed begin.
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:06.613179   235 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:06.613286   235 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:06.613301   235 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.2 port: 12001
+I0827 04:29:06.613344   235 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.2:16382
+I0827 04:29:06.613492   235 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:06.618471   235 topology.cpp:127] Device mlx5_0 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:06.618719   238 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:06.618824   238 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:06.618839   238 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.2 port: 12001
+I0827 04:29:06.618882   238 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.2:16170
+I0827 04:29:06.619020   238 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:06.624203   238 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:06.629066   238 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:06.633975   238 topology.cpp:127] Device mlx5_2 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:06.636366   240 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:06.636454   240 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:06.636466   240 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.2 port: 12001
+I0827 04:29:06.636509   240 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.2:15840
+I0827 04:29:06.636675   240 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:06.638876   238 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:06.640780   238 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:06.640802   238 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:06.640842   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.641396   240 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:06.643196   235 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:06.645701   240 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:06.649896   235 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:06.650310   240 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:06.654819   240 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:06.656375   240 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:06.656396   240 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:06.656432   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.657156   235 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:06.659863   235 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:06.659895   235 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:06.659950   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.661671   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:06.662956   240 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:02
+I0827 04:29:06.662979   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.664074   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:06.665762   238 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:02
+I0827 04:29:06.665796   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.667351   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:06.668808   235 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:02
+I0827 04:29:06.668838   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.676294   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:06.677721   235 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:02
+I0827 04:29:06.677748   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.692129   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:06.693621   240 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:02
+I0827 04:29:06.693648   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.696089   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:06.697444   238 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:02
+I0827 04:29:06.697470   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.707720   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:06.709362   235 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:02
+I0827 04:29:06.709394   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.719987   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:06.721463   240 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:02
+I0827 04:29:06.721489   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.728108   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:06.728855   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:06.729467   238 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:02
+I0827 04:29:06.729491   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.730392   240 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:02
+I0827 04:29:06.730533   240 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:06 PP3 TP1 EP1] Init torch distributed begin.
+I0827 04:29:06.736888   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:06.738317   238 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:02
+I0827 04:29:06.738422   238 transfer_engine_impl.cpp:388] installTransport, type=rdma
+I0827 04:29:06.738466   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+[2026-08-27 04:29:06 PP2 TP3 EP3] Init torch distributed begin.
+I0827 04:29:06.739869   235 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:02
+I0827 04:29:06.739982   235 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:06 PP2 TP0 EP0] Init torch distributed begin.
+[2026-08-27 04:29:26 PP3 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:26 PP2 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:27 PP2 TP3 EP3] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP2 TP1 EP1] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP2 TP2 EP2] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP2 TP0 EP0] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP3 TP3 EP3] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP3 TP0 EP0] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP3 TP2 EP2] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP3 TP1 EP1] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP3 TP3 EP3] Init torch distributed ends. elapsed=21.61 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP3 TP2 EP2] Init torch distributed ends. elapsed=21.62 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP3 TP1 EP1] Init torch distributed ends. elapsed=21.04 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP3 TP0 EP0] Init torch distributed ends. elapsed=21.61 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP2 TP2 EP2] Init torch distributed ends. elapsed=21.57 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP2 TP1 EP1] Init torch distributed ends. elapsed=21.55 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP2 TP0 EP0] Init torch distributed ends. elapsed=21.03 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP2 TP3 EP3] Init torch distributed ends. elapsed=21.04 s, mem usage=0.54 GB
+[2026-08-27 04:29:28 PP3 TP3 EP3] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP3 TP0 EP0] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP2 TP0 EP0] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP3 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP2 TP1 EP1] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP3 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP2 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP3 TP2 EP2] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP2 TP3 EP3] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP2 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP2 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP3 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP2 TP2 EP2] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP3 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP3 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP2 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP3 TP1 EP1] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP2 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP3 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP2 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP3 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP3 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP2 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP3 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP2 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP3 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP2 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP2 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP2 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP2 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP3 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP3 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:30:31 PP2 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:31 PP2 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:31 PP2 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:31 PP2 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:31 PP2 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:31 PP2 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:31 PP2 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:31 PP2 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP2 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:32 PP2 TP0 EP0] Load weight end. elapsed=64.92 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=36.14 GB, mem usage=45.91 GB.
+[2026-08-27 04:30:33 PP3 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:33 PP3 TP1 EP1] Load weight end. elapsed=65.31 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:34 PP2 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:34 PP2 TP1 EP1] Load weight end. elapsed=66.07 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=36.14 GB, mem usage=45.91 GB.
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:35 PP3 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:35 PP3 TP0 EP0] Load weight end. elapsed=67.41 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:36 PP2 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:36 PP2 TP3 EP3] Load weight end. elapsed=68.04 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=36.14 GB, mem usage=45.91 GB.
+[2026-08-27 04:30:36 PP2 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:36 PP2 TP2 EP2] Load weight end. elapsed=68.23 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=36.14 GB, mem usage=45.91 GB.
+[2026-08-27 04:30:36 PP2 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:36 PP2 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:36 PP2 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:36 PP2 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:36 PP2 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:36 PP2 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:36 PP2 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:36 PP2 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:36 PP3 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:36 PP3 TP3 EP3] Load weight end. elapsed=68.86 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:37 PP3 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:37 PP3 TP2 EP2] Load weight end. elapsed=69.68 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:38 PP3 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:38 PP3 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:38 PP3 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:38 PP3 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:38 PP3 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:38 PP3 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:38 PP3 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:38 PP3 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:54 PP3 TP3 EP3] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP3 TP2 EP2] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP3 TP1 EP1] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP2 TP3 EP3] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP3 TP0 EP0] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP2 TP2 EP2] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP2 TP1 EP1] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP2 TP0 EP0] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP2 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:54 PP2 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:54 PP3 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP3 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP2 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:54 PP3 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP3 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP2 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:54 PP3 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP3 TP3 EP3] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP2 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP2 TP2 EP2] Memory pool end. avail mem=23.25 GB
+[2026-08-27 04:30:54 PP2 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP2 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP2 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP3 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP3 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP2 TP3 EP3] Memory pool end. avail mem=23.25 GB
+[2026-08-27 04:30:54 PP2 TP1 EP1] Memory pool end. avail mem=23.25 GB
+[2026-08-27 04:30:54 PP3 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP3 TP1 EP1] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP3 TP2 EP2] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP2 TP0 EP0] Memory pool end. avail mem=23.25 GB
+[2026-08-27 04:30:54 PP3 TP0 EP0] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP3 TP3 EP3] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp3_pp3_dp0.json
+[2026-08-27 04:30:54 PP2 TP1 EP1] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp1_pp2_dp0.json
+[2026-08-27 04:30:54 PP2 TP2 EP2] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp2_pp2_dp0.json
+[2026-08-27 04:30:54 PP3 TP1 EP1] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp1_pp3_dp0.json
+[2026-08-27 04:30:54 PP2 TP3 EP3] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp3_pp2_dp0.json
+[2026-08-27 04:30:54 PP3 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:30:54 PP3 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:30:54 PP3 TP2 EP2] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp2_pp3_dp0.json
+[2026-08-27 04:30:54 PP2 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:30:54 PP2 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO[TensorRT-LLM][INFO] Set logger level to INFO
+
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[2026-08-27 04:30:54 PP3 TP0 EP0] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp0_pp3_dp0.json
+[2026-08-27 04:30:54 PP2 TP0 EP0] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp0_pp2_dp0.json
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.65profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.60profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.79profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.51profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.86profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.27profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.62profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.98profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.86profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 12.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 13.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.93profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.92profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.09profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.63profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  6.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  7.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.33profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.58profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  8.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.56profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.23profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.63profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.36profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.09profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.97profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  8.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.30profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 13.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 13.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  8.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  8.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 11.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 11.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.23profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.48profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.23profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.53profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.27profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.59profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.62profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.97profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.60profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.46profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.12profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.68profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.77profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.62profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  6.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.87profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.30profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.40profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.83profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.94profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.59profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.66profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.29profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.23profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:01,  3.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.68profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.75profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  7.01profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.22profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.65profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.62profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.46profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.22profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  7.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:01<00:06,  1.03s/profile]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:01,  3.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.50profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.93profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.48profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.11profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.12profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.30profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.68profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  7.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 11.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  9.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 12.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 11.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 11.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 11.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.11profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.31profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.65profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.31profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.35profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.49profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:01<00:01,  2.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.65profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.68profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  7.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.09profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 11.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.88profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.98profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.50profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.62profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.51profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.91profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:01,  3.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.31profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.86profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.27profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.59profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.17profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.76profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.73profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  9.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.37profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.15profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.98profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.65profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.97profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.23profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.46profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  9.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.03profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.57profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.40profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.40profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.57profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.13profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.56profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  6.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.73profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.23profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  3.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.86profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.77profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.92profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.37profile/s]
+[2026-08-27 04:31:26 PP2 TP3 EP3] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.39profile/s]
+[2026-08-27 04:31:26 PP2 TP2 EP2] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.75profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.46profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.35profile/s]
+[2026-08-27 04:31:26 PP2 TP1 EP1] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.12profile/s]
+[2026-08-27 04:31:26 PP2 TP0 EP0] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.66profile/s][2026-08-27 04:31:26 PP2 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:26 PP2 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:26 PP2 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:26 PP2 TP0 EP0] max_total_num_tokens=3801408, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=54, context_len=1048576, available_gpu_mem=20.47 GB
+[2026-08-27 04:31:26 PP2 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:26 PP2 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:26 PP2 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:26 PP2 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:26 PP2 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.93profile/s]
+[2026-08-27 04:31:27 PP3 TP1 EP1] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.16profile/s]
+[2026-08-27 04:31:27 PP3 TP0 EP0] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.43profile/s]
+[2026-08-27 04:31:27 PP3 TP2 EP2] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.05profile/s]
+[2026-08-27 04:31:27 PP3 TP3 EP3] FlashInfer autotune completed.
+[2026-08-27 04:31:27 PP3 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:27 PP3 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:27 PP3 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:27 PP3 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:27 PP3 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:27 PP3 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:27 PP3 TP0 EP0] max_total_num_tokens=3801408, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=54, context_len=1048576, available_gpu_mem=16.54 GB
+[2026-08-27 04:31:27 PP3 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:27 PP3 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+W0827 04:31:30.590376  2521 memory_location.cpp:73] Failed to get NUMA node, addr: 0x513c5ec0, len: 6912: Operation not permitted [1]
+W0827 04:31:30.590432  2522 memory_location.cpp:73] Failed to get NUMA node, addr: 0x514a0900, len: 6912: Operation not permitted [1]
+W0827 04:31:30.590495  2525 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51226dc0, len: 55296: Operation not permitted [1]
+W0827 04:31:30.590492  2523 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5128db40, len: 6912: Operation not permitted [1]
+W0827 04:31:30.590533  2524 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5156ff80, len: 6912: Operation not permitted [1]
+W0827 04:31:30.590600  2526 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51320ac0, len: 55296: Operation not permitted [1]
+W0827 04:31:30.590636  2527 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51768540, len: 6912: Operation not permitted [1]
+W0827 04:31:30.590826  2528 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51483dc0, len: 13824: Operation not permitted [1]
+W0827 04:31:30.590914  2529 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51119380, len: 6912: Operation not permitted [1]
+W0827 04:31:30.590979  2530 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50fe1e00, len: 6912: Operation not permitted [1]
+W0827 04:31:30.846120  2547 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50ceebc0, len: 6912: Operation not permitted [1]
+W0827 04:31:30.846271  2548 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50c57640, len: 6912: Operation not permitted [1]
+W0827 04:31:30.846328  2549 memory_location.cpp:73] Failed to get NUMA node, addr: 0x502ed680, len: 6912: Operation not permitted [1]
+W0827 04:31:30.846354  2550 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50d1c800, len: 6912: Operation not permitted [1]
+W0827 04:31:30.846448  2551 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50510a40, len: 55296: Operation not permitted [1]
+W0827 04:31:30.846473  2552 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5051e280, len: 55296: Operation not permitted [1]
+W0827 04:31:30.846571  2553 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50d01340, len: 6912: Operation not permitted [1]
+W0827 04:31:30.846689  2554 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50ce3440, len: 13824: Operation not permitted [1]
+W0827 04:31:30.846858  2555 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5030f440, len: 6912: Operation not permitted [1]
+W0827 04:31:30.846926  2556 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50cae980, len: 6912: Operation not permitted [1]
+W0827 04:31:31.198441  2575 memory_location.cpp:73] Failed to get NUMA node, addr: 0x514f46c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.198501  2576 memory_location.cpp:73] Failed to get NUMA node, addr: 0x509c6400, len: 6912: Operation not permitted [1]
+W0827 04:31:31.198501  2577 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51513300, len: 6912: Operation not permitted [1]
+W0827 04:31:31.198679  2578 memory_location.cpp:73] Failed to get NUMA node, addr: 0x506eef40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.198787  2579 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50dc9180, len: 55296: Operation not permitted [1]
+W0827 04:31:31.198915  2580 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50904700, len: 55296: Operation not permitted [1]
+W0827 04:31:31.198982  2581 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50aa9440, len: 6912: Operation not permitted [1]
+W0827 04:31:31.199081  2582 memory_location.cpp:73] Failed to get NUMA node, addr: 0x515012c0, len: 13824: Operation not permitted [1]
+W0827 04:31:31.199148  2583 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5161ac80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.199185  2584 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50dd69c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.763885  2603 memory_location.cpp:73] Failed to get NUMA node, addr: 0x519f1500, len: 6912: Operation not permitted [1]
+W0827 04:31:31.763998  2605 memory_location.cpp:73] Failed to get NUMA node, addr: 0x517df940, len: 6912: Operation not permitted [1]
+W0827 04:31:31.763923  2604 memory_location.cpp:73] Failed to get NUMA node, addr: 0x514ab440, len: 6912: Operation not permitted [1]
+W0827 04:31:31.764139  2606 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51556180, len: 6912: Operation not permitted [1]
+W0827 04:31:31.764169  2608 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50ed8140, len: 55296: Operation not permitted [1]
+W0827 04:31:31.764184  2607 memory_location.cpp:73] Failed to get NUMA node, addr: 0x515df040, len: 55296: Operation not permitted [1]
+W0827 04:31:31.764356  2609 memory_location.cpp:73] Failed to get NUMA node, addr: 0x517e3580, len: 6912: Operation not permitted [1]
+W0827 04:31:31.764564  2610 memory_location.cpp:73] Failed to get NUMA node, addr: 0x519cdb80, len: 13824: Operation not permitted [1]
+W0827 04:31:31.764624  2611 memory_location.cpp:73] Failed to get NUMA node, addr: 0x519dbf80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.764714  2612 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50f08d80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.868036  2629 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50a7a680, len: 6912: Operation not permitted [1]
+W0827 04:31:31.868046  2630 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50b6e400, len: 6912: Operation not permitted [1]
+W0827 04:31:31.868139  2631 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50e5f580, len: 6912: Operation not permitted [1]
+W0827 04:31:31.868261  2632 memory_location.cpp:73] Failed to get NUMA node, addr: 0x503c7100, len: 6912: Operation not permitted [1]
+W0827 04:31:31.868330  2633 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50c94ac0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.868432  2635 memory_location.cpp:73] Failed to get NUMA node, addr: 0x509b2880, len: 6912: Operation not permitted [1]
+W0827 04:31:31.868502  2634 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50b14c80, len: 55296: Operation not permitted [1]
+W0827 04:31:31.868564  2636 memory_location.cpp:73] Failed to get NUMA node, addr: 0x503ee840, len: 13824: Operation not permitted [1]
+W0827 04:31:31.868708  2637 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51057d80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.868835  2638 memory_location.cpp:73] Failed to get NUMA node, addr: 0x509cd940, len: 6912: Operation not permitted [1]
+W0827 04:31:32.003983  2656 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50715b00, len: 6912: Operation not permitted [1]
+W0827 04:31:32.004096  2657 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51346e80, len: 6912: Operation not permitted [1]
+W0827 04:31:32.004005  2655 memory_location.cpp:73] Failed to get NUMA node, addr: 0x507176c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.004179  2658 memory_location.cpp:73] Failed to get NUMA node, addr: 0x507e3a40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.004302  2659 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5078aa80, len: 55296: Operation not permitted [1]
+W0827 04:31:32.004357  2660 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5135c980, len: 55296: Operation not permitted [1]
+W0827 04:31:32.004410  2661 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5099f180, len: 6912: Operation not permitted [1]
+W0827 04:31:32.004585  2662 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50b287c0, len: 13824: Operation not permitted [1]
+W0827 04:31:32.004739  2663 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5070b6c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.004812  2664 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50753540, len: 6912: Operation not permitted [1]
+W0827 04:31:32.144867  2688 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50aed700, len: 6912: Operation not permitted [1]
+W0827 04:31:32.144946  2689 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50792740, len: 6912: Operation not permitted [1]
+W0827 04:31:32.145030  2690 memory_location.cpp:73] Failed to get NUMA node, addr: 0x508fca40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.145135  2691 memory_location.cpp:73] Failed to get NUMA node, addr: 0x505d7c40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.145318  2693 memory_location.cpp:73] Failed to get NUMA node, addr: 0x505ec340, len: 55296: Operation not permitted [1]
+W0827 04:31:32.145387  2692 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50602cc0, len: 55296: Operation not permitted [1]
+W0827 04:31:32.145479  2694 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50748c00, len: 6912: Operation not permitted [1]
+W0827 04:31:32.145632  2695 memory_location.cpp:73] Failed to get NUMA node, addr: 0x507e3480, len: 13824: Operation not permitted [1]
+W0827 04:31:32.145767  2696 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50316400, len: 6912: Operation not permitted [1]
+W0827 04:31:32.145864  2697 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50788600, len: 6912: Operation not permitted [1]
+W0827 04:31:32.177534  2726 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fff3c80, len: 6912: Operation not permitted [1]
+W0827 04:31:32.177623  2728 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fad6780, len: 6912: Operation not permitted [1]
+W0827 04:31:32.177625  2727 memory_location.cpp:73] Failed to get NUMA node, addr: 0x501dc2c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.178279  2729 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50596480, len: 6912: Operation not permitted [1]
+W0827 04:31:32.178463  2730 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50222dc0, len: 55296: Operation not permitted [1]
+W0827 04:31:32.178498  2731 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5038f6c0, len: 55296: Operation not permitted [1]
+W0827 04:31:32.178581  2732 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4ffd7ec0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.178695  2733 memory_location.cpp:73] Failed to get NUMA node, addr: 0x506d84c0, len: 13824: Operation not permitted [1]
+W0827 04:31:32.178884  2734 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5078efc0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.178922  2735 memory_location.cpp:73] Failed to get NUMA node, addr: 0x502c7b00, len: 6912: Operation not permitted [1]
+[2026-08-27 04:31:32] Dummy health check server started in background thread at 0.0.0.0:30000
+[rank10]:[W827 04:32:32.605391940 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank11]:[W827 04:32:32.605815131 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank9]:[W827 04:32:32.606041691 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank8]:[W827 04:32:32.608313727 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank11]:[W827 04:33:19.698792619 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank8]:[W827 04:33:19.699392665 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank9]:[W827 04:33:19.699501937 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank10]:[W827 04:33:19.699589651 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank14]:[W827 04:33:19.728965278 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank12]:[W827 04:33:19.729026815 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank13]:[W827 04:33:19.729092243 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank15]:[W827 04:33:19.730463091 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank14]:[W827 04:33:25.763484084 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank13]:[W827 04:33:25.764518868 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank15]:[W827 04:33:25.764776298 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank12]:[W827 04:33:25.765271843 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[2026-08-27 04:35:10 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.29
+[2026-08-27 04:35:10 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.29
+[2026-08-27 04:35:29 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.39
+[2026-08-27 04:35:29 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.39
+[2026-08-27 04:35:31 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25.86
+[2026-08-27 04:35:31 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25.87
+[2026-08-27 04:43:23 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.14
+[2026-08-27 04:43:23 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.14
+[2026-08-27 04:48:13 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.22
+[2026-08-27 04:48:13 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.22
+[2026-08-27 04:48:14 PP2 TP1 EP1] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP2 TP2 EP2] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP2 TP0 EP0] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP3 TP2 EP2] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP3 TP0 EP0] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP3 TP1 EP1] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP2 TP3 EP3] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP3 TP3 EP3] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:57:14 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.12
+[2026-08-27 04:57:14 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.12
+[2026-08-27 05:04:21 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1088, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2.55
+[2026-08-27 05:04:21 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1088, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2.55
+[2026-08-27 05:04:41 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.19
+[2026-08-27 05:04:41 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.19
+[2026-08-27 05:05:39 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1.10
+[2026-08-27 05:05:39 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1.10
+[2026-08-27 05:07:32 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 72.23
+[2026-08-27 05:07:33 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 72.17
+[2026-08-27 05:07:33 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 78417.87
+[2026-08-27 05:07:33 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 501194.56
+[2026-08-27 05:07:33 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 30385.21
+[2026-08-27 05:07:33 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32725.49
+[2026-08-27 05:07:40 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8673, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1097.27
+[2026-08-27 05:07:40 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8673, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1082.51
+[2026-08-27 05:07:40 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 481, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 68995.34
+[2026-08-27 05:07:40 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 481, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 494490.91
+[2026-08-27 05:07:40 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33832.33
+[2026-08-27 05:07:40 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33008.60
+[2026-08-27 05:07:48 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8656, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1098.69
+[2026-08-27 05:07:48 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8656, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1082.21
+[2026-08-27 05:07:48 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 464, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 62757.93
+[2026-08-27 05:07:48 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 464, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 451735.55
+[2026-08-27 05:07:48 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 30632.65
+[2026-08-27 05:07:48 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 31628.03
+[2026-08-27 05:07:55 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8663, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1101.49
+[2026-08-27 05:07:55 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8663, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1087.16
+[2026-08-27 05:07:55 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 471, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 71370.09
+[2026-08-27 05:07:55 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 471, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 487386.35
+[2026-08-27 05:07:55 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 31311.94
+[2026-08-27 05:07:55 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 31458.64
+[2026-08-27 05:08:03 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8659, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1099.50
+[2026-08-27 05:08:03 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8659, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1082.83
+[2026-08-27 05:08:03 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 467, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 61030.58
+[2026-08-27 05:08:03 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 467, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 440213.50
+[2026-08-27 05:08:03 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33725.14
+[2026-08-27 05:08:03 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32738.01
+[2026-08-27 05:08:10 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8692, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1100.46
+[2026-08-27 05:08:11 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8692, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1085.28
+[2026-08-27 05:08:11 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 500, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 67432.08
+[2026-08-27 05:08:11 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 500, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 480206.29
+[2026-08-27 05:08:11 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32427.60
+[2026-08-27 05:08:11 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 30441.07
+[2026-08-27 05:08:18 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8557, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1101.25
+[2026-08-27 05:08:18 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8557, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1089.66
+[2026-08-27 05:08:18 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 365, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 85531.52
+[2026-08-27 05:08:18 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 365, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 467687.86
+[2026-08-27 05:08:18 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 22103.80
+[2026-08-27 05:08:18 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25225.78
+[2026-08-27 05:08:26 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8683, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1099.06
+[2026-08-27 05:08:26 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8683, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1084.12
+[2026-08-27 05:08:26 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 491, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 68454.69
+[2026-08-27 05:08:26 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 491, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 485556.59
+[2026-08-27 05:08:26 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33944.36
+[2026-08-27 05:08:26 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 30281.29
+[2026-08-27 05:11:45 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.32
+[2026-08-27 05:11:45 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.32
+[2026-08-27 05:14:42 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1088, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6.16
+[2026-08-27 05:14:42 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1088, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6.16
+[2026-08-27 05:17:47 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 44.24
+[2026-08-27 05:17:47 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 44.21
+[2026-08-27 05:17:47 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 74322.16
+[2026-08-27 05:17:47 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 505572.40
+[2026-08-27 05:17:47 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 29824.24
+[2026-08-27 05:17:47 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32304.11
+[2026-08-27 05:17:55 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8673, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1100.05
+[2026-08-27 05:17:55 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8673, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1085.77
+[2026-08-27 05:17:55 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 481, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 72167.09
+[2026-08-27 05:17:55 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 481, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 477186.86
+[2026-08-27 05:17:55 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 30671.49
+[2026-08-27 05:17:55 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 34433.71
+[2026-08-27 05:18:02 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8656, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1101.74
+[2026-08-27 05:18:02 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8656, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1087.80
+[2026-08-27 05:18:02 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 464, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 74681.51
+[2026-08-27 05:18:02 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 464, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 550093.83
+[2026-08-27 05:18:02 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33172.26
+[2026-08-27 05:18:02 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32409.88
+[2026-08-27 05:18:10 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8663, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1101.73
+[2026-08-27 05:18:10 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8663, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1087.58
+[2026-08-27 05:18:10 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 471, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 73073.93
+[2026-08-27 05:18:10 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 471, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 528029.69
+[2026-08-27 05:18:10 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32561.99
+[2026-08-27 05:18:10 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 31690.72
+[2026-08-27 05:18:17 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8659, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1099.69
+[2026-08-27 05:18:17 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8659, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1085.59
+[2026-08-27 05:18:17 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 467, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 72775.07
+[2026-08-27 05:18:17 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 467, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 520101.00
+[2026-08-27 05:18:17 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 30837.63
+[2026-08-27 05:18:18 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32670.85
+[2026-08-27 05:18:25 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8692, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1100.48
+[2026-08-27 05:18:25 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8692, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1086.22
+[2026-08-27 05:18:25 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 500, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 72864.89
+[2026-08-27 05:18:25 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 500, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 524671.87
+[2026-08-27 05:18:25 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 31161.52
+[2026-08-27 05:18:25 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 35156.96
+[2026-08-27 05:18:32 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8557, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1102.45
+[2026-08-27 05:18:33 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8557, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1090.72
+[2026-08-27 05:18:33 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 365, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 85680.60
+[2026-08-27 05:18:33 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 365, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 482538.50
+[2026-08-27 05:18:33 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 24798.12
+[2026-08-27 05:18:33 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 23973.27
+[2026-08-27 05:18:40 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8683, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1102.03
+[2026-08-27 05:18:40 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8683, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1087.43
+[2026-08-27 05:18:40 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 491, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 70828.52
+[2026-08-27 05:18:40 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 491, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 529856.70
+[2026-08-27 05:18:40 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32426.64
+[2026-08-27 05:18:40 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33626.30
+[2026-08-27 05:21:44 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 2, #pending-token: 8652, #bootstrap-req: 2, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 44.47
+[2026-08-27 05:21:45 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 3, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9902.45
+[2026-08-27 05:21:45 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 2, #pending-token: 8652, #bootstrap-req: 2, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 44.27
+[2026-08-27 05:21:46 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 3, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8141.53
+[2026-08-27 05:21:46 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 3, #pending-token: 9185, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7907.49
+[2026-08-27 05:21:47 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 3, #pending-token: 9185, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10345.23
+[2026-08-27 05:21:47 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 2, #pending-token: 993, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9860.59
+[2026-08-27 05:21:48 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 2, #pending-token: 993, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8351.60
+[2026-08-27 05:21:48 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 2, #pending-token: 9680, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8645.66
+[2026-08-27 05:21:49 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 2, #pending-token: 9680, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10382.09
+[2026-08-27 05:21:49 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1488, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10477.17
+[2026-08-27 05:21:50 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1488, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8583.82
+[2026-08-27 05:21:50 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 10199, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8562.77
+[2026-08-27 05:21:51 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 18858, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9978.66
+[2026-08-27 05:21:51 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 10199, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9847.23
+[2026-08-27 05:21:51 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 18858, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9275.72
+[2026-08-27 05:21:51 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 27591, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9156.16
+[2026-08-27 05:21:52 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 53023, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10350.08
+[2026-08-27 05:21:52 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 27591, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9601.22
+[2026-08-27 05:21:53 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 44876, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9145.49
+[2026-08-27 05:21:53 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 53023, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9622.28
+[2026-08-27 05:21:54 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 36684, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9799.45
+[2026-08-27 05:21:54 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 44876, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9677.55
+[2026-08-27 05:21:55 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 28504, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10333.04
+[2026-08-27 05:21:55 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 36684, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9394.13
+[2026-08-27 05:21:56 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 28504, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10860.67
+[2026-08-27 05:21:56 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 20312, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9527.15
+[2026-08-27 05:21:57 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 20312, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9357.72
+[2026-08-27 05:21:57 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 12139, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9361.75
+[2026-08-27 05:21:57 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 12139, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 15465.96
+[2026-08-27 05:21:57 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 3947, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 15452.41
+[2026-08-27 05:21:57 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 3947, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 310446.55
+[2026-08-27 05:21:57 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 3968, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 156702.42
+[2026-08-27 05:21:57 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 3968, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 139513.02
+[2026-08-27 05:22:04 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 0, #pending-token: 8608, #bootstrap-req: 4, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1132.42
+[2026-08-27 05:22:05 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 416, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10805.74
+[2026-08-27 05:22:05 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 0, #pending-token: 8608, #bootstrap-req: 4, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1019.78
+[2026-08-27 05:22:06 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 416, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7895.05
+[2026-08-27 05:22:06 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 9149, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7392.79
+[2026-08-27 05:22:07 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 957, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11340.60
+[2026-08-27 05:22:07 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 9149, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11099.21
+[2026-08-27 05:22:08 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 957, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8063.56
+[2026-08-27 05:22:08 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 9638, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7911.72
+[2026-08-27 05:22:09 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1446, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11076.15
+[2026-08-27 05:22:09 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 9638, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10949.54
+[2026-08-27 05:22:10 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1446, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8408.34
+[2026-08-27 05:22:10 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 10033, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8234.48
+[2026-08-27 05:22:10 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1841, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10913.26
+[2026-08-27 05:22:10 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 10033, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10431.88
+[2026-08-27 05:22:11 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1856, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2259.05
+[2026-08-27 05:22:11 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1841, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9802.26
+[2026-08-27 05:22:12 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8540, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8332.93
+[2026-08-27 05:22:12 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1856, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1872.96
+[2026-08-27 05:22:13 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 17066, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9894.47
+[2026-08-27 05:22:13 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8540, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10158.45
+[2026-08-27 05:22:14 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8910, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8389.00
+[2026-08-27 05:22:14 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 17066, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8419.18
+[2026-08-27 05:22:15 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 17466, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10331.84
+[2026-08-27 05:22:15 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8910, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10077.68
+[2026-08-27 05:22:16 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9324, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9373.73
+[2026-08-27 05:22:16 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 17466, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8567.43
+[2026-08-27 05:22:16 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 17751, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10615.28
+[2026-08-27 05:22:17 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9324, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11315.42
+[2026-08-27 05:22:17 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 17751, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8963.12
+[2026-08-27 05:22:17 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9579, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8324.62
+[2026-08-27 05:22:18 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1387, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 35673.33
+[2026-08-27 05:22:18 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9579, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 35404.74
+[2026-08-27 05:22:18 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1387, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 314704.56
+[2026-08-27 05:22:18 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1408, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 52789.70
+[2026-08-27 05:22:18 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1408, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 13474.13
+[2026-08-27 05:22:25 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8547, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1143.27
+[2026-08-27 05:22:26 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 355, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10109.37
+[2026-08-27 05:22:26 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8547, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1034.06
+[2026-08-27 05:22:27 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 355, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7893.67
+[2026-08-27 05:22:27 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8932, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7504.02
+[2026-08-27 05:22:28 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 740, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11535.13
+[2026-08-27 05:22:28 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8932, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10838.23
+[2026-08-27 05:22:29 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 740, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8155.37
+[2026-08-27 05:22:29 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9259, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7789.29
+[2026-08-27 05:22:29 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1067, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10733.11
+[2026-08-27 05:22:29 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9259, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10388.64
+[2026-08-27 05:22:30 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1067, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8544.20
+[2026-08-27 05:22:30 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9745, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8357.94
+[2026-08-27 05:22:31 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1553, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10706.17
+[2026-08-27 05:22:31 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9745, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10445.99
+[2026-08-27 05:22:32 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1600, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1952.39
+[2026-08-27 05:22:32 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1553, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9931.31
+[2026-08-27 05:22:33 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8602, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8370.54
+[2026-08-27 05:22:33 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1600, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1624.80
+[2026-08-27 05:22:34 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 410, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9785.83
+[2026-08-27 05:22:34 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8602, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9969.67
+[2026-08-27 05:22:35 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8696, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8591.23
+[2026-08-27 05:22:35 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 410, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8607.27
+[2026-08-27 05:22:35 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 504, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10402.38
+[2026-08-27 05:22:35 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8696, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10102.69
+[2026-08-27 05:22:36 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9164, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9308.92
+[2026-08-27 05:22:36 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 504, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8520.94
+[2026-08-27 05:22:37 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 972, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10664.47
+[2026-08-27 05:22:37 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9164, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11299.50
+[2026-08-27 05:22:38 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 972, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8956.53
+[2026-08-27 05:22:38 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9322, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8269.42
+[2026-08-27 05:22:38 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1130, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 43785.83
+[2026-08-27 05:22:38 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9322, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 43114.10
+[2026-08-27 05:22:38 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1130, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 348034.30
+[2026-08-27 05:22:38 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1152, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 43670.18
+[2026-08-27 05:22:38 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1152, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10950.18
+[2026-08-27 05:22:45 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8602, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1146.51
+[2026-08-27 05:22:46 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 410, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10113.13
+[2026-08-27 05:22:46 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8602, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1036.82
+[2026-08-27 05:22:47 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 410, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8059.88
+[2026-08-27 05:22:47 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 9105, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7668.92
+[2026-08-27 05:22:48 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 913, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11406.02
+[2026-08-27 05:22:48 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 9105, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11028.76
+[2026-08-27 05:22:49 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 913, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8034.52
+[2026-08-27 05:22:49 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9375, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7803.47
+[2026-08-27 05:22:50 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1183, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10701.85
+[2026-08-27 05:22:50 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9375, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10338.95
+[2026-08-27 05:22:51 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1183, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8629.50
+[2026-08-27 05:22:51 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9868, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8448.89
+[2026-08-27 05:22:52 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1676, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10696.20
+[2026-08-27 05:22:52 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9868, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10383.50
+[2026-08-27 05:22:52 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2088.70
+[2026-08-27 05:22:52 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1676, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9980.67
+[2026-08-27 05:22:53 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8467, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8474.75
+[2026-08-27 05:22:53 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1765.75
+[2026-08-27 05:22:54 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 275, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10047.97
+[2026-08-27 05:22:54 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8467, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10138.75
+[2026-08-27 05:22:55 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8988, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8471.06
+[2026-08-27 05:22:55 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 275, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8570.54
+[2026-08-27 05:22:56 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 796, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10303.51
+[2026-08-27 05:22:56 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8988, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9711.56
+[2026-08-27 05:22:57 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9363, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9482.47
+[2026-08-27 05:22:57 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 796, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8960.80
+[2026-08-27 05:22:58 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 1171, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10488.36
+[2026-08-27 05:22:58 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9363, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11295.90
+[2026-08-27 05:22:59 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 1171, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8985.86
+[2026-08-27 05:22:59 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9880, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8497.44
+[2026-08-27 05:22:59 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1688, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 30171.97
+[2026-08-27 05:22:59 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9880, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 29961.56
+[2026-08-27 05:22:59 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1688, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 384839.25
+[2026-08-27 05:22:59 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 77829.47
+[2026-08-27 05:22:59 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 17255.50
+[2026-08-27 05:23:06 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8601, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1148.36
+[2026-08-27 05:23:07 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 409, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10091.44
+[2026-08-27 05:23:07 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8601, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1039.48
+[2026-08-27 05:23:08 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 409, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8052.72
+[2026-08-27 05:23:08 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 9104, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7736.57
+[2026-08-27 05:23:09 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 912, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11209.86
+[2026-08-27 05:23:09 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 9104, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10701.52
+[2026-08-27 05:23:10 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 912, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8193.31
+[2026-08-27 05:23:10 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9485, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7925.16
+[2026-08-27 05:23:10 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1293, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10717.69
+[2026-08-27 05:23:10 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9485, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10383.19
+[2026-08-27 05:23:11 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1293, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8591.94
+[2026-08-27 05:23:11 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9900, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8353.36
+[2026-08-27 05:23:12 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1708, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10835.80
+[2026-08-27 05:23:12 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9900, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10393.61
+[2026-08-27 05:23:13 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2099.70
+[2026-08-27 05:23:13 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1708, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9919.38
+[2026-08-27 05:23:14 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8525, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8406.32
+[2026-08-27 05:23:14 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1756.57
+[2026-08-27 05:23:15 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 333, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10057.88
+[2026-08-27 05:23:15 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8525, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9966.29
+[2026-08-27 05:23:16 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8909, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8449.44
+[2026-08-27 05:23:16 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 333, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8768.58
+[2026-08-27 05:23:17 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 717, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10363.27
+[2026-08-27 05:23:17 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8909, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9712.12
+[2026-08-27 05:23:17 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9452, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9349.89
+[2026-08-27 05:23:17 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 717, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8950.85
+[2026-08-27 05:23:18 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 1260, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10634.96
+[2026-08-27 05:23:18 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9452, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11269.38
+[2026-08-27 05:23:19 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 1260, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9067.98
+[2026-08-27 05:23:19 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9830, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8519.49
+[2026-08-27 05:23:19 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1638, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33116.32
+[2026-08-27 05:23:19 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9830, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32958.23
+[2026-08-27 05:23:19 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1638, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 385069.93
+[2026-08-27 05:23:19 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1664, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 72026.72
+[2026-08-27 05:23:19 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1664, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 77715.90
+[2026-08-27 05:32:52 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.11
+[2026-08-27 05:32:52 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.11
+[2026-08-27 05:35:54 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 45.18
+[2026-08-27 05:35:54 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 45.15
+[2026-08-27 05:35:54 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 72669.47
+[2026-08-27 05:35:54 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 516710.98
+[2026-08-27 05:35:54 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32925.10
+[2026-08-27 05:35:54 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 28289.90
+[2026-08-27 05:36:34 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1.59
+[2026-08-27 05:36:34 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1.59
+[2026-08-27 05:42:31 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 8652, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 22.93
+[2026-08-27 05:42:32 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 3, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10563.25
+[2026-08-27 05:42:32 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 8652, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 22.88
+[2026-08-27 05:42:33 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 3, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8086.11
+[2026-08-27 05:42:33 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 3, #pending-token: 9185, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7946.28
+[2026-08-27 05:42:34 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 2, #pending-token: 993, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10512.38
+[2026-08-27 05:42:34 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 3, #pending-token: 9185, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10479.70
+[2026-08-27 05:42:35 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 2, #pending-token: 993, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8287.73
+[2026-08-27 05:42:35 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 2, #pending-token: 9680, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8255.07
+[2026-08-27 05:42:36 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1488, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10409.89
+[2026-08-27 05:42:36 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 2, #pending-token: 9680, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10321.51
+[2026-08-27 05:42:37 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1488, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8596.01
+[2026-08-27 05:42:37 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 10199, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8531.89
+[2026-08-27 05:42:38 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 18858, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10150.76
+[2026-08-27 05:42:38 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 10199, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10112.26
+[2026-08-27 05:42:38 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 18858, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9106.30
+[2026-08-27 05:42:38 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 10707, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9092.22
+[2026-08-27 05:42:39 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 53023, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10201.63
+[2026-08-27 05:42:39 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 10707, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9511.61
+[2026-08-27 05:42:40 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 44876, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9125.12
+[2026-08-27 05:42:40 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 53023, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9650.26
+[2026-08-27 05:42:41 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 36684, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9672.38
+[2026-08-27 05:42:41 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 44876, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9545.39
+[2026-08-27 05:42:42 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 28504, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10617.25
+[2026-08-27 05:42:42 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 36684, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9479.22
+[2026-08-27 05:42:43 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 28504, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10925.86
+[2026-08-27 05:42:43 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 20312, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9493.34
+[2026-08-27 05:42:44 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 20312, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9446.28
+[2026-08-27 05:42:44 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 12139, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9440.78
+[2026-08-27 05:42:44 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 12139, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 15388.21
+[2026-08-27 05:42:44 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 3947, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 15425.68
+[2026-08-27 05:42:44 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 3968, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 193230.19
+[2026-08-27 05:42:44 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 3947, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 384008.79
+[2026-08-27 05:42:44 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 3968, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 225535.33
+[2026-08-27 05:43:18 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8608, #bootstrap-req: 1, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 241.39
+[2026-08-27 05:43:18 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8608, #bootstrap-req: 1, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 240.35
+[2026-08-27 05:43:18 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 416, #bootstrap-req: 2, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 49614.95
+[2026-08-27 05:43:18 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 416, #bootstrap-req: 2, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 436819.44
+[2026-08-27 05:43:18 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 448, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 23041.86
+[2026-08-27 05:43:18 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 448, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 23367.60
+[2026-08-27 05:43:25 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.06, #running-req: 0, #queue-req: 0, #pending-token: 8701, #bootstrap-req: 4, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1138.34
+[2026-08-27 05:43:26 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.06, #running-req: 0, #queue-req: 0, #pending-token: 509, #bootstrap-req: 4, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 13794.50
+[2026-08-27 05:43:26 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.06, #running-req: 0, #queue-req: 0, #pending-token: 8701, #bootstrap-req: 4, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1039.00
+[2026-08-27 05:43:27 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.01, mamba usage: 0.06, #running-req: 0, #queue-req: 4, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 620.23
+[2026-08-27 05:43:27 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.06, #running-req: 0, #queue-req: 0, #pending-token: 509, #bootstrap-req: 4, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9454.69
+[2026-08-27 05:43:28 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 3, #pending-token: 8678, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6776.48
+[2026-08-27 05:43:28 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.01, mamba usage: 0.06, #running-req: 0, #queue-req: 4, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 484.41
+[2026-08-27 05:43:29 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 3, #pending-token: 8678, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11332.72
+[2026-08-27 05:43:29 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 3, #pending-token: 17239, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11313.04
+[2026-08-27 05:43:30 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 3, #pending-token: 17239, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7697.99
+[2026-08-27 05:43:30 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 2, #pending-token: 9073, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7651.13
+[2026-08-27 05:43:31 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 2, #pending-token: 881, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11594.46
+[2026-08-27 05:43:31 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 2, #pending-token: 9073, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11475.27
+[2026-08-27 05:43:31 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 2, #pending-token: 881, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10069.18
+[2026-08-27 05:43:31 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 896, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1099.01
+[2026-08-27 05:43:32 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 896, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1109.06
+[2026-08-27 05:43:34 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 58625, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3228.33
+[2026-08-27 05:43:35 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 50433, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9950.31
+[2026-08-27 05:43:35 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 58625, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3166.65
+[2026-08-27 05:43:36 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 42277, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8191.42
+[2026-08-27 05:43:36 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 50433, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8327.66
+[2026-08-27 05:43:36 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 34085, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10384.47
+[2026-08-27 05:43:37 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 42277, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9958.08
+[2026-08-27 05:43:37 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 34085, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9632.24
+[2026-08-27 05:43:37 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 25943, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9081.82
+[2026-08-27 05:43:38 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 17751, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10679.25
+[2026-08-27 05:43:38 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 25943, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9927.41
+[2026-08-27 05:43:39 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 17751, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8846.04
+[2026-08-27 05:43:39 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 9579, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8328.29
+[2026-08-27 05:43:39 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 0, #pending-token: 1387, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 37248.27
+[2026-08-27 05:43:39 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 9579, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 37039.79
+[2026-08-27 05:43:39 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 0, #pending-token: 1387, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 410683.07
+[2026-08-27 05:43:39 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1408, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 67821.99
+[2026-08-27 05:43:39 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1408, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 57390.70
+[2026-08-27 05:43:49 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8547, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 809.85
+[2026-08-27 05:43:50 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8547, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 804.33
+[2026-08-27 05:43:50 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 355, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 86353.57
+[2026-08-27 05:43:50 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 355, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 517048.17
+[2026-08-27 05:43:50 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 24689.14
+[2026-08-27 05:43:50 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 23957.28
+[2026-08-27 05:44:08 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8548, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 438.31
+[2026-08-27 05:44:08 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8548, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 436.68
+[2026-08-27 05:44:08 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 356, #bootstrap-req: 1, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 94252.23
+[2026-08-27 05:44:08 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 356, #bootstrap-req: 1, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 468831.90
+[2026-08-27 05:44:08 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 23149.14
+[2026-08-27 05:44:08 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 22626.22
+[2026-08-27 05:44:16 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8491, #bootstrap-req: 1, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1131.42
+[2026-08-27 05:44:16 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8491, #bootstrap-req: 1, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1122.89
+[2026-08-27 05:44:16 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 299, #bootstrap-req: 5, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 48252.54
+[2026-08-27 05:44:16 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 299, #bootstrap-req: 5, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 12541.51
+[2026-08-27 05:44:16 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 320, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 1, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 576.12
+[2026-08-27 05:44:17 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 320, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 1, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 630.41
+[2026-08-27 05:44:23 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.06, #running-req: 0, #queue-req: 3, #pending-token: 8657, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1233.09
+[2026-08-27 05:44:24 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.06, #running-req: 0, #queue-req: 3, #pending-token: 8657, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1193.29
+[2026-08-27 05:44:24 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 2, #pending-token: 465, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11243.25
+[2026-08-27 05:44:24 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 2, #pending-token: 465, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11563.37
+[2026-08-27 05:44:24 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.01, mamba usage: 0.06, #running-req: 0, #queue-req: 2, #pending-token: 0, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 722.50
+[2026-08-27 05:44:25 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.01, mamba usage: 0.06, #running-req: 0, #queue-req: 2, #pending-token: 0, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 696.35
+[2026-08-27 05:44:28 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 58376, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2510.51
+[2026-08-27 05:44:29 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 50184, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10170.95
+[2026-08-27 05:44:29 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 58376, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2434.40
+[2026-08-27 05:44:30 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 50184, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7877.34
+[2026-08-27 05:44:30 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 42030, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7641.65
+[2026-08-27 05:44:30 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 33838, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11336.56
+[2026-08-27 05:44:30 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 42030, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10432.59
+[2026-08-27 05:44:31 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 33838, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9788.09
+[2026-08-27 05:44:31 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 25654, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9105.72
+[2026-08-27 05:44:32 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 17462, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10547.33
+[2026-08-27 05:44:32 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 25654, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8856.76
+[2026-08-27 05:44:33 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 9322, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7562.50
+[2026-08-27 05:44:33 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 17462, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8606.17
+[2026-08-27 05:44:34 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1130, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11071.46
+[2026-08-27 05:44:34 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 9322, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10245.93
+[2026-08-27 05:44:34 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1152, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2507.03
+[2026-08-27 05:44:35 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1130, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10403.29
+[2026-08-27 05:44:36 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8602, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6023.05
+[2026-08-27 05:44:36 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1152, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1203.53
+[2026-08-27 05:44:36 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8602, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 79810.68
+[2026-08-27 05:44:36 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 410, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 78639.05
+[2026-08-27 05:44:36 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 410, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 369547.92
+[2026-08-27 05:44:36 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 448, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 19914.93
+[2026-08-27 05:44:36 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 448, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 19106.69
+[2026-08-27 05:44:43 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8657, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1150.56
+[2026-08-27 05:44:43 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8657, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1136.07
+[2026-08-27 05:44:43 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 465, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 71289.71
+[2026-08-27 05:44:43 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 465, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 469218.72
+[2026-08-27 05:44:43 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 30708.34
+[2026-08-27 05:44:43 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 28589.43
+[2026-08-27 05:44:52 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8415, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 922.38
+[2026-08-27 05:44:52 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8415, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 915.18
+[2026-08-27 05:44:52 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 223, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 91921.18
+[2026-08-27 05:44:52 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 223, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 446710.96
+[2026-08-27 05:44:52 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 256, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 15114.63
+[2026-08-27 05:44:52 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 256, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 15291.41
+[2026-08-27 05:45:07 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 539.23
+[2026-08-27 05:45:07 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 535.73
+[2026-08-27 05:45:07 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 70998.26
+[2026-08-27 05:45:07 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 497029.45
+[2026-08-27 05:45:07 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 28706.58
+[2026-08-27 05:45:07 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 24009.64
+[2026-08-27 05:45:15 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8467, #bootstrap-req: 4, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1132.53
+[2026-08-27 05:45:15 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8467, #bootstrap-req: 4, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1120.29
+[2026-08-27 05:45:15 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 275, #bootstrap-req: 4, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 76264.46
+[2026-08-27 05:45:15 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 275, #bootstrap-req: 4, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 255786.52
+[2026-08-27 05:45:15 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 320, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 13135.03
+[2026-08-27 05:45:15 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 320, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 12259.67
+[2026-08-27 05:45:22 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 42247, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1113.51
+[2026-08-27 05:45:23 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 42247, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1014.85
+[2026-08-27 05:45:23 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 34055, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10965.71
+[2026-08-27 05:45:24 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 34055, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8543.60
+[2026-08-27 05:45:24 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 42692, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8383.12
+[2026-08-27 05:45:24 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 34500, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 12845.52
+[2026-08-27 05:45:25 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 42692, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9919.38
+[2026-08-27 05:45:25 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 34500, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9887.39
+[2026-08-27 05:45:25 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 26353, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8253.73
+[2026-08-27 05:45:26 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 18161, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9379.50
+[2026-08-27 05:45:26 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 26353, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8848.86
+[2026-08-27 05:45:27 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 26857, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9462.22
+[2026-08-27 05:45:27 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 18161, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8853.61
+[2026-08-27 05:45:28 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 18665, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10294.00
+[2026-08-27 05:45:28 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 26857, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11244.06
+[2026-08-27 05:45:29 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 18665, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9066.60
+[2026-08-27 05:45:29 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 10512, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8106.82
+[2026-08-27 05:45:30 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 2320, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 13213.03
+[2026-08-27 05:45:30 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 10512, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11962.83
+[2026-08-27 05:45:30 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 2320, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10926.25
+[2026-08-27 05:45:30 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 2368, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3139.62
+[2026-08-27 05:45:31 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 2368, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3314.76
+[2026-08-27 05:45:36 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.06, #running-req: 0, #queue-req: 0, #pending-token: 8525, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1442.24
+[2026-08-27 05:45:36 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.06, #running-req: 0, #queue-req: 0, #pending-token: 8525, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1615.76
+[2026-08-27 05:45:36 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.04, #running-req: 0, #queue-req: 0, #pending-token: 333, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 80517.26
+[2026-08-27 05:45:36 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.04, #running-req: 0, #queue-req: 0, #pending-token: 333, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 427738.71
+[2026-08-27 05:45:36 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.01, mamba usage: 0.04, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 19525.14
+[2026-08-27 05:45:36 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.01, mamba usage: 0.04, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 20377.23
+[2026-08-27 05:45:43 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8556, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1154.07
+[2026-08-27 05:45:43 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8556, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1139.26
+[2026-08-27 05:45:43 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 364, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 73751.78
+[2026-08-27 05:45:43 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 364, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 476927.66
+[2026-08-27 05:45:43 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 22507.33
+[2026-08-27 05:45:43 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25194.61
+[2026-08-27 05:45:52 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8525, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 998.75
+[2026-08-27 05:45:52 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8525, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 984.01
+[2026-08-27 05:45:52 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 333, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 59174.26
+[2026-08-27 05:45:52 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 333, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 503593.82
+[2026-08-27 05:45:52 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 24112.53
+[2026-08-27 05:45:52 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 21827.61
+[2026-08-27 05:46:09 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8525, #bootstrap-req: 2, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 487.30
+[2026-08-27 05:46:09 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8525, #bootstrap-req: 2, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 485.07
+[2026-08-27 05:46:09 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 333, #bootstrap-req: 2, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 86033.08
+[2026-08-27 05:46:09 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 333, #bootstrap-req: 2, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 493611.64
+[2026-08-27 05:46:09 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 23640.65
+[2026-08-27 05:46:09 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3592.53
+[2026-08-27 05:46:16 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.06, #running-req: 0, #queue-req: 0, #pending-token: 8684, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1149.55
+[2026-08-27 05:46:16 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.04, #running-req: 0, #queue-req: 0, #pending-token: 492, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10871.43
+[2026-08-27 05:46:17 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.06, #running-req: 0, #queue-req: 0, #pending-token: 8684, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1041.74
+[2026-08-27 05:46:18 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.04, #running-req: 0, #queue-req: 0, #pending-token: 492, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8633.25
+[2026-08-27 05:46:18 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.04, #running-req: 0, #queue-req: 0, #pending-token: 9062, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7876.78
+[2026-08-27 05:46:18 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.04, #running-req: 0, #queue-req: 0, #pending-token: 870, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 52518.55
+[2026-08-27 05:46:18 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.04, #running-req: 0, #queue-req: 0, #pending-token: 9062, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 51887.94
+[2026-08-27 05:46:18 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.04, #running-req: 0, #queue-req: 0, #pending-token: 870, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 444529.19
+[2026-08-27 05:46:18 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 896, #cached-token: 0, full token usage: 0.01, mamba usage: 0.04, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 46656.41
+[2026-08-27 05:46:18 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 896, #cached-token: 0, full token usage: 0.01, mamba usage: 0.04, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 49911.91
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-c8-20260827-1340/logs/manual_p_node2.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-c8-20260827-1340/logs/manual_p_node2.log
new file mode 100644
index 0000000..604ea9b
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-c8-20260827-1340/logs/manual_p_node2.log
@@ -0,0 +1,1264 @@
+/usr/local/lib/python3.12/dist-packages/torchao/quantization/quant_api.py:1731: SyntaxWarning: invalid escape sequence '\.'
+  """Configuration class for applying different quantization configs to modules or parameters based on their fully qualified names (FQNs).
+/sgl-workspace/sglang/python/sglang/launch_server.py:56: UserWarning: 'python -m sglang.launch_server' is still supported, but 'sglang serve' is the recommended entrypoint.
+  Example: sglang serve --model-path <model> [options]
+  warnings.warn(
+[2026-08-27 04:28:43] Breakable CUDA graph is incompatible with MLA attention; disabling prefill CUDA graph.
+[2026-08-27 04:28:43] Attention backend not specified. Use triton backend by default.
+[2026-08-27 04:28:43] Pipeline parallelism is incompatible with overlap schedule.
+[2026-08-27 04:28:45] server_args=ServerArgs(model_path='/data/hf_models/Kimi-K3', tokenizer_path='/data/hf_models/Kimi-K3', tokenizer_mode='auto', tokenizer_backend='huggingface', tokenizer_worker_num=1, detokenizer_worker_num=1, skip_tokenizer_init=False, load_format='auto', model_loader_extra_config='{}', trust_remote_code=True, context_length=None, is_embedding=False, enable_multimodal=None, revision=None, model_impl='auto', model_config_parser='auto', json_model_override_args='{}', dtype='auto', quantization=None, quantization_param_path=None, kv_cache_dtype='auto', enable_fp32_lm_head=False, modelopt_quant=None, modelopt_checkpoint_restore_path=None, modelopt_checkpoint_save_path=None, modelopt_export_path=None, quantize_and_serve=False, rl_quant_profile=None, enable_tf32_matmul=False, mem_fraction_static=0.88, max_running_requests=None, max_queued_requests=None, max_total_tokens=None, chunked_prefill_size=8192, enable_dynamic_chunking=False, max_prefill_tokens=16384, prefill_max_requests=None, schedule_policy='fcfs', enable_priority_scheduling=False, disable_priority_preemption=False, default_priority_value=None, abort_on_priority_when_disabled=False, schedule_low_priority_values_first=False, priority_scheduling_preemption_threshold=10, retraction_policy='length', schedule_conservativeness=1.0, page_size=64, swa_full_tokens_ratio=0.8, disable_hybrid_swa_memory=False, radix_eviction_policy='lru', prefill_only_disable_kv_cache=False, disable_radix_cache=True, enable_page_major_kv_layout=False, enable_unified_memory=False, disable_chunked_prefix_cache=False, disable_overlap_schedule=True, num_continuous_decode_steps=1, scheduler_recv_interval=1, enable_mixed_chunk=False, nccl_port=None, dist_timeout=3600, dist_init_addr='174.1.60.1:20000', nnodes=4, node_rank=2, tp_size=4, dcp_size=1, pp_size=8, pp_max_micro_batch_size=None, pp_async_batch_depth=0, dp_size=1, load_balance_method='follow_bootstrap_room', attn_cp_size=1, moe_dp_size=1, dcp_comm_backend='ag_rs', dcp_replicate_q_proj=None, dwdp_size=1, enable_prefill_cp=False, cp_strategy=None, enable_dsa_cache_layer_split=False, enable_dsa_prefill_context_parallel=False, dsa_prefill_cp_mode='round-robin-split', enable_prefill_context_parallel=False, prefill_cp_mode='in-seq-split', enable_cp_decode_attn_tp=False, enable_dp_attention=False, enable_dp_attention_local_control_broadcast=False, enable_dp_lm_head=False, enable_attn_tp_input_scattered=False, disable_attn_tp_gather=False, enable_p2p_check=False, device='cuda', base_gpu_id=0, gpu_id_step=1, random_seed=336932344, watchdog_timeout=300, soft_watchdog_timeout=None, sleep_on_idle=False, use_ray=False, custom_sigquit_handler=None, numa_node=None, gc_threshold=None, host='0.0.0.0', port=30000, fastapi_root_path='', smg_grpc_mode=False, grpc_mode=False, grpc_port=None, sidecar=None, sidecar_args=None, skip_server_warmup=False, warmups=None, enable_http2=False, ssl_keyfile=None, ssl_certfile=None, ssl_ca_certs=None, ssl_keyfile_password=None, enable_ssl_refresh=False, api_key=None, admin_api_key=None, served_model_name='kimi-k3', weight_version='default', chat_template=None, hf_chat_template_name=None, completion_template=None, file_storage_path='sglang_storage', enable_cache_report=False, reasoning_parser=None, default_chat_template_kwargs=None, strip_thinking_cache=False, enable_strict_thinking=False, tool_call_parser=None, tool_server=None, sampling_defaults='model', asr_max_buffer_seconds=60, asr_max_concurrent_sessions=32, preferred_sampling_params=None, allow_auto_truncate=False, stream_interval=1, batch_notify_size=16, stream_response_default_include_usage=False, incremental_streaming_output=False, enable_streaming_session=False, enable_session_radix_cache=False, log_level='info', log_level_http=None, log_requests=False, log_requests_level=2, log_requests_format='text', log_requests_target=None, uvicorn_access_log_exclude_prefixes=[], crash_dump_folder=None, show_time_cost=False, enable_metrics=False, smg_http_sidecar_port=None, enable_mfu_metrics=False, enable_metrics_for_all_schedulers=False, load_snapshot_publish_interval=15, tokenizer_metrics_custom_labels_header='x-custom-labels', tokenizer_metrics_allowed_custom_labels=None, extra_metric_labels=None, bucket_time_to_first_token=None, bucket_inter_token_latency=None, bucket_e2e_request_latency=None, prompt_tokens_buckets=None, generation_tokens_buckets=None, gc_warning_threshold_secs=0.0, decode_log_interval=40, enable_request_time_stats_logging=False, kv_events_config=None, enable_forward_pass_metrics=False, forward_pass_metrics_worker_id='', forward_pass_metrics_ipc_name=None, enable_trace=False, trace_modules='request', otlp_traces_endpoint='localhost:4317', export_metrics_to_file=False, export_metrics_to_file_dir=None, stat_loggers=None, constrained_json_whitespace_pattern=None, constrained_json_disable_any_whitespace=False, attention_backend='triton', decode_attention_backend=None, prefill_attention_backend=None, sampling_backend='flashinfer', grammar_backend='xgrammar', radix_cache_backend=None, mm_attention_backend=None, fp8_gemm_runner_backend='auto', fp4_gemm_runner_backend='auto', bf16_gemm_backend='auto', dsa_prefill_backend=None, dsa_decode_backend=None, dsa_paged_mqa_logits_backend='auto', dsa_topk_backend='sgl-kernel', disable_flashinfer_autotune=False, flashinfer_autotune_skip_ops=None, mamba_backend='triton', cuda_graph_config=CudaGraphConfig(decode=PhaseConfig(backend='disabled', max_bs=16, bs=[1, 2, 4, 8, 12, 16], tc_compiler='eager', full_prefill_max_req=None), prefill=PhaseConfig(backend='disabled', max_bs=2048, bs=[4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], tc_compiler='eager', full_prefill_max_req=None)), cuda_graph_backend_decode=None, cuda_graph_backend_prefill=None, cuda_graph_max_bs_decode=16, cuda_graph_max_bs_prefill=None, cuda_graph_bs_decode=None, cuda_graph_bs_prefill=None, cuda_graph_tc_compiler=None, disable_prefill_cuda_graph=False, disable_decode_cuda_graph=False, disable_cuda_graph=False, disable_cuda_graph_padding=False, enable_profile_cuda_graph=False, enable_cudagraph_gc=False, debug_cuda_graph=False, enable_layerwise_nvtx_marker=False, enable_nccl_nvls=False, enable_symm_mem=False, triton_attention_reduce_in_fp32=False, triton_attention_num_kv_splits=8, triton_attention_split_tile_size=None, flashinfer_mla_disable_ragged=False, enable_fused_qk_norm_rope=False, enable_precise_embedding_interpolation=False, enable_fused_moe_sum_all_reduce=False, enable_deepseek_v4_fp4_indexer=False, disable_custom_all_reduce=False, enable_mscclpp=False, enable_torch_symm_mem=False, enable_scattered_sconv=False, pre_warm_nccl=False, enable_quant_communications=False, enable_flashinfer_allreduce_fusion=False, enforce_disable_flashinfer_allreduce_fusion=False, flashinfer_allreduce_fusion_backend=None, enable_aiter_allreduce_fusion=False, enable_torch_compile=False, enable_torch_compile_debug_mode=False, torch_compile_max_bs=32, torchao_config='', speculative_algorithm=None, speculative_draft_model_path=None, speculative_draft_model_revision=None, speculative_draft_load_format=None, speculative_num_steps=None, speculative_eagle_topk=None, speculative_num_draft_tokens=None, speculative_dflash_block_size=None, speculative_dspark_block_size=None, speculative_dspark_sps_table_path=None, speculative_dspark_confidence_sts_path=None, speculative_dspark_align_verify_tokens_to_graph_tier=False, speculative_accept_threshold_single=1.0, speculative_accept_threshold_acc=1.0, speculative_use_rejection_sampling=False, speculative_token_map=None, speculative_attention_mode='prefill', speculative_draft_attention_backend=None, speculative_draft_window_size=None, speculative_moe_runner_backend='flashinfer_mxfp4', speculative_moe_a2a_backend=None, speculative_draft_model_quantization=None, speculative_skip_dp_mlp_sync=False, enable_multi_layer_eagle=False, speculative_adaptive=False, speculative_adaptive_config=None, decoupled_spec_bind_endpoint=None, decoupled_spec_connect_endpoints=None, decoupled_spec_rank=None, decoupled_spec_role='null', spec_trace_dir=None, speculative_ngram_min_bfs_breadth=1, speculative_ngram_max_bfs_breadth=10, speculative_ngram_match_type='BFS', speculative_ngram_max_trie_depth=18, speculative_ngram_capacity=10000000, speculative_ngram_external_corpus_path=None, speculative_ngram_external_sam_budget=0, speculative_ngram_external_corpus_max_tokens=10000000, ep_size=4, moe_a2a_backend='none', moe_runner_backend='flashinfer_mxfp4', flashinfer_mxfp4_moe_precision='default', deepep_mode='auto', fuseep_mode=2, deepep_dispatcher_output_dtype='auto', ep_num_redundant_experts=0, ep_dispatch_algorithm=None, init_expert_location='trivial', enable_eplb=False, eplb_algorithm='auto', eplb_rebalance_num_iterations=1000, eplb_rebalance_layers_per_chunk=None, eplb_min_rebalancing_utilization_threshold=1.0, expert_distribution_recorder_mode=None, expert_distribution_recorder_buffer_size=1000, enable_expert_distribution_metrics=False, deepep_config=None, moe_dense_tp_size=None, elastic_ep_backend=None, enable_elastic_expert_backup=False, mooncake_ib_device=None, enable_waterfill=False, ep_join_mode=None, ep_join_rank_offset=0, elastic_ep_initial_size=None, max_ep_size=None, elastic_ep_scale_timeout=600, elastic_ep_rejoin=False, disable_flashinfer_cutlass_moe_fp4_allgather=False, disable_shared_experts_fusion=False, enforce_shared_experts_fusion=False, max_mamba_cache_size=None, mamba_ssm_dtype=None, mamba_max_states_per_path=-1, enable_mamba_cache_stochastic_rounding=False, mamba_cache_philox_rounds=0, mamba_full_memory_ratio=0.36, mamba_radix_cache_strategy='extra_buffer_lazy', uses_mamba_radix_cache=False, mamba_track_interval=256, enable_int8_mamba_checkpoint=False, int8_mamba_ckpt_size=None, linear_attn_backend='triton', linear_attn_decode_backend=None, linear_attn_prefill_backend=None, linear_attn_verify_backend=None, enable_linear_replayssm=False, linear_replayssm_cache_len=16, enable_linear_replayssm_spec=False, enable_hierarchical_cache=False, hicache_ratio=2.0, hicache_size=0, hicache_write_policy='write_through', hicache_io_backend='kernel', hicache_mem_layout='page_first', hicache_storage_backend=None, hicache_storage_prefetch_policy='timeout', hicache_storage_backend_extra_config=None, enable_hisparse=False, hisparse_config=None, enable_broadcast_mm_inputs_process=False, enable_prefix_mm_cache=False, mm_enable_dp_encoder=False, mm_process_config={}, mm_processor_worker_num=0, mm_io_worker_num=0, limit_mm_data_per_request=None, enable_mm_global_cache=False, disable_fast_image_processor=False, mm_feature_transport='cpu', keep_mm_feature_on_device=False, enable_lora=None, enable_lora_overlap_loading=None, max_lora_rank=None, lora_target_modules=None, lora_paths=None, max_loaded_loras=None, max_loras_per_batch=8, lora_eviction_policy='lru', lora_backend='csgmv', max_lora_chunk_size=16, experts_shared_outer_loras=None, lora_use_virtual_experts=False, lora_strict_loading=False, lora_drain_wait_threshold=0.0, enable_two_batch_overlap=False, enable_single_batch_overlap=False, tbo_token_distribution_threshold=0.48, cpu_offload_gb=0, offload_group_size=-1, offload_num_in_group=1, offload_prefetch_step=1, offload_mode='cpu', enable_lmcache=False, lmcache_config_file=None, enable_flexkv=False, flexkv_config_file=None, kt_weight_path=None, kt_method='AMXINT4', kt_cpuinfer=None, kt_threadpool_count=2, kt_num_gpu_experts=None, kt_max_deferred_experts_per_token=None, dllm_algorithm=None, dllm_algorithm_config=None, dllm_fdfo=True, disaggregation_mode='prefill', disaggregation_transfer_backend='mooncake', disaggregation_bootstrap_port=28800, disaggregation_ib_device='mlx5_0,mlx5_1,mlx5_2,mlx5_3', disaggregation_decode_enable_radix_cache=False, disaggregation_decode_enable_offload_kvcache=False, num_reserved_decode_tokens=512, disaggregation_decode_extra_slots=None, disaggregation_decode_polling_interval=1, optimistic_prefill_attempts=0, encoder_only=False, language_only=False, encoder_transfer_backend='zmq_to_tokenizer', encoder_urls=[], encoder_bootstrap_port=8997, encoder_register_urls=[], enable_adaptive_dispatch_to_encoder=False, enable_pdmux=False, pdmux_config_path=None, sm_group_num=8, custom_weight_loader=[], weight_loader_disable_mmap=False, weight_loader_prefetch_checkpoints=False, weight_loader_prefetch_num_threads=4, weight_loader_drop_cache_after_load=False, remote_instance_weight_loader_seed_instance_ip=None, remote_instance_weight_loader_seed_instance_service_port=None, remote_instance_weight_loader_send_weights_group_ports=None, remote_instance_weight_loader_backend='nccl', remote_instance_weight_loader_start_seed_via_transfer_engine=False, engine_info_bootstrap_port=6789, modelexpress_config=None, download_dir=None, model_checksum=None, delete_ckpt_after_loading=False, decrypted_config_file=None, decrypted_draft_config_file=None, checkpoint_engine_wait_weights_before_ready=False, enable_prefill_delayer=False, prefill_delayer_max_delay_passes=30, prefill_delayer_token_usage_low_watermark=None, prefill_delayer_forward_passes_buckets=None, prefill_delayer_wait_seconds_buckets=None, prefill_delayer_queue_min_ratio=None, prefill_delayer_max_delay_ms=None, min_free_slots_delay=None, enable_deterministic_inference=False, rl_on_policy_target=None, kv_canary='none', kv_canary_real_data='none', kv_canary_sweep_interval=0, enable_dynamic_batch_tokenizer=False, dynamic_batch_tokenizer_batch_size=32, dynamic_batch_tokenizer_batch_timeout=0.002, enable_tokenizer_batch_encode=False, disable_tokenizer_batch_decode=False, debug_tensor_dump_output_folder=None, debug_tensor_dump_layers=None, debug_tensor_dump_input_file=None, enable_memory_saver=False, enable_weights_cpu_backup=False, enable_draft_weights_cpu_backup=False, enable_custom_logit_processor=False, enable_return_hidden_states=False, enable_return_routed_experts=False, enable_return_indexer_topk=False, disable_outlines_disk_cache=False, enable_mis=False, weight_cache_mode='off', weight_cache_socket=None, weight_cache_timeout=1800, forward_hooks=None, msprobe_dump_config=None)
+[2026-08-27 04:28:45] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:45] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:45] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:45] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:45] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:45] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:45] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:45] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:01 PP4 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:01 PP4 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:01 PP5 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:01 PP5 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:01 PP4 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:01 PP4 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:01 PP5 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:01 PP5 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:01 PP4 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:01 PP4 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:01 PP4 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:01 PP4 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:02 PP5 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:02 PP5 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:02 PP5 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:02 PP5 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:03 PP5 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:03 PP5 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:03 PP4 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:03 PP4 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:03 PP5 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:03 PP5 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:03 PP4 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:03 PP4 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:05.195727   240 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:05.195828   240 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:05.195843   240 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.3 port: 12001
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:05.195847   236 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:05.195899   240 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.3:15652
+I0827 04:29:05.195935   236 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:05.195948   236 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.3 port: 12001
+I0827 04:29:05.195989   236 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.3:15032
+I0827 04:29:05.196039   240 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:05.196110   236 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:05.206050   241 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:05.206164   241 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:05.206182   241 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.3 port: 12001
+I0827 04:29:05.206233   241 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.3:16487
+I0827 04:29:05.206414   241 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:05.206894   238 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:05.207010   238 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:05.207029   238 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.3 port: 12001
+I0827 04:29:05.207082   238 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.3:15213
+I0827 04:29:05.207248   238 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:05.213657   238 topology.cpp:127] Device mlx5_0 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:05.215921   237 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:05.216045   237 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:05.216063   237 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.3 port: 12001
+I0827 04:29:05.216125   237 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.3:16450
+I0827 04:29:05.216308   237 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:05.219630   235 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:05.219745   235 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:05.219763   235 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.3 port: 12001
+I0827 04:29:05.219815   235 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.3:15658
+I0827 04:29:05.219810   236 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:05.220000   235 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:05.222787   237 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:05.223301   240 topology.cpp:127] Device mlx5_0 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:05.223796   239 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+W0827 04:29:05.223798   242 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:05.223948   242 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:05.223948   239 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:05.223969   242 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.3 port: 12001
+I0827 04:29:05.223973   239 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.3 port: 12001
+I0827 04:29:05.224030   242 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.3:16182
+I0827 04:29:05.224030   239 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.3:15581
+I0827 04:29:05.224237   242 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:05.224242   239 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:05.226227   235 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:05.227162   236 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:05.228888   237 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:05.229261   240 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:05.232141   235 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:05.232787   239 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:05.232861   242 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:05.233434   241 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:05.233938   236 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:05.236400   237 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:05.236641   240 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:05.238986   235 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:05.241036   236 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:05.241355   238 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:05.241638   241 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:05.242061   239 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:05.242151   242 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:05.244179   236 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:05.244223   236 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:05.244287   237 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:05.244297   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.244899   240 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:05.245494   235 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:05.249099   237 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:05.249130   237 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:05.249187   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.251125   235 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:05.251137   240 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:05.251168   235 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:05.251168   240 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:05.251241   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.251241   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.251624   239 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:05.251729   242 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:05.251823   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:05.253651   236 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:03
+I0827 04:29:05.253687   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.256857   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:05.257697   239 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:05.258617   237 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:03
+I0827 04:29:05.258652   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.258757   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:05.258822   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:05.260108   239 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:05.260136   239 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:05.260191   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.260488   235 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:03
+I0827 04:29:05.260502   240 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:03
+I0827 04:29:05.260530   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.260533   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.262187   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:05.264094   236 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:03
+I0827 04:29:05.264137   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.266639   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:05.268234   237 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:03
+I0827 04:29:05.268261   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.268549   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:05.269187   241 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:05.269232   238 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:05.270207   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:05.270285   240 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:03
+I0827 04:29:05.270316   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.271827   235 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:03
+I0827 04:29:05.271884   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.275274   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:05.276836   238 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:05.276907   242 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:05.276996   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:05.277058   236 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:03
+I0827 04:29:05.277101   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.278084   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:05.278632   237 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:03
+I0827 04:29:05.278667   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.279789   240 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:03
+I0827 04:29:05.279817   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.282152   238 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:05.282193   238 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:05.282260   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.283128   242 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:05.283200   242 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:05.283263   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.286593   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:05.287469   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:05.288283   239 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:03
+I0827 04:29:05.288322   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.289064   240 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:03
+I0827 04:29:05.289194   240 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:05 PP5 TP1 EP1] Init torch distributed begin.
+I0827 04:29:05.290841   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:05.292457   242 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:03
+I0827 04:29:05.292488   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.292999   241 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:05.296203   241 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:05.296272   241 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:05.296341   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.300206   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:05.301280   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:05.301775   242 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:03
+I0827 04:29:05.301801   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.303004   235 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:03
+I0827 04:29:05.303042   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.303999   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:05.305142   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:05.305728   241 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:03
+I0827 04:29:05.305768   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.305974   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:05.306758   237 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:03
+I0827 04:29:05.306921   237 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:05 PP4 TP2 EP2] Init torch distributed begin.
+I0827 04:29:05.307497   236 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:03
+I0827 04:29:05.307627   236 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:05 PP4 TP1 EP1] Init torch distributed begin.
+I0827 04:29:05.309624   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:05.309773   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:05.311504   238 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:03
+I0827 04:29:05.311542   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.311590   242 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:03
+I0827 04:29:05.311633   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.313076   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:05.314762   239 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:03
+I0827 04:29:05.314813   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.319522   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:05.319566   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:05.321249   242 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:03
+I0827 04:29:05.321398   242 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:05 PP5 TP3 EP3] Init torch distributed begin.
+I0827 04:29:05.326727   238 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:03
+I0827 04:29:05.326759   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.329339   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:05.331007   235 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:03
+I0827 04:29:05.331142   235 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:05 PP4 TP0 EP0] Init torch distributed begin.
+I0827 04:29:05.334666   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:05.336285   238 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:03
+I0827 04:29:05.336313   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.337296   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:05.338828   241 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:03
+I0827 04:29:05.338877   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.344906   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:05.346731   239 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:03
+I0827 04:29:05.346776   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.365299   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:05.366928   238 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:03
+I0827 04:29:05.367062   238 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:05 PP4 TP3 EP3] Init torch distributed begin.
+I0827 04:29:05.368464   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:05.369644   241 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:03
+I0827 04:29:05.369663   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.377357   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:05.379019   239 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:03
+I0827 04:29:05.379146   239 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:05 PP5 TP0 EP0] Init torch distributed begin.
+I0827 04:29:05.396831   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:05.398531   241 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:03
+I0827 04:29:05.398662   241 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:05 PP5 TP2 EP2] Init torch distributed begin.
+[2026-08-27 04:29:26 PP5 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:26 PP4 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:27 PP4 TP1 EP1] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP4 TP3 EP3] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP4 TP2 EP2] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP4 TP0 EP0] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP5 TP2 EP2] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP5 TP3 EP3] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP5 TP0 EP0] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP5 TP1 EP1] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP5 TP0 EP0] Init torch distributed ends. elapsed=22.37 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP5 TP3 EP3] Init torch distributed ends. elapsed=22.42 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP5 TP2 EP2] Init torch distributed ends. elapsed=22.35 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP4 TP1 EP1] Init torch distributed ends. elapsed=22.44 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP5 TP1 EP1] Init torch distributed ends. elapsed=22.46 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP4 TP3 EP3] Init torch distributed ends. elapsed=22.38 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP4 TP2 EP2] Init torch distributed ends. elapsed=22.44 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP4 TP0 EP0] Init torch distributed ends. elapsed=22.41 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP5 TP0 EP0] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP4 TP1 EP1] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP4 TP3 EP3] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP4 TP2 EP2] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP5 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP4 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP4 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP5 TP2 EP2] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP4 TP0 EP0] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP4 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP5 TP3 EP3] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP5 TP1 EP1] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP4 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP5 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP5 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP5 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP4 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP5 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP4 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP5 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP5 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP4 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP4 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP5 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP5 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP4 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP4 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP4 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP5 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP5 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP4 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP5 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:30:33 PP4 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:33 PP4 TP2 EP2] Load weight end. elapsed=65.50 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:34 PP4 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:34 PP4 TP3 EP3] Load weight end. elapsed=66.61 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP4 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:36 PP4 TP1 EP1] Load weight end. elapsed=68.45 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:36 PP5 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:36 PP5 TP0 EP0] Load weight end. elapsed=69.00 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:37 PP5 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:37 PP5 TP3 EP3] Load weight end. elapsed=69.63 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:37 PP5 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:37 PP5 TP1 EP1] Load weight end. elapsed=69.72 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:38 PP5 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:38 PP5 TP2 EP2] Load weight end. elapsed=70.08 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:38 PP5 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:38 PP5 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:38 PP5 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:38 PP5 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:38 PP5 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:38 PP5 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:38 PP5 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:38 PP5 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:47 PP4 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:47 PP4 TP0 EP0] Load weight end. elapsed=79.31 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:47 PP4 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:47 PP4 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:47 PP4 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:47 PP4 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:47 PP4 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:47 PP4 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:47 PP4 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:47 PP4 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:54 PP5 TP3 EP3] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP5 TP1 EP1] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP5 TP2 EP2] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP5 TP0 EP0] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP4 TP3 EP3] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP4 TP2 EP2] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP4 TP1 EP1] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP4 TP0 EP0] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP5 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP5 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP4 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP4 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP4 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP5 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP4 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP5 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP5 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP5 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP4 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP5 TP3 EP3] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP4 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP4 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP4 TP1 EP1] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP5 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP5 TP2 EP2] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP4 TP3 EP3] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP4 TP0 EP0] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP5 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP5 TP1 EP1] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP5 TP0 EP0] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP4 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP4 TP2 EP2] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP4 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:30:54 PP4 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:30:54 PP4 TP0 EP0] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp0_pp4_dp0.json
+[2026-08-27 04:30:54 PP4 TP1 EP1] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp1_pp4_dp0.json
+[2026-08-27 04:30:54 PP5 TP2 EP2] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp2_pp5_dp0.json
+[2026-08-27 04:30:54 PP5 TP3 EP3] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp3_pp5_dp0.json
+[2026-08-27 04:30:54 PP5 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:30:54 PP5 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:30:54 PP5 TP0 EP0] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp0_pp5_dp0.json
+[2026-08-27 04:30:54 PP4 TP2 EP2] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp2_pp4_dp0.json
+[TensorRT-LLM][INFO] Set logger level to INFO
+[2026-08-27 04:30:54 PP4 TP3 EP3] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp3_pp4_dp0.json
+[2026-08-27 04:30:54 PP5 TP1 EP1] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp1_pp5_dp0.json
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO[TensorRT-LLM][INFO] Set logger level to INFO
+
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.37profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.12profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.87profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.32profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.91profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.77profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.94profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.88profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.20profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.87profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  7.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 10.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.46profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.67profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.91profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.29profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.49profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  7.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.30profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.94profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.47profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.12profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.06profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.35profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  9.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:01<00:02,  2.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.89profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.58profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.32profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.70profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.20profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.27profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  6.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  7.03profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.56profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.73profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.88profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.37profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:05,  1.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:01<00:02,  2.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.05profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.19profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.77profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.18profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.25profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  3.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.18profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.20profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.37profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  9.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.27profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  8.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.43profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.89profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.35profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.96profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.90profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.30profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.72profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.18profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.33profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.51profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  2.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.03profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.68profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.16profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.01profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.24profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 10.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:02,  2.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.06profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.83profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.35profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.87profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.19profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  3.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.06profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.86profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 13.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 13.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.13profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:01,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:01,  3.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.47profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  6.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.47profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.57profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.22profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.17profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.50profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.71profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.19profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.55profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.90profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:05,  1.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:01<00:06,  1.00s/profile]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.33profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.19profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.03profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.76profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.28profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.32profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.15profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.79profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.31profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.18profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 11.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:01,  3.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.06profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:01,  3.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  6.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  3.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.05profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.60profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.50profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.92profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.00profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.29profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 11.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.09profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.43profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.11profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.91profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  7.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:02,  2.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.18profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.78profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.75profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.89profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.64profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.05profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.47profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.47profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.65profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:05,  1.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.85profile/s]
+[2026-08-27 04:31:28 PP5 TP1 EP1] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.21profile/s]
+[2026-08-27 04:31:28 PP5 TP0 EP0] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.74profile/s]
+[2026-08-27 04:31:28 PP5 TP2 EP2] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.77profile/s]
+[2026-08-27 04:31:28 PP5 TP3 EP3] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.19profile/s][2026-08-27 04:31:28 PP5 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:28 PP5 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:28 PP5 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:28 PP5 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:28 PP5 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:28 PP5 TP0 EP0] max_total_num_tokens=3801408, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=54, context_len=1048576, available_gpu_mem=16.30 GB
+[2026-08-27 04:31:28 PP5 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:28 PP5 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:28 PP5 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.87profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.90profile/s]
+[2026-08-27 04:31:29 PP4 TP3 EP3] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.87profile/s]
+[2026-08-27 04:31:29 PP4 TP2 EP2] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.93profile/s]
+[2026-08-27 04:31:29 PP4 TP0 EP0] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.43profile/s]
+[2026-08-27 04:31:29 PP4 TP1 EP1] FlashInfer autotune completed.
+[2026-08-27 04:31:29 PP4 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:29 PP4 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:29 PP4 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:29 PP4 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:29 PP4 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:29 PP4 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:29 PP4 TP0 EP0] max_total_num_tokens=3801408, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=54, context_len=1048576, available_gpu_mem=16.30 GB
+[2026-08-27 04:31:29 PP4 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:29 PP4 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+W0827 04:31:30.513926  2527 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50a93b80, len: 6912: Operation not permitted [1]
+W0827 04:31:30.514082  2528 memory_location.cpp:73] Failed to get NUMA node, addr: 0x507df700, len: 6912: Operation not permitted [1]
+W0827 04:31:30.514106  2529 memory_location.cpp:73] Failed to get NUMA node, addr: 0x508aa940, len: 6912: Operation not permitted [1]
+W0827 04:31:30.514214  2530 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50f67bc0, len: 6912: Operation not permitted [1]
+W0827 04:31:30.514272  2531 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50a2f680, len: 55296: Operation not permitted [1]
+W0827 04:31:30.514393  2532 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50afecc0, len: 55296: Operation not permitted [1]
+W0827 04:31:30.514506  2533 memory_location.cpp:73] Failed to get NUMA node, addr: 0x502be580, len: 6912: Operation not permitted [1]
+W0827 04:31:30.514607  2534 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5054d3c0, len: 13824: Operation not permitted [1]
+W0827 04:31:30.514652  2535 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50ad4580, len: 6912: Operation not permitted [1]
+W0827 04:31:30.514743  2536 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50eb9880, len: 6912: Operation not permitted [1]
+W0827 04:31:31.680995  2556 memory_location.cpp:73] Failed to get NUMA node, addr: 0x501c3300, len: 6912: Operation not permitted [1]
+W0827 04:31:31.680985  2557 memory_location.cpp:73] Failed to get NUMA node, addr: 0x504ecec0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.680985  2555 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd4bac0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681169  2558 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50706900, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681272  2559 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5006d700, len: 55296: Operation not permitted [1]
+W0827 04:31:31.681360  2560 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50041cc0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.681468  2561 memory_location.cpp:73] Failed to get NUMA node, addr: 0x506f5a40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681562  2562 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fdfdf40, len: 13824: Operation not permitted [1]
+W0827 04:31:31.681607  2563 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd16080, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681685  2564 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fcfa300, len: 6912: Operation not permitted [1]
+W0827 04:31:31.733497  2583 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51554040, len: 6912: Operation not permitted [1]
+W0827 04:31:31.733552  2584 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5110a8c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.733618  2585 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51ca8e80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.733795  2586 memory_location.cpp:73] Failed to get NUMA node, addr: 0x52017680, len: 6912: Operation not permitted [1]
+W0827 04:31:31.733909  2587 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5174af80, len: 55296: Operation not permitted [1]
+W0827 04:31:31.734004  2588 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51770440, len: 55296: Operation not permitted [1]
+W0827 04:31:31.734061  2589 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51118000, len: 6912: Operation not permitted [1]
+W0827 04:31:31.734179  2590 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51564f40, len: 13824: Operation not permitted [1]
+W0827 04:31:31.734277  2591 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51f69480, len: 6912: Operation not permitted [1]
+W0827 04:31:31.734400  2592 memory_location.cpp:73] Failed to get NUMA node, addr: 0x517c2940, len: 6912: Operation not permitted [1]
+W0827 04:31:31.741586  2611 memory_location.cpp:73] Failed to get NUMA node, addr: 0x519520c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.741609  2612 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50ff8c40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.741679  2613 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51a2ad80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.741792  2614 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51b5b300, len: 6912: Operation not permitted [1]
+W0827 04:31:31.741938  2615 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50fe5b80, len: 55296: Operation not permitted [1]
+W0827 04:31:31.742043  2616 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51625140, len: 55296: Operation not permitted [1]
+W0827 04:31:31.742168  2617 memory_location.cpp:73] Failed to get NUMA node, addr: 0x516bf640, len: 6912: Operation not permitted [1]
+W0827 04:31:31.742278  2618 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51b73e80, len: 13824: Operation not permitted [1]
+W0827 04:31:31.742414  2619 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51ef60c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.742429  2620 memory_location.cpp:73] Failed to get NUMA node, addr: 0x513e95c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.900663  2641 memory_location.cpp:73] Failed to get NUMA node, addr: 0x517e8c80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.900679  2640 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5170b840, len: 6912: Operation not permitted [1]
+W0827 04:31:31.900738  2642 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5131d0c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.900669  2639 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51372400, len: 6912: Operation not permitted [1]
+W0827 04:31:31.900923  2643 memory_location.cpp:73] Failed to get NUMA node, addr: 0x512b7e00, len: 55296: Operation not permitted [1]
+W0827 04:31:31.900976  2644 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51380200, len: 55296: Operation not permitted [1]
+W0827 04:31:31.901000  2646 memory_location.cpp:73] Failed to get NUMA node, addr: 0x513151c0, len: 13824: Operation not permitted [1]
+W0827 04:31:31.901011  2645 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51194f40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.901038  2647 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51520740, len: 6912: Operation not permitted [1]
+W0827 04:31:31.901073  2648 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51724980, len: 6912: Operation not permitted [1]
+W0827 04:31:31.931138  2668 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51061000, len: 6912: Operation not permitted [1]
+W0827 04:31:31.931178  2669 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50d74c00, len: 6912: Operation not permitted [1]
+W0827 04:31:31.931098  2667 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50b2fb40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.931272  2670 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50de9000, len: 6912: Operation not permitted [1]
+W0827 04:31:31.931349  2671 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5097dc80, len: 55296: Operation not permitted [1]
+W0827 04:31:31.931519  2672 memory_location.cpp:73] Failed to get NUMA node, addr: 0x515aa000, len: 55296: Operation not permitted [1]
+W0827 04:31:31.931557  2675 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51269700, len: 6912: Operation not permitted [1]
+W0827 04:31:31.931556  2673 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5097bd00, len: 6912: Operation not permitted [1]
+W0827 04:31:31.931576  2674 memory_location.cpp:73] Failed to get NUMA node, addr: 0x514ee800, len: 13824: Operation not permitted [1]
+W0827 04:31:31.931759  2676 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50deac80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.983201  2696 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50819380, len: 6912: Operation not permitted [1]
+W0827 04:31:31.983201  2695 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50600a80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.983378  2697 memory_location.cpp:73] Failed to get NUMA node, addr: 0x506ddf40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.983438  2698 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50716800, len: 6912: Operation not permitted [1]
+W0827 04:31:31.983490  2699 memory_location.cpp:73] Failed to get NUMA node, addr: 0x501575c0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.983624  2700 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50863340, len: 55296: Operation not permitted [1]
+W0827 04:31:31.983737  2701 memory_location.cpp:73] Failed to get NUMA node, addr: 0x508d7a40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.983930  2702 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50668c40, len: 13824: Operation not permitted [1]
+W0827 04:31:31.984035  2703 memory_location.cpp:73] Failed to get NUMA node, addr: 0x506841c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.984107  2704 memory_location.cpp:73] Failed to get NUMA node, addr: 0x508f16c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.001853  2725 memory_location.cpp:73] Failed to get NUMA node, addr: 0x507c8d40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.001852  2724 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50b39240, len: 6912: Operation not permitted [1]
+W0827 04:31:32.002048  2727 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50a8ecc0, len: 55296: Operation not permitted [1]
+W0827 04:31:32.001888  2726 memory_location.cpp:73] Failed to get NUMA node, addr: 0x512a8140, len: 6912: Operation not permitted [1]
+W0827 04:31:32.002044  2728 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50a68880, len: 55296: Operation not permitted [1]
+W0827 04:31:32.002257  2729 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5087b880, len: 6912: Operation not permitted [1]
+W0827 04:31:32.002264  2731 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50fc6980, len: 6912: Operation not permitted [1]
+W0827 04:31:32.002266  2730 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51331100, len: 13824: Operation not permitted [1]
+W0827 04:31:32.002470  2732 memory_location.cpp:73] Failed to get NUMA node, addr: 0x513384c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.002780  2723 memory_location.cpp:73] Failed to get NUMA node, addr: 0x512c6940, len: 6912: Operation not permitted [1]
+[2026-08-27 04:31:32] Dummy health check server started in background thread at 0.0.0.0:30000
+[rank16]:[W827 04:33:25.972222579 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank17]:[W827 04:33:25.980800100 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank19]:[W827 04:33:25.981413562 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank18]:[W827 04:33:26.644294178 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank16]:[W827 04:34:10.467297472 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank18]:[W827 04:34:10.467837454 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank17]:[W827 04:34:10.468151853 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank19]:[W827 04:34:10.468427857 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank22]:[W827 04:34:10.495396864 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank23]:[W827 04:34:10.495456053 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank21]:[W827 04:34:10.495795969 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank20]:[W827 04:34:10.495814123 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank22]:[W827 04:34:16.703081720 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank20]:[W827 04:34:16.703915664 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank21]:[W827 04:34:16.704033167 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank23]:[W827 04:34:16.704314444 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[2026-08-27 04:35:10 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.29
+[2026-08-27 04:35:10 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.29
+[2026-08-27 04:35:29 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.39
+[2026-08-27 04:35:29 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.39
+[2026-08-27 04:35:31 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25.89
+[2026-08-27 04:35:31 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25.91
+[2026-08-27 04:43:23 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.14
+[2026-08-27 04:43:23 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.14
+[2026-08-27 04:48:13 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.22
+[2026-08-27 04:48:14 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.22
+[2026-08-27 04:48:14 PP4 TP1 EP1] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP4 TP3 EP3] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP4 TP2 EP2] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP4 TP0 EP0] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP5 TP1 EP1] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP5 TP2 EP2] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP5 TP0 EP0] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP5 TP3 EP3] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:57:14 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.12
+[2026-08-27 04:57:14 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.12
+[2026-08-27 05:04:21 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1088, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2.55
+[2026-08-27 05:04:21 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1088, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2.55
+[2026-08-27 05:04:41 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.19
+[2026-08-27 05:04:41 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.19
+[2026-08-27 05:05:39 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1.10
+[2026-08-27 05:05:39 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1.10
+[2026-08-27 05:07:33 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 72.17
+[2026-08-27 05:07:33 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 72.17
+[2026-08-27 05:07:33 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 437051.18
+[2026-08-27 05:07:33 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 375850.16
+[2026-08-27 05:07:33 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25700.77
+[2026-08-27 05:07:33 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33225.98
+[2026-08-27 05:07:40 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8673, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1083.39
+[2026-08-27 05:07:40 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8673, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1083.24
+[2026-08-27 05:07:40 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 481, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 429004.90
+[2026-08-27 05:07:40 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 481, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 414364.03
+[2026-08-27 05:07:40 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 30845.47
+[2026-08-27 05:07:40 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 34555.92
+[2026-08-27 05:07:48 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8656, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1082.66
+[2026-08-27 05:07:48 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8656, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1082.26
+[2026-08-27 05:07:48 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 464, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 397827.41
+[2026-08-27 05:07:48 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 464, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 437968.14
+[2026-08-27 05:07:48 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32533.07
+[2026-08-27 05:07:48 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 31758.27
+[2026-08-27 05:07:55 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8663, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1087.59
+[2026-08-27 05:07:55 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8663, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1087.59
+[2026-08-27 05:07:55 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 471, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 394787.42
+[2026-08-27 05:07:55 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 471, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 405862.61
+[2026-08-27 05:07:55 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32866.57
+[2026-08-27 05:07:55 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32717.90
+[2026-08-27 05:08:03 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8659, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1082.97
+[2026-08-27 05:08:03 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8659, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1083.02
+[2026-08-27 05:08:03 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 467, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 419865.10
+[2026-08-27 05:08:03 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 467, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 410377.15
+[2026-08-27 05:08:03 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33296.73
+[2026-08-27 05:08:03 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 28803.86
+[2026-08-27 05:08:11 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8692, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1085.66
+[2026-08-27 05:08:11 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8692, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1085.97
+[2026-08-27 05:08:11 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 500, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 394417.45
+[2026-08-27 05:08:11 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 500, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 404424.58
+[2026-08-27 05:08:11 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33136.89
+[2026-08-27 05:08:11 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 31089.88
+[2026-08-27 05:08:18 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8557, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1090.03
+[2026-08-27 05:08:18 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8557, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1089.95
+[2026-08-27 05:08:18 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 365, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 409432.91
+[2026-08-27 05:08:18 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 365, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 521781.11
+[2026-08-27 05:08:18 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 30469.13
+[2026-08-27 05:08:18 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 22116.29
+[2026-08-27 05:08:26 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8683, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1083.95
+[2026-08-27 05:08:26 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8683, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1084.25
+[2026-08-27 05:08:26 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 491, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 406971.55
+[2026-08-27 05:08:26 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 491, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 413225.93
+[2026-08-27 05:08:26 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33768.16
+[2026-08-27 05:08:26 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32002.71
+[2026-08-27 05:11:45 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.32
+[2026-08-27 05:11:45 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.32
+[2026-08-27 05:14:42 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1088, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6.16
+[2026-08-27 05:14:42 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1088, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6.16
+[2026-08-27 05:17:47 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 44.22
+[2026-08-27 05:17:47 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 44.22
+[2026-08-27 05:17:47 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 414289.62
+[2026-08-27 05:17:47 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 418259.05
+[2026-08-27 05:17:47 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 31189.29
+[2026-08-27 05:17:47 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 34799.42
+[2026-08-27 05:17:55 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8673, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1086.30
+[2026-08-27 05:17:55 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8673, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1086.07
+[2026-08-27 05:17:55 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 481, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 412548.65
+[2026-08-27 05:17:55 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 481, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 453654.88
+[2026-08-27 05:17:55 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33203.44
+[2026-08-27 05:17:55 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32631.53
+[2026-08-27 05:18:02 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8656, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1088.44
+[2026-08-27 05:18:02 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8656, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1088.37
+[2026-08-27 05:18:02 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 464, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 420591.31
+[2026-08-27 05:18:02 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 464, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 421477.03
+[2026-08-27 05:18:02 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33878.95
+[2026-08-27 05:18:02 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 35064.03
+[2026-08-27 05:18:10 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8663, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1088.01
+[2026-08-27 05:18:10 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8663, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1087.84
+[2026-08-27 05:18:10 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 471, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 429396.76
+[2026-08-27 05:18:10 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 471, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 403808.84
+[2026-08-27 05:18:10 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 30291.78
+[2026-08-27 05:18:10 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 35867.18
+[2026-08-27 05:18:18 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8659, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1086.23
+[2026-08-27 05:18:18 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8659, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1086.07
+[2026-08-27 05:18:18 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 467, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 416829.87
+[2026-08-27 05:18:18 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 467, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 367035.35
+[2026-08-27 05:18:18 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 27946.41
+[2026-08-27 05:18:18 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32311.83
+[2026-08-27 05:18:25 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8692, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1087.22
+[2026-08-27 05:18:25 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8692, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1087.19
+[2026-08-27 05:18:25 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 500, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 454585.15
+[2026-08-27 05:18:25 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 500, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 478466.41
+[2026-08-27 05:18:25 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 34596.06
+[2026-08-27 05:18:25 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 36601.41
+[2026-08-27 05:18:33 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8557, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1090.87
+[2026-08-27 05:18:33 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8557, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1090.67
+[2026-08-27 05:18:33 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 365, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 423632.77
+[2026-08-27 05:18:33 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 365, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 427798.84
+[2026-08-27 05:18:33 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 24938.13
+[2026-08-27 05:18:33 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 26814.84
+[2026-08-27 05:18:40 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8683, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1087.87
+[2026-08-27 05:18:40 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8683, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1087.73
+[2026-08-27 05:18:40 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 491, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 435790.11
+[2026-08-27 05:18:40 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 491, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 437235.07
+[2026-08-27 05:18:40 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33178.64
+[2026-08-27 05:18:40 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 30137.19
+[2026-08-27 05:21:46 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 2, #pending-token: 8652, #bootstrap-req: 2, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 44.04
+[2026-08-27 05:21:47 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 3, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9796.50
+[2026-08-27 05:21:47 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 2, #pending-token: 8652, #bootstrap-req: 2, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 43.85
+[2026-08-27 05:21:48 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 3, #pending-token: 9185, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8786.35
+[2026-08-27 05:21:48 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 3, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8753.58
+[2026-08-27 05:21:49 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 2, #pending-token: 993, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9748.57
+[2026-08-27 05:21:49 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 3, #pending-token: 9185, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9863.29
+[2026-08-27 05:21:50 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 2, #pending-token: 9680, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8707.37
+[2026-08-27 05:21:50 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 2, #pending-token: 993, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8549.77
+[2026-08-27 05:21:51 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1488, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9553.86
+[2026-08-27 05:21:51 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 2, #pending-token: 9680, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9316.43
+[2026-08-27 05:21:52 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1488, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9810.72
+[2026-08-27 05:21:52 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 10199, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8877.35
+[2026-08-27 05:21:52 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 18858, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10683.27
+[2026-08-27 05:21:52 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 10199, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9914.60
+[2026-08-27 05:21:53 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 27591, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9125.55
+[2026-08-27 05:21:53 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 18858, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9121.64
+[2026-08-27 05:21:54 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 53023, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9612.99
+[2026-08-27 05:21:54 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 27591, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9653.85
+[2026-08-27 05:21:55 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 44876, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9352.82
+[2026-08-27 05:21:55 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 53023, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9371.66
+[2026-08-27 05:21:56 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 36684, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9652.74
+[2026-08-27 05:21:56 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 44876, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9597.62
+[2026-08-27 05:21:57 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 28504, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10734.61
+[2026-08-27 05:21:57 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 36684, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9728.96
+[2026-08-27 05:21:57 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 20312, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 15406.33
+[2026-08-27 05:21:57 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 28504, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 18481.89
+[2026-08-27 05:21:57 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 20312, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 294820.71
+[2026-08-27 05:21:57 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 12139, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 258255.44
+[2026-08-27 05:21:57 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 3947, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 296280.34
+[2026-08-27 05:21:57 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 12139, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 273258.91
+[2026-08-27 05:21:57 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 3968, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 41900.16
+[2026-08-27 05:21:57 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 3947, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 83041.18
+[2026-08-27 05:21:58 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 3968, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7314.63
+[2026-08-27 05:22:06 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 0, #pending-token: 8608, #bootstrap-req: 4, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 922.15
+[2026-08-27 05:22:07 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 416, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9857.85
+[2026-08-27 05:22:07 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 0, #pending-token: 8608, #bootstrap-req: 4, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 892.94
+[2026-08-27 05:22:08 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 9149, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8818.94
+[2026-08-27 05:22:08 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 416, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8761.69
+[2026-08-27 05:22:09 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 957, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10110.54
+[2026-08-27 05:22:09 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 9149, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10200.43
+[2026-08-27 05:22:10 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 9638, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8782.22
+[2026-08-27 05:22:10 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 957, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8750.32
+[2026-08-27 05:22:10 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1446, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9843.04
+[2026-08-27 05:22:10 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 9638, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9909.13
+[2026-08-27 05:22:11 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1446, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9535.57
+[2026-08-27 05:22:11 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 10033, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9304.25
+[2026-08-27 05:22:12 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1841, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8648.37
+[2026-08-27 05:22:12 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 10033, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8117.84
+[2026-08-27 05:22:13 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1841, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10198.16
+[2026-08-27 05:22:13 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1856, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2161.72
+[2026-08-27 05:22:14 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8540, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8634.86
+[2026-08-27 05:22:14 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1856, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1928.73
+[2026-08-27 05:22:15 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8540, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9967.52
+[2026-08-27 05:22:15 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 17066, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9942.20
+[2026-08-27 05:22:16 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8910, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8633.09
+[2026-08-27 05:22:16 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 17066, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8614.10
+[2026-08-27 05:22:17 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 17466, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10316.02
+[2026-08-27 05:22:17 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8910, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10031.51
+[2026-08-27 05:22:18 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9324, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9877.92
+[2026-08-27 05:22:18 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 17466, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9424.86
+[2026-08-27 05:22:18 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 17751, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 36929.95
+[2026-08-27 05:22:18 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9324, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 50710.16
+[2026-08-27 05:22:18 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 17751, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 310670.60
+[2026-08-27 05:22:18 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9579, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 230464.10
+[2026-08-27 05:22:18 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1387, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 75748.54
+[2026-08-27 05:22:18 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9579, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 70984.49
+[2026-08-27 05:22:18 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1408, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2600.96
+[2026-08-27 05:22:18 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1387, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 15177.10
+[2026-08-27 05:22:19 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1408, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1879.05
+[2026-08-27 05:22:27 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8547, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 982.50
+[2026-08-27 05:22:28 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 355, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9870.85
+[2026-08-27 05:22:28 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8547, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 973.15
+[2026-08-27 05:22:29 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8932, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8633.06
+[2026-08-27 05:22:29 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 355, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8626.74
+[2026-08-27 05:22:29 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 740, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10223.25
+[2026-08-27 05:22:29 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8932, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10140.49
+[2026-08-27 05:22:30 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 740, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8655.21
+[2026-08-27 05:22:30 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9259, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8412.98
+[2026-08-27 05:22:31 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1067, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10205.64
+[2026-08-27 05:22:31 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9259, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9957.18
+[2026-08-27 05:22:32 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1067, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9648.18
+[2026-08-27 05:22:32 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9745, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9472.33
+[2026-08-27 05:22:33 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1553, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8576.88
+[2026-08-27 05:22:33 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9745, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8156.56
+[2026-08-27 05:22:34 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1600, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1870.62
+[2026-08-27 05:22:34 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1553, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9817.59
+[2026-08-27 05:22:35 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8602, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8886.75
+[2026-08-27 05:22:35 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1600, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1744.17
+[2026-08-27 05:22:36 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 410, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9856.51
+[2026-08-27 05:22:36 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8602, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9924.95
+[2026-08-27 05:22:36 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8696, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8614.79
+[2026-08-27 05:22:36 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 410, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8588.26
+[2026-08-27 05:22:37 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 504, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10212.49
+[2026-08-27 05:22:37 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8696, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10054.15
+[2026-08-27 05:22:38 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9164, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9916.25
+[2026-08-27 05:22:38 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 504, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9469.58
+[2026-08-27 05:22:38 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 972, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 45807.47
+[2026-08-27 05:22:38 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9164, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 65745.12
+[2026-08-27 05:22:38 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 972, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 271786.52
+[2026-08-27 05:22:38 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9322, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 228784.17
+[2026-08-27 05:22:38 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1130, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 76741.41
+[2026-08-27 05:22:38 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9322, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 73462.62
+[2026-08-27 05:22:39 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1152, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2130.27
+[2026-08-27 05:22:39 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1130, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 15175.73
+[2026-08-27 05:22:40 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1152, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1541.83
+[2026-08-27 05:22:47 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8602, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 986.08
+[2026-08-27 05:22:48 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 410, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9910.97
+[2026-08-27 05:22:48 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8602, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 976.33
+[2026-08-27 05:22:49 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 9105, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8736.80
+[2026-08-27 05:22:49 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 410, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8755.48
+[2026-08-27 05:22:50 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 913, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10163.94
+[2026-08-27 05:22:50 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 9105, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10126.04
+[2026-08-27 05:22:51 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9375, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8789.11
+[2026-08-27 05:22:51 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 913, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8802.92
+[2026-08-27 05:22:52 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1183, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9769.42
+[2026-08-27 05:22:52 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9375, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9789.67
+[2026-08-27 05:22:52 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9868, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9680.86
+[2026-08-27 05:22:52 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1183, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9551.84
+[2026-08-27 05:22:53 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1676, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8479.61
+[2026-08-27 05:22:53 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9868, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8580.43
+[2026-08-27 05:22:54 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1676, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9905.15
+[2026-08-27 05:22:54 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2025.45
+[2026-08-27 05:22:55 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8467, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8799.98
+[2026-08-27 05:22:55 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1790.28
+[2026-08-27 05:22:56 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8467, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9998.59
+[2026-08-27 05:22:56 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 275, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9869.57
+[2026-08-27 05:22:57 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8988, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8731.62
+[2026-08-27 05:22:57 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 275, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8714.49
+[2026-08-27 05:22:58 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 796, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10327.91
+[2026-08-27 05:22:58 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8988, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10150.84
+[2026-08-27 05:22:59 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9363, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9995.05
+[2026-08-27 05:22:59 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 796, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9408.40
+[2026-08-27 05:22:59 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 1171, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 30814.55
+[2026-08-27 05:22:59 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9363, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 40670.98
+[2026-08-27 05:22:59 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 1171, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 312400.93
+[2026-08-27 05:22:59 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9880, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 273542.20
+[2026-08-27 05:22:59 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1688, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 81674.04
+[2026-08-27 05:22:59 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9880, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 78852.53
+[2026-08-27 05:23:00 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3180.06
+[2026-08-27 05:23:00 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1688, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 15111.86
+[2026-08-27 05:23:00 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2313.52
+[2026-08-27 05:23:08 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8601, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 987.24
+[2026-08-27 05:23:09 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 409, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9965.41
+[2026-08-27 05:23:09 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8601, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 978.45
+[2026-08-27 05:23:10 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 9104, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8686.72
+[2026-08-27 05:23:10 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 409, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8681.63
+[2026-08-27 05:23:10 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 912, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10219.36
+[2026-08-27 05:23:10 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 9104, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10129.55
+[2026-08-27 05:23:11 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9485, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8726.43
+[2026-08-27 05:23:11 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 912, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8797.95
+[2026-08-27 05:23:12 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1293, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9968.13
+[2026-08-27 05:23:12 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9485, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9970.37
+[2026-08-27 05:23:13 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1293, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9561.18
+[2026-08-27 05:23:13 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9900, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9322.88
+[2026-08-27 05:23:14 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1708, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8660.45
+[2026-08-27 05:23:14 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9900, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8445.20
+[2026-08-27 05:23:15 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1708, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9863.39
+[2026-08-27 05:23:15 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2040.28
+[2026-08-27 05:23:16 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8525, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8723.80
+[2026-08-27 05:23:16 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1809.05
+[2026-08-27 05:23:17 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8525, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9978.38
+[2026-08-27 05:23:17 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 333, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9952.63
+[2026-08-27 05:23:18 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 333, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8701.62
+[2026-08-27 05:23:18 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8909, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8696.74
+[2026-08-27 05:23:18 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 717, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10354.85
+[2026-08-27 05:23:18 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8909, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10184.48
+[2026-08-27 05:23:19 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9452, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10071.93
+[2026-08-27 05:23:19 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 717, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9491.81
+[2026-08-27 05:23:19 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 1260, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33416.21
+[2026-08-27 05:23:19 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9452, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 44390.95
+[2026-08-27 05:23:19 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 1260, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 436736.09
+[2026-08-27 05:23:19 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9830, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 336391.66
+[2026-08-27 05:23:19 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9830, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 339855.11
+[2026-08-27 05:23:19 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1638, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 394477.69
+[2026-08-27 05:23:19 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1638, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 408438.83
+[2026-08-27 05:23:19 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1664, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 78605.25
+[2026-08-27 05:23:19 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1664, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 99592.51
+[2026-08-27 05:32:52 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.11
+[2026-08-27 05:32:52 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.11
+[2026-08-27 05:35:54 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 45.15
+[2026-08-27 05:35:54 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 45.15
+[2026-08-27 05:35:54 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 377056.19
+[2026-08-27 05:35:54 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 398020.47
+[2026-08-27 05:35:54 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 31553.00
+[2026-08-27 05:35:54 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 31134.23
+[2026-08-27 05:36:34 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1.59
+[2026-08-27 05:36:34 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1.59
+[2026-08-27 05:42:33 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 8652, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 22.82
+[2026-08-27 05:42:34 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 3, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9822.04
+[2026-08-27 05:42:34 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 8652, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 22.76
+[2026-08-27 05:42:35 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 3, #pending-token: 9185, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8745.08
+[2026-08-27 05:42:35 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 3, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8645.29
+[2026-08-27 05:42:36 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 2, #pending-token: 993, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9986.78
+[2026-08-27 05:42:36 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 3, #pending-token: 9185, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10117.95
+[2026-08-27 05:42:37 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 2, #pending-token: 9680, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8688.58
+[2026-08-27 05:42:37 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 2, #pending-token: 993, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8650.16
+[2026-08-27 05:42:38 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1488, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10001.69
+[2026-08-27 05:42:38 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 2, #pending-token: 9680, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9962.44
+[2026-08-27 05:42:38 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 10199, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8984.82
+[2026-08-27 05:42:38 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1488, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8923.09
+[2026-08-27 05:42:39 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 18858, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9817.35
+[2026-08-27 05:42:39 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 10199, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9828.78
+[2026-08-27 05:42:40 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 10707, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9100.24
+[2026-08-27 05:42:40 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 18858, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9203.99
+[2026-08-27 05:42:41 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 53023, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9565.34
+[2026-08-27 05:42:41 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 10707, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9596.45
+[2026-08-27 05:42:42 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 44876, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9454.02
+[2026-08-27 05:42:42 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 53023, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9324.98
+[2026-08-27 05:42:43 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 36684, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9959.93
+[2026-08-27 05:42:43 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 44876, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9808.24
+[2026-08-27 05:42:44 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 28504, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10345.51
+[2026-08-27 05:42:44 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 36684, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9684.87
+[2026-08-27 05:42:44 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 20312, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 15470.88
+[2026-08-27 05:42:44 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 28504, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 18125.71
+[2026-08-27 05:42:44 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 20312, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 432980.54
+[2026-08-27 05:42:44 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 12139, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 359018.12
+[2026-08-27 05:42:44 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 3947, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 407919.15
+[2026-08-27 05:42:44 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 12139, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 343399.93
+[2026-08-27 05:42:44 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 3968, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 200008.91
+[2026-08-27 05:42:44 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 3947, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 446241.18
+[2026-08-27 05:42:44 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 3968, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 279949.99
+[2026-08-27 05:43:18 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8608, #bootstrap-req: 1, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 240.35
+[2026-08-27 05:43:18 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8608, #bootstrap-req: 1, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 240.33
+[2026-08-27 05:43:18 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 416, #bootstrap-req: 2, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 442863.57
+[2026-08-27 05:43:18 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 416, #bootstrap-req: 2, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 420812.16
+[2026-08-27 05:43:18 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 448, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 23008.35
+[2026-08-27 05:43:18 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 448, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4452.74
+[2026-08-27 05:43:27 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.06, #running-req: 0, #queue-req: 0, #pending-token: 8701, #bootstrap-req: 4, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 937.67
+[2026-08-27 05:43:28 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.06, #running-req: 0, #queue-req: 0, #pending-token: 509, #bootstrap-req: 4, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8411.92
+[2026-08-27 05:43:28 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.06, #running-req: 0, #queue-req: 0, #pending-token: 8701, #bootstrap-req: 4, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 852.24
+[2026-08-27 05:43:29 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.01, mamba usage: 0.06, #running-req: 0, #queue-req: 4, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 633.17
+[2026-08-27 05:43:29 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.06, #running-req: 0, #queue-req: 0, #pending-token: 509, #bootstrap-req: 4, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9884.35
+[2026-08-27 05:43:30 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 3, #pending-token: 8678, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9522.15
+[2026-08-27 05:43:30 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.01, mamba usage: 0.06, #running-req: 0, #queue-req: 4, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 572.37
+[2026-08-27 05:43:30 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 3, #pending-token: 8678, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 12544.42
+[2026-08-27 05:43:30 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 3, #pending-token: 17239, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9703.75
+[2026-08-27 05:43:31 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 3, #pending-token: 17239, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8904.23
+[2026-08-27 05:43:31 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 2, #pending-token: 9073, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10234.26
+[2026-08-27 05:43:32 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 2, #pending-token: 881, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9944.81
+[2026-08-27 05:43:32 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 2, #pending-token: 9073, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9600.09
+[2026-08-27 05:43:33 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 2, #pending-token: 881, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9693.16
+[2026-08-27 05:43:33 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 896, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1042.71
+[2026-08-27 05:43:34 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 896, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 933.39
+[2026-08-27 05:43:36 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 58625, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2944.07
+[2026-08-27 05:43:37 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 58625, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3078.50
+[2026-08-27 05:43:37 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 50433, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9773.75
+[2026-08-27 05:43:38 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 42277, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8739.12
+[2026-08-27 05:43:38 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 50433, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8709.60
+[2026-08-27 05:43:38 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 34085, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10301.77
+[2026-08-27 05:43:38 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 42277, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9989.54
+[2026-08-27 05:43:39 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 25943, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9684.90
+[2026-08-27 05:43:39 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 34085, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9324.44
+[2026-08-27 05:43:39 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 17751, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 36814.90
+[2026-08-27 05:43:39 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 25943, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 49985.91
+[2026-08-27 05:43:39 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 17751, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 424363.30
+[2026-08-27 05:43:39 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 9579, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 376476.70
+[2026-08-27 05:43:39 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 0, #pending-token: 1387, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 311796.53
+[2026-08-27 05:43:39 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 9579, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 274989.36
+[2026-08-27 05:43:39 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 0, #pending-token: 1387, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 451611.60
+[2026-08-27 05:43:39 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1408, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 64130.72
+[2026-08-27 05:43:39 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1408, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 70140.77
+[2026-08-27 05:43:50 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8547, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 805.37
+[2026-08-27 05:43:50 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8547, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 805.79
+[2026-08-27 05:43:50 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 355, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 441275.22
+[2026-08-27 05:43:50 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 355, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 401841.37
+[2026-08-27 05:43:50 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 21505.07
+[2026-08-27 05:43:50 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 23670.17
+[2026-08-27 05:44:08 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8548, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 436.69
+[2026-08-27 05:44:08 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8548, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 436.67
+[2026-08-27 05:44:08 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 356, #bootstrap-req: 1, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 464575.87
+[2026-08-27 05:44:08 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 356, #bootstrap-req: 1, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 413994.51
+[2026-08-27 05:44:08 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 19485.30
+[2026-08-27 05:44:08 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 21478.12
+[2026-08-27 05:44:16 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8491, #bootstrap-req: 1, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1105.71
+[2026-08-27 05:44:16 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8491, #bootstrap-req: 1, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1030.57
+[2026-08-27 05:44:17 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 299, #bootstrap-req: 5, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7961.03
+[2026-08-27 05:44:17 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 299, #bootstrap-req: 5, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7897.41
+[2026-08-27 05:44:17 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 320, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 1, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 557.73
+[2026-08-27 05:44:18 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 320, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 1, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 506.39
+[2026-08-27 05:44:24 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.06, #running-req: 0, #queue-req: 3, #pending-token: 8657, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1181.25
+[2026-08-27 05:44:25 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.06, #running-req: 0, #queue-req: 3, #pending-token: 8657, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1154.32
+[2026-08-27 05:44:25 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 2, #pending-token: 465, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10317.46
+[2026-08-27 05:44:26 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.01, mamba usage: 0.06, #running-req: 0, #queue-req: 2, #pending-token: 0, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 617.31
+[2026-08-27 05:44:26 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 2, #pending-token: 465, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9762.61
+[2026-08-27 05:44:27 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.01, mamba usage: 0.06, #running-req: 0, #queue-req: 2, #pending-token: 0, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 597.57
+[2026-08-27 05:44:30 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 58376, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2316.27
+[2026-08-27 05:44:30 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 50184, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9835.68
+[2026-08-27 05:44:30 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 58376, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2331.69
+[2026-08-27 05:44:31 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 42030, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8574.76
+[2026-08-27 05:44:31 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 50184, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8573.52
+[2026-08-27 05:44:32 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 33838, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10138.01
+[2026-08-27 05:44:32 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 42030, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9968.30
+[2026-08-27 05:44:33 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 33838, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8609.51
+[2026-08-27 05:44:33 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 25654, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8481.29
+[2026-08-27 05:44:34 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 17462, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10122.75
+[2026-08-27 05:44:34 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 25654, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10060.20
+[2026-08-27 05:44:35 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 9322, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9787.56
+[2026-08-27 05:44:35 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 17462, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9509.72
+[2026-08-27 05:44:36 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1130, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9309.56
+[2026-08-27 05:44:36 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 9322, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9044.44
+[2026-08-27 05:44:36 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1152, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11494.77
+[2026-08-27 05:44:36 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1130, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 174215.73
+[2026-08-27 05:44:36 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8602, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 285270.59
+[2026-08-27 05:44:36 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1152, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 39002.78
+[2026-08-27 05:44:36 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 410, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 355883.67
+[2026-08-27 05:44:36 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8602, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 369490.65
+[2026-08-27 05:44:36 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 448, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 20791.36
+[2026-08-27 05:44:36 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 410, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 398534.60
+[2026-08-27 05:44:36 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 448, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4590.27
+[2026-08-27 05:44:43 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8657, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1137.76
+[2026-08-27 05:44:43 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8657, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1150.97
+[2026-08-27 05:44:43 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 465, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 378568.20
+[2026-08-27 05:44:43 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 465, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 398179.79
+[2026-08-27 05:44:43 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 31349.47
+[2026-08-27 05:44:43 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 30414.26
+[2026-08-27 05:44:52 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8415, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 915.39
+[2026-08-27 05:44:52 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8415, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 915.39
+[2026-08-27 05:44:52 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 223, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 396454.15
+[2026-08-27 05:44:52 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 223, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 392203.52
+[2026-08-27 05:44:52 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 256, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 15256.77
+[2026-08-27 05:44:52 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 256, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 17845.64
+[2026-08-27 05:45:07 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 535.69
+[2026-08-27 05:45:07 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 535.49
+[2026-08-27 05:45:07 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 413632.80
+[2026-08-27 05:45:07 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 451095.61
+[2026-08-27 05:45:07 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 26026.23
+[2026-08-27 05:45:07 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 23636.92
+[2026-08-27 05:45:15 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8467, #bootstrap-req: 4, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1117.01
+[2026-08-27 05:45:15 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8467, #bootstrap-req: 4, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1117.68
+[2026-08-27 05:45:15 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 275, #bootstrap-req: 4, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 491292.57
+[2026-08-27 05:45:15 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 320, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3159.34
+[2026-08-27 05:45:15 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 275, #bootstrap-req: 4, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 78694.24
+[2026-08-27 05:45:15 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 320, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 595.19
+[2026-08-27 05:45:24 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 42247, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 918.09
+[2026-08-27 05:45:25 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 34055, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9798.76
+[2026-08-27 05:45:25 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 42247, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 888.45
+[2026-08-27 05:45:26 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 42692, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8759.63
+[2026-08-27 05:45:26 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 34055, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8750.88
+[2026-08-27 05:45:26 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 34500, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9957.11
+[2026-08-27 05:45:26 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 42692, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9922.60
+[2026-08-27 05:45:27 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 26353, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8703.38
+[2026-08-27 05:45:27 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 34500, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8732.23
+[2026-08-27 05:45:28 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 18161, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10238.25
+[2026-08-27 05:45:28 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 26353, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10063.74
+[2026-08-27 05:45:29 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 26857, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9987.60
+[2026-08-27 05:45:29 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 18161, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9220.37
+[2026-08-27 05:45:29 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 18665, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 23258.84
+[2026-08-27 05:45:29 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 26857, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 30179.73
+[2026-08-27 05:45:30 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 18665, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 24148.98
+[2026-08-27 05:45:30 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 10512, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9875.05
+[2026-08-27 05:45:31 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 10512, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7776.66
+[2026-08-27 05:45:31 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 2320, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9207.39
+[2026-08-27 05:45:32 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 2320, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7381.89
+[2026-08-27 05:45:32 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 2368, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2982.68
+[2026-08-27 05:45:33 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 2368, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2810.80
+[2026-08-27 05:45:36 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.06, #running-req: 0, #queue-req: 0, #pending-token: 8525, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1878.72
+[2026-08-27 05:45:36 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.06, #running-req: 0, #queue-req: 0, #pending-token: 8525, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2311.08
+[2026-08-27 05:45:36 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.04, #running-req: 0, #queue-req: 0, #pending-token: 333, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 373455.20
+[2026-08-27 05:45:36 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.01, mamba usage: 0.04, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3147.78
+[2026-08-27 05:45:36 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.04, #running-req: 0, #queue-req: 0, #pending-token: 333, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 65056.22
+[2026-08-27 05:45:37 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.01, mamba usage: 0.04, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 725.18
+[2026-08-27 05:45:43 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8556, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1156.86
+[2026-08-27 05:45:43 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8556, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1247.36
+[2026-08-27 05:45:43 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 364, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 400455.32
+[2026-08-27 05:45:43 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 364, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 405684.64
+[2026-08-27 05:45:43 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 22719.78
+[2026-08-27 05:45:43 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 23655.83
+[2026-08-27 05:45:52 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8525, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 984.55
+[2026-08-27 05:45:52 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8525, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 984.44
+[2026-08-27 05:45:52 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 333, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 384092.80
+[2026-08-27 05:45:52 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 333, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 391180.99
+[2026-08-27 05:45:52 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 22940.70
+[2026-08-27 05:45:52 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25412.01
+[2026-08-27 05:46:09 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8525, #bootstrap-req: 2, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 485.13
+[2026-08-27 05:46:09 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 333, #bootstrap-req: 2, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 73512.03
+[2026-08-27 05:46:09 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8525, #bootstrap-req: 2, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 482.32
+[2026-08-27 05:46:09 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 333, #bootstrap-req: 2, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 15060.69
+[2026-08-27 05:46:09 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 702.69
+[2026-08-27 05:46:10 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 511.46
+[2026-08-27 05:46:18 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.06, #running-req: 0, #queue-req: 0, #pending-token: 8684, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 991.74
+[2026-08-27 05:46:18 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.04, #running-req: 0, #queue-req: 0, #pending-token: 492, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 113175.52
+[2026-08-27 05:46:18 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.06, #running-req: 0, #queue-req: 0, #pending-token: 8684, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1080.24
+[2026-08-27 05:46:18 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.04, #running-req: 0, #queue-req: 0, #pending-token: 492, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 76392.94
+[2026-08-27 05:46:18 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.04, #running-req: 0, #queue-req: 0, #pending-token: 9062, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 74901.84
+[2026-08-27 05:46:18 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.04, #running-req: 0, #queue-req: 0, #pending-token: 870, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 455065.05
+[2026-08-27 05:46:18 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.04, #running-req: 0, #queue-req: 0, #pending-token: 9062, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 408722.82
+[2026-08-27 05:46:18 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 896, #cached-token: 0, full token usage: 0.01, mamba usage: 0.04, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 49452.06
+[2026-08-27 05:46:18 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.04, #running-req: 0, #queue-req: 0, #pending-token: 870, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 454561.87
+[2026-08-27 05:46:18 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 896, #cached-token: 0, full token usage: 0.01, mamba usage: 0.04, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 45892.86
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-c8-20260827-1340/logs/manual_p_node3.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-c8-20260827-1340/logs/manual_p_node3.log
new file mode 100644
index 0000000..f3f077f
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-c8-20260827-1340/logs/manual_p_node3.log
@@ -0,0 +1,1267 @@
+/usr/local/lib/python3.12/dist-packages/torchao/quantization/quant_api.py:1731: SyntaxWarning: invalid escape sequence '\.'
+  """Configuration class for applying different quantization configs to modules or parameters based on their fully qualified names (FQNs).
+/sgl-workspace/sglang/python/sglang/launch_server.py:56: UserWarning: 'python -m sglang.launch_server' is still supported, but 'sglang serve' is the recommended entrypoint.
+  Example: sglang serve --model-path <model> [options]
+  warnings.warn(
+[2026-08-27 04:28:46] Breakable CUDA graph is incompatible with MLA attention; disabling prefill CUDA graph.
+[2026-08-27 04:28:47] Attention backend not specified. Use triton backend by default.
+[2026-08-27 04:28:47] Pipeline parallelism is incompatible with overlap schedule.
+[2026-08-27 04:28:49] server_args=ServerArgs(model_path='/data/hf_models/Kimi-K3', tokenizer_path='/data/hf_models/Kimi-K3', tokenizer_mode='auto', tokenizer_backend='huggingface', tokenizer_worker_num=1, detokenizer_worker_num=1, skip_tokenizer_init=False, load_format='auto', model_loader_extra_config='{}', trust_remote_code=True, context_length=None, is_embedding=False, enable_multimodal=None, revision=None, model_impl='auto', model_config_parser='auto', json_model_override_args='{}', dtype='auto', quantization=None, quantization_param_path=None, kv_cache_dtype='auto', enable_fp32_lm_head=False, modelopt_quant=None, modelopt_checkpoint_restore_path=None, modelopt_checkpoint_save_path=None, modelopt_export_path=None, quantize_and_serve=False, rl_quant_profile=None, enable_tf32_matmul=False, mem_fraction_static=0.88, max_running_requests=None, max_queued_requests=None, max_total_tokens=None, chunked_prefill_size=8192, enable_dynamic_chunking=False, max_prefill_tokens=16384, prefill_max_requests=None, schedule_policy='fcfs', enable_priority_scheduling=False, disable_priority_preemption=False, default_priority_value=None, abort_on_priority_when_disabled=False, schedule_low_priority_values_first=False, priority_scheduling_preemption_threshold=10, retraction_policy='length', schedule_conservativeness=1.0, page_size=64, swa_full_tokens_ratio=0.8, disable_hybrid_swa_memory=False, radix_eviction_policy='lru', prefill_only_disable_kv_cache=False, disable_radix_cache=True, enable_page_major_kv_layout=False, enable_unified_memory=False, disable_chunked_prefix_cache=False, disable_overlap_schedule=True, num_continuous_decode_steps=1, scheduler_recv_interval=1, enable_mixed_chunk=False, nccl_port=None, dist_timeout=3600, dist_init_addr='174.1.60.1:20000', nnodes=4, node_rank=3, tp_size=4, dcp_size=1, pp_size=8, pp_max_micro_batch_size=None, pp_async_batch_depth=0, dp_size=1, load_balance_method='follow_bootstrap_room', attn_cp_size=1, moe_dp_size=1, dcp_comm_backend='ag_rs', dcp_replicate_q_proj=None, dwdp_size=1, enable_prefill_cp=False, cp_strategy=None, enable_dsa_cache_layer_split=False, enable_dsa_prefill_context_parallel=False, dsa_prefill_cp_mode='round-robin-split', enable_prefill_context_parallel=False, prefill_cp_mode='in-seq-split', enable_cp_decode_attn_tp=False, enable_dp_attention=False, enable_dp_attention_local_control_broadcast=False, enable_dp_lm_head=False, enable_attn_tp_input_scattered=False, disable_attn_tp_gather=False, enable_p2p_check=False, device='cuda', base_gpu_id=0, gpu_id_step=1, random_seed=927276655, watchdog_timeout=300, soft_watchdog_timeout=None, sleep_on_idle=False, use_ray=False, custom_sigquit_handler=None, numa_node=None, gc_threshold=None, host='0.0.0.0', port=30000, fastapi_root_path='', smg_grpc_mode=False, grpc_mode=False, grpc_port=None, sidecar=None, sidecar_args=None, skip_server_warmup=False, warmups=None, enable_http2=False, ssl_keyfile=None, ssl_certfile=None, ssl_ca_certs=None, ssl_keyfile_password=None, enable_ssl_refresh=False, api_key=None, admin_api_key=None, served_model_name='kimi-k3', weight_version='default', chat_template=None, hf_chat_template_name=None, completion_template=None, file_storage_path='sglang_storage', enable_cache_report=False, reasoning_parser=None, default_chat_template_kwargs=None, strip_thinking_cache=False, enable_strict_thinking=False, tool_call_parser=None, tool_server=None, sampling_defaults='model', asr_max_buffer_seconds=60, asr_max_concurrent_sessions=32, preferred_sampling_params=None, allow_auto_truncate=False, stream_interval=1, batch_notify_size=16, stream_response_default_include_usage=False, incremental_streaming_output=False, enable_streaming_session=False, enable_session_radix_cache=False, log_level='info', log_level_http=None, log_requests=False, log_requests_level=2, log_requests_format='text', log_requests_target=None, uvicorn_access_log_exclude_prefixes=[], crash_dump_folder=None, show_time_cost=False, enable_metrics=False, smg_http_sidecar_port=None, enable_mfu_metrics=False, enable_metrics_for_all_schedulers=False, load_snapshot_publish_interval=15, tokenizer_metrics_custom_labels_header='x-custom-labels', tokenizer_metrics_allowed_custom_labels=None, extra_metric_labels=None, bucket_time_to_first_token=None, bucket_inter_token_latency=None, bucket_e2e_request_latency=None, prompt_tokens_buckets=None, generation_tokens_buckets=None, gc_warning_threshold_secs=0.0, decode_log_interval=40, enable_request_time_stats_logging=False, kv_events_config=None, enable_forward_pass_metrics=False, forward_pass_metrics_worker_id='', forward_pass_metrics_ipc_name=None, enable_trace=False, trace_modules='request', otlp_traces_endpoint='localhost:4317', export_metrics_to_file=False, export_metrics_to_file_dir=None, stat_loggers=None, constrained_json_whitespace_pattern=None, constrained_json_disable_any_whitespace=False, attention_backend='triton', decode_attention_backend=None, prefill_attention_backend=None, sampling_backend='flashinfer', grammar_backend='xgrammar', radix_cache_backend=None, mm_attention_backend=None, fp8_gemm_runner_backend='auto', fp4_gemm_runner_backend='auto', bf16_gemm_backend='auto', dsa_prefill_backend=None, dsa_decode_backend=None, dsa_paged_mqa_logits_backend='auto', dsa_topk_backend='sgl-kernel', disable_flashinfer_autotune=False, flashinfer_autotune_skip_ops=None, mamba_backend='triton', cuda_graph_config=CudaGraphConfig(decode=PhaseConfig(backend='disabled', max_bs=16, bs=[1, 2, 4, 8, 12, 16], tc_compiler='eager', full_prefill_max_req=None), prefill=PhaseConfig(backend='disabled', max_bs=2048, bs=[4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], tc_compiler='eager', full_prefill_max_req=None)), cuda_graph_backend_decode=None, cuda_graph_backend_prefill=None, cuda_graph_max_bs_decode=16, cuda_graph_max_bs_prefill=None, cuda_graph_bs_decode=None, cuda_graph_bs_prefill=None, cuda_graph_tc_compiler=None, disable_prefill_cuda_graph=False, disable_decode_cuda_graph=False, disable_cuda_graph=False, disable_cuda_graph_padding=False, enable_profile_cuda_graph=False, enable_cudagraph_gc=False, debug_cuda_graph=False, enable_layerwise_nvtx_marker=False, enable_nccl_nvls=False, enable_symm_mem=False, triton_attention_reduce_in_fp32=False, triton_attention_num_kv_splits=8, triton_attention_split_tile_size=None, flashinfer_mla_disable_ragged=False, enable_fused_qk_norm_rope=False, enable_precise_embedding_interpolation=False, enable_fused_moe_sum_all_reduce=False, enable_deepseek_v4_fp4_indexer=False, disable_custom_all_reduce=False, enable_mscclpp=False, enable_torch_symm_mem=False, enable_scattered_sconv=False, pre_warm_nccl=False, enable_quant_communications=False, enable_flashinfer_allreduce_fusion=False, enforce_disable_flashinfer_allreduce_fusion=False, flashinfer_allreduce_fusion_backend=None, enable_aiter_allreduce_fusion=False, enable_torch_compile=False, enable_torch_compile_debug_mode=False, torch_compile_max_bs=32, torchao_config='', speculative_algorithm=None, speculative_draft_model_path=None, speculative_draft_model_revision=None, speculative_draft_load_format=None, speculative_num_steps=None, speculative_eagle_topk=None, speculative_num_draft_tokens=None, speculative_dflash_block_size=None, speculative_dspark_block_size=None, speculative_dspark_sps_table_path=None, speculative_dspark_confidence_sts_path=None, speculative_dspark_align_verify_tokens_to_graph_tier=False, speculative_accept_threshold_single=1.0, speculative_accept_threshold_acc=1.0, speculative_use_rejection_sampling=False, speculative_token_map=None, speculative_attention_mode='prefill', speculative_draft_attention_backend=None, speculative_draft_window_size=None, speculative_moe_runner_backend='flashinfer_mxfp4', speculative_moe_a2a_backend=None, speculative_draft_model_quantization=None, speculative_skip_dp_mlp_sync=False, enable_multi_layer_eagle=False, speculative_adaptive=False, speculative_adaptive_config=None, decoupled_spec_bind_endpoint=None, decoupled_spec_connect_endpoints=None, decoupled_spec_rank=None, decoupled_spec_role='null', spec_trace_dir=None, speculative_ngram_min_bfs_breadth=1, speculative_ngram_max_bfs_breadth=10, speculative_ngram_match_type='BFS', speculative_ngram_max_trie_depth=18, speculative_ngram_capacity=10000000, speculative_ngram_external_corpus_path=None, speculative_ngram_external_sam_budget=0, speculative_ngram_external_corpus_max_tokens=10000000, ep_size=4, moe_a2a_backend='none', moe_runner_backend='flashinfer_mxfp4', flashinfer_mxfp4_moe_precision='default', deepep_mode='auto', fuseep_mode=2, deepep_dispatcher_output_dtype='auto', ep_num_redundant_experts=0, ep_dispatch_algorithm=None, init_expert_location='trivial', enable_eplb=False, eplb_algorithm='auto', eplb_rebalance_num_iterations=1000, eplb_rebalance_layers_per_chunk=None, eplb_min_rebalancing_utilization_threshold=1.0, expert_distribution_recorder_mode=None, expert_distribution_recorder_buffer_size=1000, enable_expert_distribution_metrics=False, deepep_config=None, moe_dense_tp_size=None, elastic_ep_backend=None, enable_elastic_expert_backup=False, mooncake_ib_device=None, enable_waterfill=False, ep_join_mode=None, ep_join_rank_offset=0, elastic_ep_initial_size=None, max_ep_size=None, elastic_ep_scale_timeout=600, elastic_ep_rejoin=False, disable_flashinfer_cutlass_moe_fp4_allgather=False, disable_shared_experts_fusion=False, enforce_shared_experts_fusion=False, max_mamba_cache_size=None, mamba_ssm_dtype=None, mamba_max_states_per_path=-1, enable_mamba_cache_stochastic_rounding=False, mamba_cache_philox_rounds=0, mamba_full_memory_ratio=0.36, mamba_radix_cache_strategy='extra_buffer_lazy', uses_mamba_radix_cache=False, mamba_track_interval=256, enable_int8_mamba_checkpoint=False, int8_mamba_ckpt_size=None, linear_attn_backend='triton', linear_attn_decode_backend=None, linear_attn_prefill_backend=None, linear_attn_verify_backend=None, enable_linear_replayssm=False, linear_replayssm_cache_len=16, enable_linear_replayssm_spec=False, enable_hierarchical_cache=False, hicache_ratio=2.0, hicache_size=0, hicache_write_policy='write_through', hicache_io_backend='kernel', hicache_mem_layout='page_first', hicache_storage_backend=None, hicache_storage_prefetch_policy='timeout', hicache_storage_backend_extra_config=None, enable_hisparse=False, hisparse_config=None, enable_broadcast_mm_inputs_process=False, enable_prefix_mm_cache=False, mm_enable_dp_encoder=False, mm_process_config={}, mm_processor_worker_num=0, mm_io_worker_num=0, limit_mm_data_per_request=None, enable_mm_global_cache=False, disable_fast_image_processor=False, mm_feature_transport='cpu', keep_mm_feature_on_device=False, enable_lora=None, enable_lora_overlap_loading=None, max_lora_rank=None, lora_target_modules=None, lora_paths=None, max_loaded_loras=None, max_loras_per_batch=8, lora_eviction_policy='lru', lora_backend='csgmv', max_lora_chunk_size=16, experts_shared_outer_loras=None, lora_use_virtual_experts=False, lora_strict_loading=False, lora_drain_wait_threshold=0.0, enable_two_batch_overlap=False, enable_single_batch_overlap=False, tbo_token_distribution_threshold=0.48, cpu_offload_gb=0, offload_group_size=-1, offload_num_in_group=1, offload_prefetch_step=1, offload_mode='cpu', enable_lmcache=False, lmcache_config_file=None, enable_flexkv=False, flexkv_config_file=None, kt_weight_path=None, kt_method='AMXINT4', kt_cpuinfer=None, kt_threadpool_count=2, kt_num_gpu_experts=None, kt_max_deferred_experts_per_token=None, dllm_algorithm=None, dllm_algorithm_config=None, dllm_fdfo=True, disaggregation_mode='prefill', disaggregation_transfer_backend='mooncake', disaggregation_bootstrap_port=28800, disaggregation_ib_device='mlx5_0,mlx5_1,mlx5_2,mlx5_3', disaggregation_decode_enable_radix_cache=False, disaggregation_decode_enable_offload_kvcache=False, num_reserved_decode_tokens=512, disaggregation_decode_extra_slots=None, disaggregation_decode_polling_interval=1, optimistic_prefill_attempts=0, encoder_only=False, language_only=False, encoder_transfer_backend='zmq_to_tokenizer', encoder_urls=[], encoder_bootstrap_port=8997, encoder_register_urls=[], enable_adaptive_dispatch_to_encoder=False, enable_pdmux=False, pdmux_config_path=None, sm_group_num=8, custom_weight_loader=[], weight_loader_disable_mmap=False, weight_loader_prefetch_checkpoints=False, weight_loader_prefetch_num_threads=4, weight_loader_drop_cache_after_load=False, remote_instance_weight_loader_seed_instance_ip=None, remote_instance_weight_loader_seed_instance_service_port=None, remote_instance_weight_loader_send_weights_group_ports=None, remote_instance_weight_loader_backend='nccl', remote_instance_weight_loader_start_seed_via_transfer_engine=False, engine_info_bootstrap_port=6789, modelexpress_config=None, download_dir=None, model_checksum=None, delete_ckpt_after_loading=False, decrypted_config_file=None, decrypted_draft_config_file=None, checkpoint_engine_wait_weights_before_ready=False, enable_prefill_delayer=False, prefill_delayer_max_delay_passes=30, prefill_delayer_token_usage_low_watermark=None, prefill_delayer_forward_passes_buckets=None, prefill_delayer_wait_seconds_buckets=None, prefill_delayer_queue_min_ratio=None, prefill_delayer_max_delay_ms=None, min_free_slots_delay=None, enable_deterministic_inference=False, rl_on_policy_target=None, kv_canary='none', kv_canary_real_data='none', kv_canary_sweep_interval=0, enable_dynamic_batch_tokenizer=False, dynamic_batch_tokenizer_batch_size=32, dynamic_batch_tokenizer_batch_timeout=0.002, enable_tokenizer_batch_encode=False, disable_tokenizer_batch_decode=False, debug_tensor_dump_output_folder=None, debug_tensor_dump_layers=None, debug_tensor_dump_input_file=None, enable_memory_saver=False, enable_weights_cpu_backup=False, enable_draft_weights_cpu_backup=False, enable_custom_logit_processor=False, enable_return_hidden_states=False, enable_return_routed_experts=False, enable_return_indexer_topk=False, disable_outlines_disk_cache=False, enable_mis=False, weight_cache_mode='off', weight_cache_socket=None, weight_cache_timeout=1800, forward_hooks=None, msprobe_dump_config=None)
+[2026-08-27 04:28:49] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:49] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:49] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:49] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:49] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:49] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:49] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:49] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:05 PP6 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:05 PP6 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:07 PP6 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:07 PP6 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:07 PP6 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:07 PP7 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:07 PP7 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:07 PP7 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:07 PP7 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:07 PP6 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:07 PP6 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:07.662256   238 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:07.662379   238 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:07.662392   238 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.4 port: 12001
+I0827 04:29:07.662429   238 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.4:15914
+I0827 04:29:07.662514   238 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:07.685189   238 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:07.689950   238 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:07.695539   238 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:07.701294   238 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:07.703691   238 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:07.703715   238 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:07.703760   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:07.710492   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:07.711953   238 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:04
+I0827 04:29:07.711989   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:07.735716   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:07.737195   238 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:04
+I0827 04:29:07.737223   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:07.763653   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:07.765146   238 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:04
+I0827 04:29:07.765172   238 rdma_context.cpp:205] Using SIEVE endpoint store
+[2026-08-27 04:29:07 PP7 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:07 PP7 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:07 PP7 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:07 PP7 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+I0827 04:29:07.791589   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:07.793025   238 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:04
+I0827 04:29:07.793139   238 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:07 PP6 TP3 EP3] Init torch distributed begin.
+[2026-08-27 04:29:07 PP6 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:07 PP6 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:08 PP6 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:08.916008   235 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:08.916072   235 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:08.916082   235 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.4 port: 12001
+I0827 04:29:08.916121   235 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.4:15737
+I0827 04:29:08.916215   235 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:08.941840   235 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:08.950320   235 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:08.979278   235 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:08.988204   235 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:08.991855   235 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:08.991879   235 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:08.991928   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:08.999985   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:09.001416   235 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:04
+I0827 04:29:09.001446   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:09.027477   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:09.028954   235 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:04
+I0827 04:29:09.028980   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:09.055524   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:09.056981   235 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:04
+I0827 04:29:09.057020   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:09.083544   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:09.084911   235 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:04
+I0827 04:29:09.085031   235 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:09 PP6 TP0 EP0] Init torch distributed begin.
+[2026-08-27 04:29:09 PP7 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:09 PP6 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:09 PP7 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:09 PP6 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:09 PP7 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:09 PP7 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:10.000486   242 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:10.000609   242 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:10.000628   242 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.4 port: 12001
+I0827 04:29:10.000689   242 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.4:16890
+I0827 04:29:10.000851   242 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:10.006592   236 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:10.006711   236 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:10.006728   236 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.4 port: 12001
+I0827 04:29:10.006786   236 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.4:15893
+I0827 04:29:10.006963   236 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:10.022238   241 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:10.022354   241 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:10.022370   241 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.4 port: 12001
+I0827 04:29:10.022423   241 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.4:16277
+I0827 04:29:10.022593   241 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:10.032052   242 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:10.032101   236 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:10.039772   236 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:10.039858   242 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:10.047292   236 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:10.047341   242 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:10.047785   241 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:10.054806   242 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:10.054838   236 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:10.055280   241 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:10.059140   236 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:10.059171   236 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:10.059175   242 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:10.059201   242 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:10.059227   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.059260   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.062610   241 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:10.066787   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:10.066826   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:10.068500   242 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:04
+I0827 04:29:10.068532   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.068558   236 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:04
+I0827 04:29:10.068590   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.069866   241 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:10.072566   241 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:10.072592   241 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:10.072649   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.080149   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:10.081809   241 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:04
+I0827 04:29:10.081843   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.095701   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:10.097368   236 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:04
+I0827 04:29:10.097400   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.099974   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:10.101672   242 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:04
+I0827 04:29:10.101717   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.107705   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:10.109316   241 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:04
+I0827 04:29:10.109347   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.123868   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:10.125504   236 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:04
+I0827 04:29:10.125536   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.127957   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:10.129673   242 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:04
+I0827 04:29:10.129703   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.139813   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:10.141404   241 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:04
+I0827 04:29:10.141431   241 rdma_context.cpp:205] Using SIEVE endpoint store
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:10.150050   237 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:10.150159   237 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:10.150173   237 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.4 port: 12001
+I0827 04:29:10.150219   237 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.4:15413
+I0827 04:29:10.150358   237 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:10.151880   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:10.153475   236 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:04
+I0827 04:29:10.153601   236 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:10 PP6 TP1 EP1] Init torch distributed begin.
+I0827 04:29:10.155849   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:10.155864   237 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:10.157462   242 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:04
+I0827 04:29:10.157585   242 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:10 PP7 TP3 EP3] Init torch distributed begin.
+I0827 04:29:10.171923   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:10.172516   240 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:10.172667   240 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:10.172688   240 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.4 port: 12001
+I0827 04:29:10.172762   240 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.4:15705
+I0827 04:29:10.172977   240 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:10.173573   241 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:04
+I0827 04:29:10.173692   241 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:10 PP7 TP2 EP2] Init torch distributed begin.
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:10.177717   239 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:10.177865   239 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:10.177886   239 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.4 port: 12001
+I0827 04:29:10.177956   239 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.4:16192
+I0827 04:29:10.178146   239 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:10.183638   237 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:10.187090   239 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:10.191001   237 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:10.195258   239 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:10.198271   237 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:10.200984   237 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:10.201025   237 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:10.201079   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.203014   239 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:10.203156   240 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:10.209589   240 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:10.210305   239 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:10.212793   239 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:10.212821   239 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:10.212878   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.216092   240 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:10.219946   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:10.221297   239 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:04
+I0827 04:29:10.221323   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.222187   240 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:10.224284   240 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:10.224304   240 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:10.224344   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.230640   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:10.231778   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:10.231937   240 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:04
+I0827 04:29:10.231962   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.233362   237 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:04
+I0827 04:29:10.233397   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.236478   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:10.237651   240 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:04
+I0827 04:29:10.237670   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.250197   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:10.251711   239 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:04
+I0827 04:29:10.251737   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.263743   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:10.265311   237 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:04
+I0827 04:29:10.265339   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.267139   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:10.268607   240 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:04
+I0827 04:29:10.268632   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.273075   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:10.274682   237 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:04
+I0827 04:29:10.274710   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.276602   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:10.278275   239 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:04
+I0827 04:29:10.278314   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.295709   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:10.297385   240 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:04
+I0827 04:29:10.297510   240 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:10 PP7 TP1 EP1] Init torch distributed begin.
+I0827 04:29:10.299893   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:10.301563   237 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:04
+I0827 04:29:10.301689   237 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:10 PP6 TP2 EP2] Init torch distributed begin.
+I0827 04:29:10.303737   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:10.305307   239 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:04
+I0827 04:29:10.305433   239 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:10 PP7 TP0 EP0] Init torch distributed begin.
+[2026-08-27 04:29:26 PP7 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:26 PP6 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:27 PP6 TP0 EP0] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP6 TP3 EP3] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP6 TP2 EP2] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP6 TP1 EP1] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP7 TP1 EP1] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP7 TP0 EP0] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP7 TP3 EP3] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP7 TP2 EP2] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP7 TP1 EP1] Init torch distributed ends. elapsed=17.45 s, mem usage=0.46 GB
+[2026-08-27 04:29:27 PP7 TP3 EP3] Init torch distributed ends. elapsed=17.59 s, mem usage=0.46 GB
+[2026-08-27 04:29:27 PP7 TP2 EP2] Init torch distributed ends. elapsed=17.57 s, mem usage=0.46 GB
+[2026-08-27 04:29:27 PP7 TP0 EP0] Init torch distributed ends. elapsed=17.44 s, mem usage=0.46 GB
+[2026-08-27 04:29:27 PP6 TP1 EP1] Init torch distributed ends. elapsed=17.59 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP6 TP3 EP3] Init torch distributed ends. elapsed=19.95 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP6 TP0 EP0] Init torch distributed ends. elapsed=18.66 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP6 TP2 EP2] Init torch distributed ends. elapsed=17.44 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP6 TP1 EP1] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP7 TP1 EP1] Load weight begin. avail mem=82.12 GB
+[2026-08-27 04:29:27 PP6 TP0 EP0] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP6 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP7 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP7 TP3 EP3] Load weight begin. avail mem=82.12 GB
+[2026-08-27 04:29:27 PP6 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP7 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP6 TP2 EP2] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP6 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP7 TP2 EP2] Load weight begin. avail mem=82.12 GB
+[2026-08-27 04:29:28 PP7 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP6 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP7 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP6 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP7 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP7 TP0 EP0] Load weight begin. avail mem=82.12 GB
+[2026-08-27 04:29:28 PP6 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP7 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP7 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP6 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP7 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP6 TP3 EP3] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP6 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP6 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP6 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP7 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP7 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP7 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP6 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP7 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP6 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:30:40 PP7 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:40 PP7 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:40 PP7 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:40 PP7 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:40 PP7 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:40 PP7 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:40 PP7 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:40 PP7 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:41 PP7 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:41 PP7 TP1 EP1] Load weight end. elapsed=73.17 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=31.91 GB, mem usage=50.21 GB.
+[2026-08-27 04:30:41 PP7 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:41 PP7 TP0 EP0] Load weight end. elapsed=73.80 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=31.91 GB, mem usage=50.21 GB.
+[2026-08-27 04:30:42 PP6 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:42 PP6 TP1 EP1] Load weight end. elapsed=74.47 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:45 PP6 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:45 PP6 TP2 EP2] Load weight end. elapsed=77.20 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:50 PP7 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:50 PP7 TP3 EP3] Load weight end. elapsed=82.79 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=31.91 GB, mem usage=50.21 GB.
+[2026-08-27 04:30:51 PP6 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:51 PP6 TP3 EP3] Load weight end. elapsed=83.19 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:51 PP7 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:51 PP7 TP2 EP2] Load weight end. elapsed=83.68 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=31.91 GB, mem usage=50.21 GB.
+[2026-08-27 04:30:51 PP7 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:51 PP7 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:52 PP7 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:52 PP7 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:52 PP7 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:52 PP7 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:52 PP7 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:52 PP7 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:53 PP6 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:53 PP6 TP0 EP0] Load weight end. elapsed=85.47 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:53 PP6 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:53 PP6 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:53 PP6 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:53 PP6 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:53 PP6 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:53 PP6 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:53 PP6 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:53 PP6 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:54 PP7 TP3 EP3] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP7 TP2 EP2] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP7 TP1 EP1] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP7 TP0 EP0] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP6 TP3 EP3] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP6 TP2 EP2] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP6 TP1 EP1] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP6 TP0 EP0] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP6 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP6 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP7 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:54 PP6 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP7 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:54 PP7 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:54 PP7 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:54 PP6 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP6 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP6 TP3 EP3] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP6 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP6 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP6 TP2 EP2] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP6 TP1 EP1] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP7 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 16.31 GB
+[2026-08-27 04:30:54 PP7 TP2 EP2] Memory pool end. avail mem=14.95 GB
+[2026-08-27 04:30:54 PP6 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP6 TP0 EP0] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP7 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 16.31 GB
+[2026-08-27 04:30:54 PP7 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 16.31 GB
+[2026-08-27 04:30:54 PP7 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 16.31 GB
+[2026-08-27 04:30:54 PP7 TP3 EP3] Memory pool end. avail mem=14.95 GB
+[2026-08-27 04:30:54 PP7 TP0 EP0] Memory pool end. avail mem=14.95 GB
+[2026-08-27 04:30:54 PP7 TP1 EP1] Memory pool end. avail mem=14.95 GB
+[2026-08-27 04:30:54 PP7 TP2 EP2] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp2_pp7_dp0.json
+[2026-08-27 04:30:54 PP6 TP3 EP3] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp3_pp6_dp0.json
+[2026-08-27 04:30:54 PP6 TP1 EP1] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp1_pp6_dp0.json
+[2026-08-27 04:30:54 PP7 TP3 EP3] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp3_pp7_dp0.json
+[2026-08-27 04:30:54 PP6 TP2 EP2] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp2_pp6_dp0.json
+[2026-08-27 04:30:54 PP6 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:30:54 PP6 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:30:54 PP6 TP0 EP0] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp0_pp6_dp0.json
+[2026-08-27 04:30:54 PP7 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:30:54 PP7 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:30:54 PP7 TP0 EP0] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp0_pp7_dp0.json
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[2026-08-27 04:30:54 PP7 TP1 EP1] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp1_pp7_dp0.json
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 11.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 13.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.76profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.62profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.66profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.72profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.68profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.55profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.12profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  6.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  7.16profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.92profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.15profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.59profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.56profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 19.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  8.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.67profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.40profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.88profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  8.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  6.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.83profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.50profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.32profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.92profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.77profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.94profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.57profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 12.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 12.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 13.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.36profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.66profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  6.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.72profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  6.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.22profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.65profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.13profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.03profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.46profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.59profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.16profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  6.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.96profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.76profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.67profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  8.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.27profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.88profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.01profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.57profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.12profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 11.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  2.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.68profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.90profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.43profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.92profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  8.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.66profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  6.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  7.13profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.32profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.71profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:05,  1.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.16profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.31profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.77profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.28profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.48profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.87profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.40profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 13.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.35profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.98profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.88profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.73profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  3.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.88profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.09profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.20profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.97profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.83profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.47profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.40profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.60profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  8.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  7.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.97profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.05profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.05profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  6.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.49profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.72profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.19profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 10.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.91profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.64profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.58profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  3.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.35profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  9.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 12.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.09profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:02,  2.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 12.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.15profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.76profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.49profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.03profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.79profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.96profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.47profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.65profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.55profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.03profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  9.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.63profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.98profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:05,  1.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:01,  3.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.50profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  6.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.94profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.13profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.78profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.53profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.35profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.56profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.36profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:01<00:06,  1.06s/profile]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  8.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.91profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.66profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  8.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  7.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:02,  2.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.83profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.01profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.66profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.13profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.59profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  8.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  8.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.71profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.29profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.97profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.88profile/s]
+[2026-08-27 04:31:27 PP6 TP3 EP3] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.86profile/s]
+[2026-08-27 04:31:27 PP6 TP1 EP1] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.24profile/s]
+[2026-08-27 04:31:27 PP6 TP2 EP2] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.33profile/s]
+[2026-08-27 04:31:27 PP6 TP0 EP0] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.56profile/s][2026-08-27 04:31:27 PP6 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:27 PP6 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:27 PP6 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:27 PP6 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:27 PP6 TP0 EP0] max_total_num_tokens=3801408, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=54, context_len=1048576, available_gpu_mem=16.52 GB
+[2026-08-27 04:31:27 PP6 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:27 PP6 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:27 PP6 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:27 PP6 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
+[2026-08-27 04:31:28 PP7 TP2 EP2] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.32profile/s]
+[2026-08-27 04:31:28 PP7 TP3 EP3] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.60profile/s]
+[2026-08-27 04:31:28 PP7 TP0 EP0] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.60profile/s]
+[2026-08-27 04:31:28 PP7 TP1 EP1] FlashInfer autotune completed.
+[2026-08-27 04:31:28 PP7 TP2 EP2] multimem all-gather disabled (no multicast for world_size=4)
+[2026-08-27 04:31:28 PP7 TP3 EP3] multimem all-gather disabled (no multicast for world_size=4)
+[2026-08-27 04:31:28 PP7 TP1 EP1] multimem all-gather disabled (no multicast for world_size=4)
+[2026-08-27 04:31:28 PP7 TP0 EP0] multimem all-gather disabled (no multicast for world_size=4)
+[2026-08-27 04:31:28 PP7 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:28 PP7 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:28 PP7 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:28 PP7 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:28 PP7 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:28 PP7 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:28 PP7 TP0 EP0] max_total_num_tokens=3801408, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=54, context_len=1048576, available_gpu_mem=12.14 GB
+[2026-08-27 04:31:28 PP7 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:28 PP7 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+W0827 04:31:31.959518  2596 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5efc8ec0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.959637  2598 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ee654c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.959640  2597 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5eda3c80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.959542  2595 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5efc6f80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.959741  2599 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5eeaf500, len: 55296: Operation not permitted [1]
+W0827 04:31:31.959794  2600 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ef4e380, len: 55296: Operation not permitted [1]
+W0827 04:31:31.959827  2601 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5f548500, len: 6912: Operation not permitted [1]
+W0827 04:31:31.959918  2602 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5eec3bc0, len: 13824: Operation not permitted [1]
+W0827 04:31:31.960085  2603 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5f01bcc0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.960222  2604 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5f4c8ac0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.980162  2623 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60ba35c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.980152  2624 memory_location.cpp:73] Failed to get NUMA node, addr: 0x609ccf40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.980242  2625 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60d38600, len: 6912: Operation not permitted [1]
+W0827 04:31:31.980361  2626 memory_location.cpp:73] Failed to get NUMA node, addr: 0x609546c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.980516  2628 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60c49a00, len: 55296: Operation not permitted [1]
+W0827 04:31:31.980537  2627 memory_location.cpp:73] Failed to get NUMA node, addr: 0x607c6800, len: 55296: Operation not permitted [1]
+W0827 04:31:31.980547  2629 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60d11800, len: 6912: Operation not permitted [1]
+W0827 04:31:31.980834  2630 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60a1b740, len: 13824: Operation not permitted [1]
+W0827 04:31:31.981053  2632 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60ce5000, len: 6912: Operation not permitted [1]
+W0827 04:31:31.981062  2631 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60aef300, len: 6912: Operation not permitted [1]
+W0827 04:31:32.117022  2651 memory_location.cpp:73] Failed to get NUMA node, addr: 0x74c79d40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.117132  2652 memory_location.cpp:73] Failed to get NUMA node, addr: 0x74f99700, len: 6912: Operation not permitted [1]
+W0827 04:31:32.117317  2653 memory_location.cpp:73] Failed to get NUMA node, addr: 0x7546c740, len: 6912: Operation not permitted [1]
+W0827 04:31:32.117331  2654 memory_location.cpp:73] Failed to get NUMA node, addr: 0x74f4f140, len: 6912: Operation not permitted [1]
+W0827 04:31:32.117414  2656 memory_location.cpp:73] Failed to get NUMA node, addr: 0x750037c0, len: 55296: Operation not permitted [1]
+W0827 04:31:32.117445  2655 memory_location.cpp:73] Failed to get NUMA node, addr: 0x74bf1200, len: 55296: Operation not permitted [1]
+W0827 04:31:32.117705  2657 memory_location.cpp:73] Failed to get NUMA node, addr: 0x74f80080, len: 6912: Operation not permitted [1]
+W0827 04:31:32.117738  2658 memory_location.cpp:73] Failed to get NUMA node, addr: 0x74fbd2c0, len: 13824: Operation not permitted [1]
+W0827 04:31:32.117790  2659 memory_location.cpp:73] Failed to get NUMA node, addr: 0x74fa5e40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.118012  2660 memory_location.cpp:73] Failed to get NUMA node, addr: 0x74c24880, len: 6912: Operation not permitted [1]
+W0827 04:31:32.137427  2677 memory_location.cpp:73] Failed to get NUMA node, addr: 0x685bea40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.137485  2678 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6865c240, len: 6912: Operation not permitted [1]
+W0827 04:31:32.137612  2680 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6832f280, len: 6912: Operation not permitted [1]
+W0827 04:31:32.137683  2679 memory_location.cpp:73] Failed to get NUMA node, addr: 0x68511ec0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.137791  2681 memory_location.cpp:73] Failed to get NUMA node, addr: 0x687ae280, len: 55296: Operation not permitted [1]
+W0827 04:31:32.137800  2682 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6863aec0, len: 55296: Operation not permitted [1]
+W0827 04:31:32.137928  2683 memory_location.cpp:73] Failed to get NUMA node, addr: 0x685f4540, len: 6912: Operation not permitted [1]
+W0827 04:31:32.138145  2685 memory_location.cpp:73] Failed to get NUMA node, addr: 0x68b70f80, len: 6912: Operation not permitted [1]
+W0827 04:31:32.138144  2684 memory_location.cpp:73] Failed to get NUMA node, addr: 0x686e87c0, len: 13824: Operation not permitted [1]
+W0827 04:31:32.138219  2686 memory_location.cpp:73] Failed to get NUMA node, addr: 0x685491c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.187127  2706 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8eaeff40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.187187  2707 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8eaa1740, len: 6912: Operation not permitted [1]
+W0827 04:31:32.187136  2705 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8e5f2600, len: 6912: Operation not permitted [1]
+W0827 04:31:32.187413  2709 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ec4eb00, len: 55296: Operation not permitted [1]
+W0827 04:31:32.187405  2708 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ea3b800, len: 6912: Operation not permitted [1]
+W0827 04:31:32.187651  2710 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ed92ec0, len: 55296: Operation not permitted [1]
+W0827 04:31:32.187799  2711 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8eb3fbc0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.188133  2712 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ed0e1c0, len: 13824: Operation not permitted [1]
+W0827 04:31:32.188231  2713 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8e9d20c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.188272  2714 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8f2c7a80, len: 6912: Operation not permitted [1]
+W0827 04:31:32.252096  2732 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72b6b180, len: 6912: Operation not permitted [1]
+W0827 04:31:32.252050  2731 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72be0600, len: 6912: Operation not permitted [1]
+W0827 04:31:32.252147  2733 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72716c00, len: 6912: Operation not permitted [1]
+W0827 04:31:32.252276  2734 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72739940, len: 6912: Operation not permitted [1]
+W0827 04:31:32.252379  2735 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72da8a80, len: 55296: Operation not permitted [1]
+W0827 04:31:32.252439  2736 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72d15800, len: 55296: Operation not permitted [1]
+W0827 04:31:32.252678  2737 memory_location.cpp:73] Failed to get NUMA node, addr: 0x7315db40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.252701  2738 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72e0e700, len: 13824: Operation not permitted [1]
+W0827 04:31:32.252745  2739 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72f5f480, len: 6912: Operation not permitted [1]
+W0827 04:31:32.252832  2740 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72b295c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.268508  2760 memory_location.cpp:73] Failed to get NUMA node, addr: 0x69eefc00, len: 6912: Operation not permitted [1]
+W0827 04:31:32.268532  2759 memory_location.cpp:73] Failed to get NUMA node, addr: 0x69ec8680, len: 6912: Operation not permitted [1]
+W0827 04:31:32.268577  2761 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6a02e280, len: 6912: Operation not permitted [1]
+W0827 04:31:32.268708  2762 memory_location.cpp:73] Failed to get NUMA node, addr: 0x69d75dc0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.268766  2763 memory_location.cpp:73] Failed to get NUMA node, addr: 0x69fc4880, len: 55296: Operation not permitted [1]
+W0827 04:31:32.268886  2764 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6a0c8540, len: 55296: Operation not permitted [1]
+W0827 04:31:32.268975  2765 memory_location.cpp:73] Failed to get NUMA node, addr: 0x69c75940, len: 6912: Operation not permitted [1]
+W0827 04:31:32.269125  2766 memory_location.cpp:73] Failed to get NUMA node, addr: 0x698ae1c0, len: 13824: Operation not permitted [1]
+W0827 04:31:32.269232  2767 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6a5457c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.269354  2768 memory_location.cpp:73] Failed to get NUMA node, addr: 0x69c8fe80, len: 6912: Operation not permitted [1]
+W0827 04:31:32.281724  2785 memory_location.cpp:73] Failed to get NUMA node, addr: 0x678e47c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.281751  2786 memory_location.cpp:73] Failed to get NUMA node, addr: 0x67a39b80, len: 6912: Operation not permitted [1]
+W0827 04:31:32.281852  2787 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6798d9c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.281994  2788 memory_location.cpp:73] Failed to get NUMA node, addr: 0x67981dc0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.282085  2789 memory_location.cpp:73] Failed to get NUMA node, addr: 0x67823300, len: 55296: Operation not permitted [1]
+W0827 04:31:32.282146  2790 memory_location.cpp:73] Failed to get NUMA node, addr: 0x67a3bb00, len: 55296: Operation not permitted [1]
+W0827 04:31:32.282214  2791 memory_location.cpp:73] Failed to get NUMA node, addr: 0x67cdad40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.282285  2792 memory_location.cpp:73] Failed to get NUMA node, addr: 0x68028e40, len: 13824: Operation not permitted [1]
+W0827 04:31:32.282544  2794 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6796cf40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.282545  2793 memory_location.cpp:73] Failed to get NUMA node, addr: 0x67fa6600, len: 6912: Operation not permitted [1]
+[2026-08-27 04:31:32] Dummy health check server started in background thread at 0.0.0.0:30000
+[rank27]:[W827 04:34:16.127713860 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank25]:[W827 04:34:16.128748517 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank24]:[W827 04:34:16.129831915 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank26]:[W827 04:34:16.136239904 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank26]:[W827 04:35:04.964905528 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank25]:[W827 04:35:04.966225881 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank27]:[W827 04:35:04.966239242 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank24]:[W827 04:35:04.966301664 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank31]:[W827 04:35:04.990701389 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank28]:[W827 04:35:04.991914270 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank29]:[W827 04:35:04.994546241 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank30]:[W827 04:35:04.999020079 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank31]:[W827 04:35:10.873647776 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank30]:[W827 04:35:10.873901815 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank28]:[W827 04:35:10.874227564 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank29]:[W827 04:35:10.885318244 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[2026-08-27 04:35:10 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.29
+[2026-08-27 04:35:10 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.29
+[2026-08-27 04:35:29 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.39
+[2026-08-27 04:35:29 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.39
+[2026-08-27 04:35:31 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25.93
+[2026-08-27 04:35:31 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25.95
+[2026-08-27 04:43:23 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.14
+[2026-08-27 04:43:23 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.14
+[2026-08-27 04:48:14 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.22
+[2026-08-27 04:48:14 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.22
+[2026-08-27 04:48:14 PP6 TP1 EP1] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP6 TP2 EP2] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP7 TP2 EP2] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP6 TP0 EP0] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP7 TP0 EP0] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP7 TP1 EP1] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP6 TP3 EP3] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP7 TP3 EP3] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:57:14 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.12
+[2026-08-27 04:57:14 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.12
+[2026-08-27 05:04:21 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1088, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2.55
+[2026-08-27 05:04:21 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1088, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2.55
+[2026-08-27 05:04:41 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.19
+[2026-08-27 05:04:41 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.18
+[2026-08-27 05:05:39 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1.10
+[2026-08-27 05:05:39 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1.10
+[2026-08-27 05:07:33 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 72.18
+[2026-08-27 05:07:33 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 72.18
+[2026-08-27 05:07:33 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 389087.41
+[2026-08-27 05:07:33 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 459604.95
+[2026-08-27 05:07:33 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32068.03
+[2026-08-27 05:07:33 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 36484.58
+[2026-08-27 05:07:40 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8673, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1083.15
+[2026-08-27 05:07:40 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8673, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1082.87
+[2026-08-27 05:07:40 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 481, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 423255.51
+[2026-08-27 05:07:40 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 481, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 454364.97
+[2026-08-27 05:07:40 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 34494.75
+[2026-08-27 05:07:40 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33090.16
+[2026-08-27 05:07:48 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8656, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1082.19
+[2026-08-27 05:07:48 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8656, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1082.17
+[2026-08-27 05:07:48 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 464, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 443415.27
+[2026-08-27 05:07:48 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 464, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 422320.55
+[2026-08-27 05:07:48 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 30852.82
+[2026-08-27 05:07:48 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33407.06
+[2026-08-27 05:07:55 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8663, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1087.54
+[2026-08-27 05:07:55 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8663, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1087.42
+[2026-08-27 05:07:55 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 471, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 437681.61
+[2026-08-27 05:07:55 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 471, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 442367.54
+[2026-08-27 05:07:55 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 31561.40
+[2026-08-27 05:07:55 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 31433.36
+[2026-08-27 05:08:03 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8659, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1083.03
+[2026-08-27 05:08:03 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8659, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1082.87
+[2026-08-27 05:08:03 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 467, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 376910.00
+[2026-08-27 05:08:03 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 467, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 408825.56
+[2026-08-27 05:08:03 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32102.57
+[2026-08-27 05:08:03 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33296.68
+[2026-08-27 05:08:11 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8692, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1085.73
+[2026-08-27 05:08:11 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8692, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1085.61
+[2026-08-27 05:08:11 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 500, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 434356.57
+[2026-08-27 05:08:11 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 500, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 383780.71
+[2026-08-27 05:08:11 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 27941.29
+[2026-08-27 05:08:11 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32876.54
+[2026-08-27 05:08:18 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8557, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1090.32
+[2026-08-27 05:08:18 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8557, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1090.65
+[2026-08-27 05:08:18 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 365, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 477756.06
+[2026-08-27 05:08:18 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 365, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 457694.49
+[2026-08-27 05:08:18 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 27080.22
+[2026-08-27 05:08:18 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 27388.57
+[2026-08-27 05:08:26 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8683, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1083.79
+[2026-08-27 05:08:26 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8683, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1083.51
+[2026-08-27 05:08:26 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 491, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 434750.14
+[2026-08-27 05:08:26 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 491, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 447315.13
+[2026-08-27 05:08:26 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33020.44
+[2026-08-27 05:08:26 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33110.34
+[2026-08-27 05:11:45 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.32
+[2026-08-27 05:11:45 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.32
+[2026-08-27 05:14:42 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1088, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6.16
+[2026-08-27 05:14:42 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1088, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6.16
+[2026-08-27 05:17:47 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 44.21
+[2026-08-27 05:17:47 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 44.21
+[2026-08-27 05:17:47 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 461237.58
+[2026-08-27 05:17:47 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 443473.87
+[2026-08-27 05:17:47 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 30106.68
+[2026-08-27 05:17:47 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32700.87
+[2026-08-27 05:17:55 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8673, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1086.24
+[2026-08-27 05:17:55 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8673, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1086.14
+[2026-08-27 05:17:55 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 481, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 438855.61
+[2026-08-27 05:17:55 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 481, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 448853.01
+[2026-08-27 05:17:55 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32917.40
+[2026-08-27 05:17:55 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 35669.59
+[2026-08-27 05:18:02 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8656, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1088.38
+[2026-08-27 05:18:02 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8656, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1088.12
+[2026-08-27 05:18:02 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 464, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 432657.94
+[2026-08-27 05:18:02 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 464, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 441680.19
+[2026-08-27 05:18:02 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32420.83
+[2026-08-27 05:18:02 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33219.83
+[2026-08-27 05:18:10 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8663, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1087.85
+[2026-08-27 05:18:10 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8663, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1087.68
+[2026-08-27 05:18:10 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 471, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 433552.65
+[2026-08-27 05:18:10 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 471, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 448022.33
+[2026-08-27 05:18:10 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32265.40
+[2026-08-27 05:18:10 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32427.35
+[2026-08-27 05:18:18 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8659, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1086.07
+[2026-08-27 05:18:18 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8659, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1085.68
+[2026-08-27 05:18:18 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 467, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 365245.86
+[2026-08-27 05:18:18 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 467, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 360696.15
+[2026-08-27 05:18:18 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 27416.04
+[2026-08-27 05:18:18 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 35182.90
+[2026-08-27 05:18:25 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8692, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1087.81
+[2026-08-27 05:18:25 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8692, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1087.87
+[2026-08-27 05:18:25 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 500, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 494405.53
+[2026-08-27 05:18:25 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 500, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 406159.77
+[2026-08-27 05:18:25 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 28692.54
+[2026-08-27 05:18:25 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 35731.06
+[2026-08-27 05:18:33 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8557, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1090.88
+[2026-08-27 05:18:33 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8557, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1090.72
+[2026-08-27 05:18:33 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 365, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 476480.55
+[2026-08-27 05:18:33 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 365, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 465689.14
+[2026-08-27 05:18:33 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 24743.50
+[2026-08-27 05:18:33 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 26986.87
+[2026-08-27 05:18:40 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8683, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1087.65
+[2026-08-27 05:18:40 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8683, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1087.47
+[2026-08-27 05:18:40 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 491, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 444510.88
+[2026-08-27 05:18:40 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 491, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 381334.71
+[2026-08-27 05:18:40 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 27475.75
+[2026-08-27 05:18:40 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 35546.91
+[2026-08-27 05:21:48 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 2, #pending-token: 8652, #bootstrap-req: 2, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 43.63
+[2026-08-27 05:21:49 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 3, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9876.33
+[2026-08-27 05:21:50 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 2, #pending-token: 8652, #bootstrap-req: 2, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 43.26
+[2026-08-27 05:21:50 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 3, #pending-token: 9185, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8052.60
+[2026-08-27 05:21:50 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 3, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9996.10
+[2026-08-27 05:21:51 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 2, #pending-token: 993, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9980.46
+[2026-08-27 05:21:51 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 3, #pending-token: 9185, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8828.21
+[2026-08-27 05:21:52 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 2, #pending-token: 9680, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9798.99
+[2026-08-27 05:21:52 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 2, #pending-token: 993, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9906.89
+[2026-08-27 05:21:52 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1488, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9933.78
+[2026-08-27 05:21:53 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 2, #pending-token: 9680, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9164.38
+[2026-08-27 05:21:53 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 10199, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9100.99
+[2026-08-27 05:21:54 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1488, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9595.39
+[2026-08-27 05:21:54 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 18858, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9669.02
+[2026-08-27 05:21:55 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 10199, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9348.80
+[2026-08-27 05:21:55 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 27591, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9301.63
+[2026-08-27 05:21:56 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 18858, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9457.16
+[2026-08-27 05:21:56 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 53023, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9526.90
+[2026-08-27 05:21:57 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 27591, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9435.09
+[2026-08-27 05:21:57 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 44876, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9774.31
+[2026-08-27 05:21:57 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 53023, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 15230.20
+[2026-08-27 05:21:57 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 36684, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 18786.53
+[2026-08-27 05:21:57 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 44876, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 521404.18
+[2026-08-27 05:21:57 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 28504, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 355637.74
+[2026-08-27 05:21:57 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 36684, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 388735.22
+[2026-08-27 05:21:57 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 20312, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 74956.78
+[2026-08-27 05:21:57 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 28504, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 73931.58
+[2026-08-27 05:21:58 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 12139, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 16102.75
+[2026-08-27 05:21:58 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 20312, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 14970.35
+[2026-08-27 05:21:58 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 3947, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 12109.59
+[2026-08-27 05:21:59 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 12139, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10923.81
+[2026-08-27 05:21:59 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 3968, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 5070.20
+[2026-08-27 05:21:59 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 3947, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11542.46
+[2026-08-27 05:22:00 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 3968, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4785.41
+[2026-08-27 05:22:08 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 0, #pending-token: 8608, #bootstrap-req: 4, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 940.83
+[2026-08-27 05:22:09 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 416, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10204.68
+[2026-08-27 05:22:09 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 0, #pending-token: 8608, #bootstrap-req: 4, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 868.76
+[2026-08-27 05:22:10 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 9149, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8740.29
+[2026-08-27 05:22:10 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 416, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 29834.98
+[2026-08-27 05:22:10 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 957, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9976.76
+[2026-08-27 05:22:11 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 9149, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7962.19
+[2026-08-27 05:22:11 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 9638, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9468.71
+[2026-08-27 05:22:12 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 957, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6057.13
+[2026-08-27 05:22:12 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1446, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8241.57
+[2026-08-27 05:22:13 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 9638, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10874.51
+[2026-08-27 05:22:13 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 10033, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10016.49
+[2026-08-27 05:22:14 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1446, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7794.33
+[2026-08-27 05:22:14 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1841, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8525.06
+[2026-08-27 05:22:15 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 10033, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10735.67
+[2026-08-27 05:22:15 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1856, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2257.01
+[2026-08-27 05:22:16 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1841, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8114.90
+[2026-08-27 05:22:16 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8540, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8613.20
+[2026-08-27 05:22:16 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1856, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2421.25
+[2026-08-27 05:22:17 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 17066, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10036.34
+[2026-08-27 05:22:17 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8540, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8278.93
+[2026-08-27 05:22:18 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8910, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9399.94
+[2026-08-27 05:22:18 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 17066, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 36997.59
+[2026-08-27 05:22:18 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 17466, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 51312.40
+[2026-08-27 05:22:18 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8910, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 411464.53
+[2026-08-27 05:22:18 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9324, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 66779.98
+[2026-08-27 05:22:18 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 17466, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 68447.00
+[2026-08-27 05:22:18 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 17751, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 16323.38
+[2026-08-27 05:22:18 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9324, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 15086.19
+[2026-08-27 05:22:19 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9579, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 12098.16
+[2026-08-27 05:22:19 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 17751, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10943.79
+[2026-08-27 05:22:20 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1387, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10522.42
+[2026-08-27 05:22:20 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9579, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11560.43
+[2026-08-27 05:22:20 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1408, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2179.97
+[2026-08-27 05:22:21 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1387, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9951.54
+[2026-08-27 05:22:21 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1408, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1996.17
+[2026-08-27 05:22:29 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8547, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1014.77
+[2026-08-27 05:22:29 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 355, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10133.87
+[2026-08-27 05:22:30 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8547, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 936.03
+[2026-08-27 05:22:30 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8932, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8653.72
+[2026-08-27 05:22:30 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 355, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 34268.89
+[2026-08-27 05:22:31 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 740, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10007.94
+[2026-08-27 05:22:31 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8932, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7704.93
+[2026-08-27 05:22:32 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9259, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9600.28
+[2026-08-27 05:22:33 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 740, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6120.45
+[2026-08-27 05:22:33 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1067, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8368.71
+[2026-08-27 05:22:34 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9259, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10965.05
+[2026-08-27 05:22:34 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9745, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9275.86
+[2026-08-27 05:22:35 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1067, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7846.08
+[2026-08-27 05:22:35 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1553, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9172.31
+[2026-08-27 05:22:35 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9745, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10985.13
+[2026-08-27 05:22:36 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1600, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1929.98
+[2026-08-27 05:22:36 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1553, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7933.67
+[2026-08-27 05:22:36 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8602, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8619.09
+[2026-08-27 05:22:37 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1600, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2101.55
+[2026-08-27 05:22:37 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 410, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10046.41
+[2026-08-27 05:22:38 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8602, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8223.98
+[2026-08-27 05:22:38 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8696, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9435.25
+[2026-08-27 05:22:38 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 410, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 45497.16
+[2026-08-27 05:22:38 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 504, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 66281.64
+[2026-08-27 05:22:38 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8696, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 425064.62
+[2026-08-27 05:22:38 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9164, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 67906.51
+[2026-08-27 05:22:38 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 504, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 69370.77
+[2026-08-27 05:22:39 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 972, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 16327.79
+[2026-08-27 05:22:39 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9164, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 15103.13
+[2026-08-27 05:22:40 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9322, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 12119.18
+[2026-08-27 05:22:40 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 972, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10991.29
+[2026-08-27 05:22:40 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1130, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10534.55
+[2026-08-27 05:22:40 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9322, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11588.54
+[2026-08-27 05:22:41 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1152, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1784.29
+[2026-08-27 05:22:41 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1130, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10014.78
+[2026-08-27 05:22:42 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1152, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1638.04
+[2026-08-27 05:22:49 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8602, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1020.06
+[2026-08-27 05:22:50 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 410, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10100.36
+[2026-08-27 05:22:51 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8602, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 939.88
+[2026-08-27 05:22:51 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 9105, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8814.04
+[2026-08-27 05:22:51 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 410, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 31402.72
+[2026-08-27 05:22:52 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 913, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9823.70
+[2026-08-27 05:22:52 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 9105, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7950.01
+[2026-08-27 05:22:52 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9375, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9520.99
+[2026-08-27 05:22:53 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 913, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6088.53
+[2026-08-27 05:22:53 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1183, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8593.45
+[2026-08-27 05:22:54 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9375, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11034.51
+[2026-08-27 05:22:54 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9868, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9888.23
+[2026-08-27 05:22:55 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1183, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7825.73
+[2026-08-27 05:22:55 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1676, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8507.30
+[2026-08-27 05:22:56 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9868, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10702.83
+[2026-08-27 05:22:56 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2096.94
+[2026-08-27 05:22:57 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1676, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8175.44
+[2026-08-27 05:22:57 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8467, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8745.10
+[2026-08-27 05:22:58 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2235.70
+[2026-08-27 05:22:58 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 275, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9728.14
+[2026-08-27 05:22:59 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8467, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8423.31
+[2026-08-27 05:22:59 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8988, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9773.55
+[2026-08-27 05:22:59 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 275, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 30877.72
+[2026-08-27 05:22:59 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 796, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 41016.53
+[2026-08-27 05:22:59 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8988, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 458929.33
+[2026-08-27 05:22:59 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9363, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 73474.43
+[2026-08-27 05:22:59 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 796, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 74700.60
+[2026-08-27 05:22:59 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 1171, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 16297.37
+[2026-08-27 05:23:00 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9363, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 15107.75
+[2026-08-27 05:23:00 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9880, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 12178.33
+[2026-08-27 05:23:00 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 1171, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11020.95
+[2026-08-27 05:23:01 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1688, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10567.86
+[2026-08-27 05:23:01 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9880, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11615.07
+[2026-08-27 05:23:02 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2677.95
+[2026-08-27 05:23:02 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1688, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10008.16
+[2026-08-27 05:23:02 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2457.28
+[2026-08-27 05:23:10 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8601, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1020.50
+[2026-08-27 05:23:10 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 409, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10128.83
+[2026-08-27 05:23:11 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8601, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 941.05
+[2026-08-27 05:23:11 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 9104, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8792.53
+[2026-08-27 05:23:11 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 409, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 31651.22
+[2026-08-27 05:23:12 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 912, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9996.90
+[2026-08-27 05:23:12 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 9104, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7936.01
+[2026-08-27 05:23:13 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9485, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9519.44
+[2026-08-27 05:23:14 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 912, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6079.95
+[2026-08-27 05:23:14 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1293, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8474.47
+[2026-08-27 05:23:15 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9485, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10878.89
+[2026-08-27 05:23:15 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9900, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9816.19
+[2026-08-27 05:23:16 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1293, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7901.51
+[2026-08-27 05:23:16 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1708, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8613.59
+[2026-08-27 05:23:16 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9900, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10766.20
+[2026-08-27 05:23:17 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2099.82
+[2026-08-27 05:23:17 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1708, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8087.47
+[2026-08-27 05:23:18 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8525, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8707.95
+[2026-08-27 05:23:18 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2244.91
+[2026-08-27 05:23:18 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 333, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10166.80
+[2026-08-27 05:23:19 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8525, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8506.15
+[2026-08-27 05:23:19 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8909, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9459.25
+[2026-08-27 05:23:19 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 333, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33322.73
+[2026-08-27 05:23:19 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 717, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 45670.17
+[2026-08-27 05:23:19 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8909, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 667975.55
+[2026-08-27 05:23:19 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9452, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 458625.25
+[2026-08-27 05:23:19 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 717, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 444740.14
+[2026-08-27 05:23:19 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9452, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 446180.29
+[2026-08-27 05:23:19 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 1260, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 337352.56
+[2026-08-27 05:23:19 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 1260, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 382567.64
+[2026-08-27 05:23:19 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9830, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 429427.21
+[2026-08-27 05:23:19 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1638, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 448443.92
+[2026-08-27 05:23:19 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9830, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 396809.23
+[2026-08-27 05:23:19 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1664, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 93514.06
+[2026-08-27 05:23:19 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1638, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 464006.76
+[2026-08-27 05:23:20 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1664, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 108432.29
+[2026-08-27 05:32:52 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.11
+[2026-08-27 05:32:52 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.11
+[2026-08-27 05:35:54 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 45.15
+[2026-08-27 05:35:54 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 45.15
+[2026-08-27 05:35:54 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 412328.11
+[2026-08-27 05:35:54 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 421944.54
+[2026-08-27 05:35:54 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 31302.30
+[2026-08-27 05:35:54 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32994.85
+[2026-08-27 05:36:34 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1.59
+[2026-08-27 05:36:34 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1.59
+[2026-08-27 05:42:35 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 8652, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 22.70
+[2026-08-27 05:42:36 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 3, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10126.71
+[2026-08-27 05:42:37 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 8652, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 22.60
+[2026-08-27 05:42:37 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 3, #pending-token: 9185, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8640.69
+[2026-08-27 05:42:37 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 3, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10092.22
+[2026-08-27 05:42:38 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 2, #pending-token: 993, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9974.31
+[2026-08-27 05:42:38 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 3, #pending-token: 9185, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8772.78
+[2026-08-27 05:42:39 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 2, #pending-token: 9680, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8911.62
+[2026-08-27 05:42:39 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 2, #pending-token: 993, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9785.60
+[2026-08-27 05:42:39 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1488, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9844.23
+[2026-08-27 05:42:40 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 2, #pending-token: 9680, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9174.59
+[2026-08-27 05:42:40 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 10199, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9078.31
+[2026-08-27 05:42:41 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1488, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9572.33
+[2026-08-27 05:42:41 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 18858, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9739.02
+[2026-08-27 05:42:42 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 10199, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9373.97
+[2026-08-27 05:42:42 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 10707, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9312.43
+[2026-08-27 05:42:43 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 18858, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9508.95
+[2026-08-27 05:42:43 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 53023, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9641.34
+[2026-08-27 05:42:44 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 10707, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9420.68
+[2026-08-27 05:42:44 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 44876, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9809.06
+[2026-08-27 05:42:44 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 53023, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 15384.37
+[2026-08-27 05:42:44 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 36684, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 18299.65
+[2026-08-27 05:42:44 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 44876, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 729022.60
+[2026-08-27 05:42:44 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 36684, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 584814.06
+[2026-08-27 05:42:44 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 28504, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 485869.39
+[2026-08-27 05:42:44 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 28504, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 479480.92
+[2026-08-27 05:42:44 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 20312, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 381450.12
+[2026-08-27 05:42:44 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 20312, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 342991.68
+[2026-08-27 05:42:44 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 12139, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 390968.12
+[2026-08-27 05:42:44 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 12139, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 482447.39
+[2026-08-27 05:42:44 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 3947, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 522368.06
+[2026-08-27 05:42:44 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 3947, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 481188.96
+[2026-08-27 05:42:44 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 3968, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 246103.15
+[2026-08-27 05:42:44 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 3968, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 371450.51
+[2026-08-27 05:43:18 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8608, #bootstrap-req: 1, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 239.71
+[2026-08-27 05:43:19 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 416, #bootstrap-req: 2, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 16127.61
+[2026-08-27 05:43:19 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8608, #bootstrap-req: 1, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 235.99
+[2026-08-27 05:43:20 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 448, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 671.79
+[2026-08-27 05:43:20 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 416, #bootstrap-req: 2, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10816.40
+[2026-08-27 05:43:20 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 448, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 667.60
+[2026-08-27 05:43:29 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.06, #running-req: 0, #queue-req: 0, #pending-token: 8701, #bootstrap-req: 4, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 882.19
+[2026-08-27 05:43:30 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.06, #running-req: 0, #queue-req: 0, #pending-token: 509, #bootstrap-req: 4, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8918.66
+[2026-08-27 05:43:30 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.06, #running-req: 0, #queue-req: 0, #pending-token: 8701, #bootstrap-req: 4, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 861.04
+[2026-08-27 05:43:31 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.01, mamba usage: 0.06, #running-req: 0, #queue-req: 4, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 660.42
+[2026-08-27 05:43:31 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.06, #running-req: 0, #queue-req: 0, #pending-token: 509, #bootstrap-req: 4, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11530.04
+[2026-08-27 05:43:31 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 3, #pending-token: 8678, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11490.79
+[2026-08-27 05:43:31 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.01, mamba usage: 0.06, #running-req: 0, #queue-req: 4, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 626.20
+[2026-08-27 05:43:32 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 3, #pending-token: 8678, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11633.75
+[2026-08-27 05:43:32 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 3, #pending-token: 17239, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9140.68
+[2026-08-27 05:43:33 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 2, #pending-token: 9073, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9249.67
+[2026-08-27 05:43:34 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 3, #pending-token: 17239, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4790.93
+[2026-08-27 05:43:34 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 2, #pending-token: 881, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8775.12
+[2026-08-27 05:43:35 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 2, #pending-token: 9073, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10809.52
+[2026-08-27 05:43:35 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 896, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1051.38
+[2026-08-27 05:43:36 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 2, #pending-token: 881, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7651.56
+[2026-08-27 05:43:36 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 896, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1174.25
+[2026-08-27 05:43:38 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 58625, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2979.25
+[2026-08-27 05:43:38 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 50433, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9894.26
+[2026-08-27 05:43:39 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 58625, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2938.37
+[2026-08-27 05:43:39 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 42277, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9321.97
+[2026-08-27 05:43:39 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 50433, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 36944.53
+[2026-08-27 05:43:39 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 42277, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 767120.00
+[2026-08-27 05:43:39 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 34085, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 52557.72
+[2026-08-27 05:43:39 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 34085, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 412026.53
+[2026-08-27 05:43:39 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 25943, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 326957.01
+[2026-08-27 05:43:39 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 25943, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 274941.05
+[2026-08-27 05:43:39 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 17751, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 343648.36
+[2026-08-27 05:43:39 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 17751, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 448059.58
+[2026-08-27 05:43:39 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 9579, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 409851.57
+[2026-08-27 05:43:39 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 0, #pending-token: 1387, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 448745.58
+[2026-08-27 05:43:39 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 9579, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 386427.27
+[2026-08-27 05:43:40 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 0, #pending-token: 1387, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 490764.62
+[2026-08-27 05:43:40 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1408, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 82817.87
+[2026-08-27 05:43:40 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1408, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 97282.15
+[2026-08-27 05:43:50 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8547, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 806.00
+[2026-08-27 05:43:50 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8547, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 805.80
+[2026-08-27 05:43:50 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 355, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 379337.52
+[2026-08-27 05:43:50 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 355, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 414759.69
+[2026-08-27 05:43:50 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 23329.01
+[2026-08-27 05:43:50 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25907.81
+[2026-08-27 05:44:08 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8548, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 436.72
+[2026-08-27 05:44:08 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8548, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 436.57
+[2026-08-27 05:44:08 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 356, #bootstrap-req: 1, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 350628.97
+[2026-08-27 05:44:09 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 356, #bootstrap-req: 1, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 403205.84
+[2026-08-27 05:44:09 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 22048.39
+[2026-08-27 05:44:09 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4949.90
+[2026-08-27 05:44:18 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8491, #bootstrap-req: 1, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 888.39
+[2026-08-27 05:44:18 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 299, #bootstrap-req: 5, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 21692.49
+[2026-08-27 05:44:19 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8491, #bootstrap-req: 1, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 824.92
+[2026-08-27 05:44:19 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 299, #bootstrap-req: 5, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 34995.55
+[2026-08-27 05:44:19 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 320, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 1, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 488.98
+[2026-08-27 05:44:20 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 320, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 1, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 374.65
+[2026-08-27 05:44:26 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.06, #running-req: 0, #queue-req: 3, #pending-token: 8657, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1124.61
+[2026-08-27 05:44:27 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 2, #pending-token: 465, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9603.31
+[2026-08-27 05:44:27 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.06, #running-req: 0, #queue-req: 3, #pending-token: 8657, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1045.84
+[2026-08-27 05:44:28 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.01, mamba usage: 0.06, #running-req: 0, #queue-req: 2, #pending-token: 0, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 609.24
+[2026-08-27 05:44:28 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 2, #pending-token: 465, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10933.24
+[2026-08-27 05:44:29 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.01, mamba usage: 0.06, #running-req: 0, #queue-req: 2, #pending-token: 0, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 413.94
+[2026-08-27 05:44:31 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 58376, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2254.96
+[2026-08-27 05:44:32 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 50184, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9769.76
+[2026-08-27 05:44:33 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 58376, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2290.60
+[2026-08-27 05:44:33 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 42030, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8755.23
+[2026-08-27 05:44:33 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 50184, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 44086.34
+[2026-08-27 05:44:34 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 33838, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10091.78
+[2026-08-27 05:44:34 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 42030, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7339.62
+[2026-08-27 05:44:35 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 25654, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9419.96
+[2026-08-27 05:44:36 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 33838, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6021.13
+[2026-08-27 05:44:36 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 17462, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9053.20
+[2026-08-27 05:44:36 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 25654, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 82486.95
+[2026-08-27 05:44:36 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 9322, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 184356.37
+[2026-08-27 05:44:36 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 17462, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 493997.44
+[2026-08-27 05:44:36 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 9322, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 327564.69
+[2026-08-27 05:44:36 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1130, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 313269.46
+[2026-08-27 05:44:36 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1130, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 329615.22
+[2026-08-27 05:44:36 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1152, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 46517.13
+[2026-08-27 05:44:36 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8602, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 82360.09
+[2026-08-27 05:44:36 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1152, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11364.86
+[2026-08-27 05:44:36 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 410, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 16282.84
+[2026-08-27 05:44:36 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8602, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 15163.02
+[2026-08-27 05:44:37 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 448, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 674.57
+[2026-08-27 05:44:37 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 410, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10855.77
+[2026-08-27 05:44:38 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 448, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 694.16
+[2026-08-27 05:44:43 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8657, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1369.35
+[2026-08-27 05:44:43 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8657, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1568.46
+[2026-08-27 05:44:43 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 465, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 401366.39
+[2026-08-27 05:44:43 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 465, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 402067.29
+[2026-08-27 05:44:43 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32182.07
+[2026-08-27 05:44:43 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33329.12
+[2026-08-27 05:44:52 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8415, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 915.37
+[2026-08-27 05:44:52 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8415, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 915.28
+[2026-08-27 05:44:52 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 223, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 426783.19
+[2026-08-27 05:44:52 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 223, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 438981.49
+[2026-08-27 05:44:52 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 256, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 15535.29
+[2026-08-27 05:44:52 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 256, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 17522.02
+[2026-08-27 05:45:07 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 535.41
+[2026-08-27 05:45:07 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 535.18
+[2026-08-27 05:45:07 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 405105.28
+[2026-08-27 05:45:07 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 452249.88
+[2026-08-27 05:45:07 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 26546.28
+[2026-08-27 05:45:07 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 23977.04
+[2026-08-27 05:45:15 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8467, #bootstrap-req: 4, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1036.10
+[2026-08-27 05:45:16 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 275, #bootstrap-req: 4, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 12005.20
+[2026-08-27 05:45:16 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8467, #bootstrap-req: 4, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 943.37
+[2026-08-27 05:45:17 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 320, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 404.21
+[2026-08-27 05:45:17 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 275, #bootstrap-req: 4, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11467.16
+[2026-08-27 05:45:18 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 320, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 381.21
+[2026-08-27 05:45:26 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 42247, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 938.30
+[2026-08-27 05:45:26 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 34055, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9933.54
+[2026-08-27 05:45:27 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 42247, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 864.71
+[2026-08-27 05:45:27 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 42692, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8673.14
+[2026-08-27 05:45:28 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 34055, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10278.25
+[2026-08-27 05:45:28 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 34500, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10096.23
+[2026-08-27 05:45:29 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 42692, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8651.64
+[2026-08-27 05:45:29 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 26353, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9233.46
+[2026-08-27 05:45:29 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 34500, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 22667.74
+[2026-08-27 05:45:30 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 18161, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 14748.30
+[2026-08-27 05:45:30 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 26353, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 24606.02
+[2026-08-27 05:45:30 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 26857, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 12396.15
+[2026-08-27 05:45:30 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 18161, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10937.36
+[2026-08-27 05:45:31 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 18665, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11952.78
+[2026-08-27 05:45:31 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 26857, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 12714.11
+[2026-08-27 05:45:31 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 10512, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 20836.89
+[2026-08-27 05:45:32 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 18665, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11789.57
+[2026-08-27 05:45:32 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 10512, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 31975.94
+[2026-08-27 05:45:33 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 2320, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6211.92
+[2026-08-27 05:45:33 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 2320, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 5364.33
+[2026-08-27 05:45:33 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 2368, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2783.79
+[2026-08-27 05:45:34 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 2368, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2789.77
+[2026-08-27 05:45:37 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.06, #running-req: 0, #queue-req: 0, #pending-token: 8525, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2482.76
+[2026-08-27 05:45:37 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.04, #running-req: 0, #queue-req: 0, #pending-token: 333, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 12384.46
+[2026-08-27 05:45:38 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.06, #running-req: 0, #queue-req: 0, #pending-token: 8525, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2522.81
+[2026-08-27 05:45:38 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.01, mamba usage: 0.04, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 551.26
+[2026-08-27 05:45:38 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.04, #running-req: 0, #queue-req: 0, #pending-token: 333, #bootstrap-req: 1, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 12673.24
+[2026-08-27 05:45:39 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.01, mamba usage: 0.04, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 553.52
+[2026-08-27 05:45:43 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8556, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1552.84
+[2026-08-27 05:45:43 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8556, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1814.40
+[2026-08-27 05:45:43 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 364, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 407741.53
+[2026-08-27 05:45:43 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 364, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 404680.96
+[2026-08-27 05:45:43 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 22227.18
+[2026-08-27 05:45:43 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 21957.12
+[2026-08-27 05:45:52 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8525, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 984.51
+[2026-08-27 05:45:52 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8525, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 984.47
+[2026-08-27 05:45:52 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 333, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 418586.12
+[2026-08-27 05:45:52 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 333, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 441542.39
+[2026-08-27 05:45:52 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 22458.89
+[2026-08-27 05:45:52 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 22117.35
+[2026-08-27 05:46:10 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8525, #bootstrap-req: 2, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 451.61
+[2026-08-27 05:46:11 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 333, #bootstrap-req: 2, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10372.87
+[2026-08-27 05:46:11 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8525, #bootstrap-req: 2, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 432.12
+[2026-08-27 05:46:11 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 607.04
+[2026-08-27 05:46:12 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 333, #bootstrap-req: 2, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9858.89
+[2026-08-27 05:46:12 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 593.37
+[2026-08-27 05:46:18 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.06, #running-req: 0, #queue-req: 0, #pending-token: 8684, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1282.65
+[2026-08-27 05:46:18 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.06, #running-req: 0, #queue-req: 0, #pending-token: 8684, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1485.37
+[2026-08-27 05:46:18 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.04, #running-req: 0, #queue-req: 0, #pending-token: 492, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 348073.37
+[2026-08-27 05:46:18 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.04, #running-req: 0, #queue-req: 0, #pending-token: 492, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 334339.13
+[2026-08-27 05:46:18 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.04, #running-req: 0, #queue-req: 0, #pending-token: 9062, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 403326.36
+[2026-08-27 05:46:18 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.04, #running-req: 0, #queue-req: 0, #pending-token: 870, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 425779.98
+[2026-08-27 05:46:18 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.04, #running-req: 0, #queue-req: 0, #pending-token: 9062, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 388882.76
+[2026-08-27 05:46:18 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.01, mamba usage: 0.04, #running-req: 0, #queue-req: 0, #pending-token: 870, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 428926.87
+[2026-08-27 05:46:18 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 896, #cached-token: 0, full token usage: 0.01, mamba usage: 0.04, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 45842.45
+[2026-08-27 05:46:18 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 896, #cached-token: 0, full token usage: 0.01, mamba usage: 0.04, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 62359.04
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-c8-20260827-1340/logs/manual_router.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-c8-20260827-1340/logs/manual_router.log
new file mode 100644
index 0000000..5ea37ca
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-c8-20260827-1340/logs/manual_router.log
@@ -0,0 +1,188 @@
+MiniLB is only for debugging purposes, it only supports random policy!
+[MiniLB] Overriding policy to random
+INFO:     Started server process [1]
+INFO:     Waiting for application startup.
+INFO:     Application startup complete.
+INFO:     174.1.60.1:51774 - "GET /health HTTP/1.1" 200 OK
+INFO:     174.1.60.1:39128 - "GET /health HTTP/1.1" 200 OK
+INFO:     174.1.60.1:43568 - "GET /health HTTP/1.1" 200 OK
+INFO:     174.1.60.1:58114 - "GET /v1/models HTTP/1.1" 200 OK
+INFO:     174.1.60.1:58122 - "GET /v1/models HTTP/1.1" 200 OK
+INFO:     174.1.60.1:34030 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:52970 - "GET /server_info HTTP/1.1" 200 OK
+INFO:     174.1.60.1:52972 - "GET /server_info HTTP/1.1" 200 OK
+INFO:     Uvicorn running on http://0.0.0.0:31000 (Press CTRL+C to quit)
+/usr/local/lib/python3.12/dist-packages/fastapi/routing.py:350: FastAPIDeprecationWarning: ORJSONResponse is deprecated, FastAPI now serializes data directly to JSON bytes via Pydantic when a return type or response model is set, which is faster and doesn't need a custom response class. Read more in the FastAPI docs: https://fastapi.tiangolo.com/advanced/custom-response/#orjson-or-response-model and https://fastapi.tiangolo.com/tutorial/response-model/
+  return await dependant.call(**values)
+INFO:     174.1.60.1:43934 - "GET /health HTTP/1.1" 200 OK
+INFO:     174.1.60.1:57120 - "GET /v1/models HTTP/1.1" 200 OK
+INFO:     174.1.60.1:57126 - "GET /v1/models HTTP/1.1" 200 OK
+INFO:     174.1.60.1:55592 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:55598 - "GET /server_info HTTP/1.1" 200 OK
+INFO:     174.1.60.1:55600 - "GET /server_info HTTP/1.1" 200 OK
+INFO:     174.1.60.1:39092 - "GET /v1/models HTTP/1.1" 200 OK
+INFO:     174.1.60.1:39100 - "GET /v1/models HTTP/1.1" 200 OK
+INFO:     174.1.60.1:42280 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:51506 - "GET /server_info HTTP/1.1" 200 OK
+INFO:     174.1.60.1:51522 - "GET /server_info HTTP/1.1" 200 OK
+INFO:     174.1.60.1:59130 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:54978 - "GET /v1/models HTTP/1.1" 200 OK
+/usr/local/lib/python3.12/dist-packages/sglang_router/mini_lb.py:390: FastAPIDeprecationWarning: ORJSONResponse is deprecated, FastAPI now serializes data directly to JSON bytes via Pydantic when a return type or response model is set, which is faster and doesn't need a custom response class. Read more in the FastAPI docs: https://fastapi.tiangolo.com/advanced/custom-response/#orjson-or-response-model and https://fastapi.tiangolo.com/tutorial/response-model/
+  return await lb.generate(
+INFO:     174.1.60.1:54992 - "GET /v1/models HTTP/1.1" 200 OK
+INFO:     174.1.60.1:45422 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:49916 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:52084 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:52096 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:45122 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:35498 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:50726 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:52480 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:52490 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:40770 - "GET /server_info HTTP/1.1" 200 OK
+INFO:     174.1.60.1:40776 - "GET /server_info HTTP/1.1" 200 OK
+INFO:     174.1.60.1:41252 - "GET /health HTTP/1.1" 200 OK
+INFO:     174.1.60.1:53822 - "GET /v1/models HTTP/1.1" 200 OK
+INFO:     174.1.60.1:53832 - "GET /v1/models HTTP/1.1" 200 OK
+INFO:     174.1.60.1:36760 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:36764 - "GET /server_info HTTP/1.1" 200 OK
+INFO:     174.1.60.1:36772 - "GET /server_info HTTP/1.1" 200 OK
+INFO:     174.1.60.1:47514 - "GET /v1/models HTTP/1.1" 200 OK
+INFO:     174.1.60.1:47520 - "GET /v1/models HTTP/1.1" 200 OK
+INFO:     174.1.60.1:35562 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:35578 - "GET /server_info HTTP/1.1" 200 OK
+INFO:     174.1.60.1:35586 - "GET /server_info HTTP/1.1" 200 OK
+INFO:     174.1.60.1:40864 - "GET /v1/models HTTP/1.1" 200 OK
+INFO:     174.1.60.1:40878 - "GET /v1/models HTTP/1.1" 200 OK
+INFO:     174.1.60.1:35998 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:47822 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:41780 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:49022 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:49034 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:36558 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:49056 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:50776 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:50790 - "GET /server_info HTTP/1.1" 200 OK
+INFO:     174.1.60.1:50802 - "GET /server_info HTTP/1.1" 200 OK
+INFO:     174.1.60.1:45642 - "GET /v1/models HTTP/1.1" 200 OK
+INFO:     174.1.60.1:45656 - "GET /v1/models HTTP/1.1" 200 OK
+INFO:     174.1.60.1:47786 - "POST /generate HTTP/1.1" 200 OK
+Unclosed connection
+client_connection: Connection<ConnectionKey(host='174.1.60.1', port=30000, is_ssl=False, ssl=True, proxy=None, proxy_auth=None, proxy_headers_hash=None, server_hostname=None)>
+INFO:     174.1.60.1:47794 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:47804 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:47806 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:47812 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:47816 - "POST /generate HTTP/1.1" 200 OK
+Unclosed connection
+client_connection: Connection<ConnectionKey(host='174.1.60.1', port=30000, is_ssl=False, ssl=True, proxy=None, proxy_auth=None, proxy_headers_hash=None, server_hostname=None)>
+INFO:     174.1.60.1:47822 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:47836 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:47582 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:47594 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:47602 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:47604 - "POST /generate HTTP/1.1" 200 OK
+Unclosed connection
+client_connection: Connection<ConnectionKey(host='174.1.60.1', port=30000, is_ssl=False, ssl=True, proxy=None, proxy_auth=None, proxy_headers_hash=None, server_hostname=None)>
+Unclosed connection
+client_connection: Connection<ConnectionKey(host='174.1.60.1', port=30000, is_ssl=False, ssl=True, proxy=None, proxy_auth=None, proxy_headers_hash=None, server_hostname=None)>
+Unclosed connection
+INFO:     174.1.60.1:47618 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:47626 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:47640 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:47644 - "POST /generate HTTP/1.1" 200 OK
+client_connection: Connection<ConnectionKey(host='174.1.60.1', port=30000, is_ssl=False, ssl=True, proxy=None, proxy_auth=None, proxy_headers_hash=None, server_hostname=None)>
+INFO:     174.1.60.1:50356 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:50366 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:50374 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:50376 - "POST /generate HTTP/1.1" 200 OK
+Unclosed connection
+client_connection: Connection<ConnectionKey(host='174.1.60.1', port=30000, is_ssl=False, ssl=True, proxy=None, proxy_auth=None, proxy_headers_hash=None, server_hostname=None)>
+INFO:     174.1.60.1:50388 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:50402 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:50414 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:46792 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:53680 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:53696 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:53708 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:53714 - "POST /generate HTTP/1.1" 200 OK
+Unclosed connection
+INFO:     174.1.60.1:53718 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:53728 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:53742 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:57586 - "POST /generate HTTP/1.1" 200 OK
+client_connection: Connection<ConnectionKey(host='174.1.60.1', port=30000, is_ssl=False, ssl=True, proxy=None, proxy_auth=None, proxy_headers_hash=None, server_hostname=None)>
+INFO:     174.1.60.1:53888 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:53898 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:53900 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:53906 - "POST /generate HTTP/1.1" 200 OK
+Unclosed connection
+client_connection: Connection<ConnectionKey(host='174.1.60.1', port=30000, is_ssl=False, ssl=True, proxy=None, proxy_auth=None, proxy_headers_hash=None, server_hostname=None)>
+INFO:     174.1.60.1:53920 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:53934 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:47682 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:47698 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:41500 - "GET /server_info HTTP/1.1" 200 OK
+INFO:     174.1.60.1:41510 - "GET /server_info HTTP/1.1" 200 OK
+INFO:     174.1.60.1:51428 - "GET /health HTTP/1.1" 200 OK
+INFO:     174.1.60.1:51434 - "GET /health HTTP/1.1" 200 OK
+INFO:     174.1.60.1:41938 - "GET /v1/models HTTP/1.1" 200 OK
+INFO:     174.1.60.1:41948 - "GET /v1/models HTTP/1.1" 200 OK
+INFO:     174.1.60.1:43850 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:52950 - "GET /server_info HTTP/1.1" 200 OK
+INFO:     174.1.60.1:52964 - "GET /server_info HTTP/1.1" 200 OK
+INFO:     174.1.60.1:40292 - "GET /health HTTP/1.1" 200 OK
+INFO:     174.1.60.1:40304 - "GET /server_info HTTP/1.1" 200 OK
+INFO:     174.1.60.1:41682 - "GET /health HTTP/1.1" 200 OK
+INFO:     174.1.60.1:54172 - "GET /v1/models HTTP/1.1" 200 OK
+INFO:     174.1.60.1:54176 - "GET /v1/models HTTP/1.1" 200 OK
+INFO:     174.1.60.1:43036 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:43038 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:43040 - "POST /generate HTTP/1.1" 200 OK
+Unclosed connection
+client_connection: Connection<ConnectionKey(host='174.1.60.1', port=30000, is_ssl=False, ssl=True, proxy=None, proxy_auth=None, proxy_headers_hash=None, server_hostname=None)>
+Unclosed connection
+client_connection: Connection<ConnectionKey(host='174.1.60.1', port=30000, is_ssl=False, ssl=True, proxy=None, proxy_auth=None, proxy_headers_hash=None, server_hostname=None)>
+Unclosed connection
+INFO:     174.1.60.1:43056 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:43068 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:43072 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:43086 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:43100 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:46890 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:46894 - "POST /generate HTTP/1.1" 200 OK
+client_connection: Connection<ConnectionKey(host='174.1.60.1', port=30000, is_ssl=False, ssl=True, proxy=None, proxy_auth=None, proxy_headers_hash=None, server_hostname=None)>
+Unclosed connection
+client_connection: Connection<ConnectionKey(host='174.1.60.1', port=30000, is_ssl=False, ssl=True, proxy=None, proxy_auth=None, proxy_headers_hash=None, server_hostname=None)>
+INFO:     174.1.60.1:46904 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:46914 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:46924 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:46938 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:46942 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:46956 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:52362 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:43772 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:43778 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:43794 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:32984 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:32994 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:33002 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:33006 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:33016 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:56076 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:45176 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:54036 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:54064 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:54068 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:54074 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:54088 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:58126 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:58130 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:35214 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:55756 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:41930 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:35350 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:35354 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:35368 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:59190 - "GET /server_info HTTP/1.1" 200 OK
+INFO:     174.1.60.1:59196 - "GET /server_info HTTP/1.1" 200 OK
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-c8-20260827-1340/metadata/fatal_pattern_scan.txt b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-c8-20260827-1340/metadata/fatal_pattern_scan.txt
new file mode 100644
index 0000000..e69de29
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-c8-20260827-1340/metadata/router_server_info.json b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-c8-20260827-1340/metadata/router_server_info.json
new file mode 100644
index 0000000..5640a7e
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-c8-20260827-1340/metadata/router_server_info.json
@@ -0,0 +1 @@
+{"internal_states":[{"model_path":"/data/hf_models/Kimi-K3","tokenizer_path":"/data/hf_models/Kimi-K3","tokenizer_mode":"auto","tokenizer_backend":"huggingface","tokenizer_worker_num":1,"detokenizer_worker_num":1,"skip_tokenizer_init":false,"load_format":"auto","model_loader_extra_config":"{}","trust_remote_code":true,"context_length":null,"is_embedding":false,"enable_multimodal":null,"revision":null,"model_impl":"auto","model_config_parser":"auto","json_model_override_args":"{}","dtype":"auto","quantization":null,"quantization_param_path":null,"kv_cache_dtype":"auto","enable_fp32_lm_head":false,"modelopt_quant":null,"modelopt_checkpoint_restore_path":null,"modelopt_checkpoint_save_path":null,"modelopt_export_path":null,"quantize_and_serve":false,"rl_quant_profile":null,"enable_tf32_matmul":false,"mem_fraction_static":0.88,"max_running_requests":null,"max_queued_requests":null,"max_total_tokens":null,"chunked_prefill_size":8192,"enable_dynamic_chunking":false,"max_prefill_tokens":16384,"prefill_max_requests":null,"schedule_policy":"fcfs","enable_priority_scheduling":false,"disable_priority_preemption":false,"default_priority_value":null,"abort_on_priority_when_disabled":false,"schedule_low_priority_values_first":false,"priority_scheduling_preemption_threshold":10,"retraction_policy":"length","schedule_conservativeness":1.0,"page_size":64,"swa_full_tokens_ratio":0.8,"disable_hybrid_swa_memory":false,"radix_eviction_policy":"lru","prefill_only_disable_kv_cache":false,"disable_radix_cache":true,"enable_page_major_kv_layout":false,"enable_unified_memory":false,"disable_chunked_prefix_cache":true,"disable_overlap_schedule":false,"num_continuous_decode_steps":1,"scheduler_recv_interval":1,"enable_mixed_chunk":false,"nccl_port":null,"dist_timeout":3600,"dist_init_addr":"174.1.60.5:20000","nnodes":4,"node_rank":0,"tp_size":32,"dcp_size":1,"pp_size":1,"pp_max_micro_batch_size":181,"pp_async_batch_depth":0,"dp_size":1,"load_balance_method":"round_robin","attn_cp_size":1,"moe_dp_size":1,"dcp_comm_backend":"ag_rs","dcp_replicate_q_proj":null,"dwdp_size":1,"enable_prefill_cp":false,"cp_strategy":null,"enable_dsa_cache_layer_split":false,"enable_dsa_prefill_context_parallel":false,"dsa_prefill_cp_mode":"round-robin-split","enable_prefill_context_parallel":false,"prefill_cp_mode":"in-seq-split","enable_cp_decode_attn_tp":false,"enable_dp_attention":false,"enable_dp_attention_local_control_broadcast":false,"enable_dp_lm_head":false,"enable_attn_tp_input_scattered":false,"disable_attn_tp_gather":false,"enable_p2p_check":false,"device":"cuda","base_gpu_id":0,"gpu_id_step":1,"random_seed":369235288,"watchdog_timeout":300,"soft_watchdog_timeout":null,"sleep_on_idle":false,"use_ray":false,"numa_node":null,"gc_threshold":null,"host":"0.0.0.0","port":30000,"fastapi_root_path":"","smg_grpc_mode":false,"grpc_mode":false,"grpc_port":null,"sidecar":null,"sidecar_args":null,"skip_server_warmup":false,"warmups":null,"enable_http2":false,"ssl_keyfile":null,"ssl_certfile":null,"ssl_ca_certs":null,"ssl_keyfile_password":null,"enable_ssl_refresh":false,"api_key":null,"admin_api_key":null,"served_model_name":"kimi-k3","weight_version":"default","chat_template":null,"hf_chat_template_name":null,"completion_template":null,"file_storage_path":"sglang_storage","enable_cache_report":false,"reasoning_parser":null,"default_chat_template_kwargs":null,"strip_thinking_cache":false,"enable_strict_thinking":false,"tool_call_parser":null,"tool_server":null,"sampling_defaults":"model","asr_max_buffer_seconds":60,"asr_max_concurrent_sessions":32,"preferred_sampling_params":null,"allow_auto_truncate":false,"stream_interval":1,"batch_notify_size":16,"stream_response_default_include_usage":false,"incremental_streaming_output":false,"enable_streaming_session":false,"enable_session_radix_cache":false,"log_level":"info","log_level_http":null,"log_requests":false,"log_requests_level":2,"log_requests_format":"text","log_requests_target":null,"uvicorn_access_log_exclude_prefixes":[],"crash_dump_folder":null,"show_time_cost":false,"enable_metrics":false,"smg_http_sidecar_port":null,"enable_mfu_metrics":false,"enable_metrics_for_all_schedulers":false,"load_snapshot_publish_interval":15,"tokenizer_metrics_custom_labels_header":"x-custom-labels","tokenizer_metrics_allowed_custom_labels":null,"extra_metric_labels":null,"bucket_time_to_first_token":null,"bucket_inter_token_latency":null,"bucket_e2e_request_latency":null,"prompt_tokens_buckets":null,"generation_tokens_buckets":null,"gc_warning_threshold_secs":0.0,"decode_log_interval":40,"enable_request_time_stats_logging":false,"kv_events_config":null,"enable_forward_pass_metrics":false,"forward_pass_metrics_worker_id":"","forward_pass_metrics_ipc_name":null,"enable_trace":false,"trace_modules":"request","otlp_traces_endpoint":"localhost:4317","export_metrics_to_file":false,"export_metrics_to_file_dir":null,"stat_loggers":null,"constrained_json_whitespace_pattern":null,"constrained_json_disable_any_whitespace":false,"attention_backend":"triton","decode_attention_backend":null,"prefill_attention_backend":null,"sampling_backend":"flashinfer","grammar_backend":"xgrammar","radix_cache_backend":null,"mm_attention_backend":null,"fp8_gemm_runner_backend":"auto","fp4_gemm_runner_backend":"auto","bf16_gemm_backend":"auto","dsa_prefill_backend":null,"dsa_decode_backend":null,"dsa_paged_mqa_logits_backend":"auto","dsa_topk_backend":"sgl-kernel","disable_flashinfer_autotune":false,"flashinfer_autotune_skip_ops":null,"mamba_backend":"triton","cuda_graph_config":{"decode":{"backend":"full","max_bs":16,"bs":[1,2,4,8,12,16],"tc_compiler":"eager","full_prefill_max_req":null},"prefill":{"backend":"disabled","max_bs":2048,"bs":[4,8,12,16,20,24,28,32,48,64,80,96,112,128,144,160,176,192,208,224,240,256,288,320,352,384,416,448,480,512,576,640,704,768,832,896,960,1024,1280,1536,1792,2048],"tc_compiler":"eager","full_prefill_max_req":null}},"cuda_graph_backend_decode":null,"cuda_graph_backend_prefill":null,"cuda_graph_max_bs_decode":16,"cuda_graph_max_bs_prefill":null,"cuda_graph_bs_decode":null,"cuda_graph_bs_prefill":null,"cuda_graph_tc_compiler":null,"disable_prefill_cuda_graph":false,"disable_decode_cuda_graph":false,"disable_cuda_graph":false,"disable_cuda_graph_padding":false,"enable_profile_cuda_graph":false,"enable_cudagraph_gc":false,"debug_cuda_graph":false,"enable_layerwise_nvtx_marker":false,"enable_nccl_nvls":false,"enable_symm_mem":false,"triton_attention_reduce_in_fp32":false,"triton_attention_num_kv_splits":8,"triton_attention_split_tile_size":null,"flashinfer_mla_disable_ragged":false,"enable_fused_qk_norm_rope":false,"enable_precise_embedding_interpolation":false,"enable_fused_moe_sum_all_reduce":false,"enable_deepseek_v4_fp4_indexer":false,"disable_custom_all_reduce":false,"enable_mscclpp":false,"enable_torch_symm_mem":false,"enable_scattered_sconv":false,"pre_warm_nccl":false,"enable_quant_communications":false,"enable_flashinfer_allreduce_fusion":false,"enforce_disable_flashinfer_allreduce_fusion":false,"flashinfer_allreduce_fusion_backend":null,"enable_aiter_allreduce_fusion":false,"enable_torch_compile":false,"enable_torch_compile_debug_mode":false,"torch_compile_max_bs":32,"torchao_config":"","speculative_algorithm":null,"speculative_draft_model_path":null,"speculative_draft_model_revision":null,"speculative_draft_load_format":null,"speculative_num_steps":null,"speculative_eagle_topk":null,"speculative_num_draft_tokens":null,"speculative_dflash_block_size":null,"speculative_dspark_block_size":null,"speculative_dspark_sps_table_path":null,"speculative_dspark_confidence_sts_path":null,"speculative_dspark_align_verify_tokens_to_graph_tier":false,"speculative_accept_threshold_single":1.0,"speculative_accept_threshold_acc":1.0,"speculative_use_rejection_sampling":false,"speculative_token_map":null,"speculative_attention_mode":"prefill","speculative_draft_attention_backend":null,"speculative_draft_window_size":null,"speculative_moe_runner_backend":"marlin","speculative_moe_a2a_backend":null,"speculative_draft_model_quantization":null,"speculative_skip_dp_mlp_sync":false,"enable_multi_layer_eagle":false,"speculative_adaptive":false,"speculative_adaptive_config":null,"decoupled_spec_bind_endpoint":null,"decoupled_spec_connect_endpoints":null,"decoupled_spec_rank":null,"decoupled_spec_role":"null","spec_trace_dir":null,"speculative_ngram_min_bfs_breadth":1,"speculative_ngram_max_bfs_breadth":10,"speculative_ngram_match_type":"BFS","speculative_ngram_max_trie_depth":18,"speculative_ngram_capacity":10000000,"speculative_ngram_external_corpus_path":null,"speculative_ngram_external_sam_budget":0,"speculative_ngram_external_corpus_max_tokens":10000000,"ep_size":32,"moe_a2a_backend":"none","moe_runner_backend":"marlin","flashinfer_mxfp4_moe_precision":"default","deepep_mode":"auto","fuseep_mode":2,"deepep_dispatcher_output_dtype":"auto","ep_num_redundant_experts":0,"ep_dispatch_algorithm":null,"init_expert_location":"trivial","enable_eplb":false,"eplb_algorithm":"auto","eplb_rebalance_num_iterations":1000,"eplb_rebalance_layers_per_chunk":null,"eplb_min_rebalancing_utilization_threshold":1.0,"expert_distribution_recorder_mode":null,"expert_distribution_recorder_buffer_size":1000,"enable_expert_distribution_metrics":false,"deepep_config":null,"moe_dense_tp_size":null,"elastic_ep_backend":null,"enable_elastic_expert_backup":false,"mooncake_ib_device":null,"enable_waterfill":false,"ep_join_mode":null,"ep_join_rank_offset":0,"elastic_ep_initial_size":null,"max_ep_size":null,"elastic_ep_scale_timeout":600,"elastic_ep_rejoin":false,"disable_flashinfer_cutlass_moe_fp4_allgather":false,"disable_shared_experts_fusion":false,"enforce_shared_experts_fusion":false,"max_mamba_cache_size":181,"mamba_ssm_dtype":null,"mamba_max_states_per_path":-1,"enable_mamba_cache_stochastic_rounding":false,"mamba_cache_philox_rounds":0,"mamba_full_memory_ratio":0.21,"mamba_radix_cache_strategy":"extra_buffer_lazy","uses_mamba_radix_cache":false,"mamba_track_interval":256,"enable_int8_mamba_checkpoint":false,"int8_mamba_ckpt_size":null,"linear_attn_backend":"triton","linear_attn_decode_backend":null,"linear_attn_prefill_backend":null,"linear_attn_verify_backend":null,"enable_linear_replayssm":false,"linear_replayssm_cache_len":16,"enable_linear_replayssm_spec":false,"enable_hierarchical_cache":false,"hicache_ratio":2.0,"hicache_size":0,"hicache_write_policy":"write_through","hicache_io_backend":"kernel","hicache_mem_layout":"page_first","hicache_storage_backend":null,"hicache_storage_prefetch_policy":"timeout","hicache_storage_backend_extra_config":null,"enable_hisparse":false,"hisparse_config":null,"enable_broadcast_mm_inputs_process":false,"enable_prefix_mm_cache":false,"mm_enable_dp_encoder":false,"mm_process_config":{},"mm_processor_worker_num":0,"mm_io_worker_num":0,"limit_mm_data_per_request":null,"enable_mm_global_cache":false,"disable_fast_image_processor":false,"mm_feature_transport":"cpu","keep_mm_feature_on_device":false,"enable_lora":null,"enable_lora_overlap_loading":null,"max_lora_rank":null,"lora_target_modules":null,"lora_paths":null,"max_loaded_loras":null,"max_loras_per_batch":8,"lora_eviction_policy":"lru","lora_backend":"csgmv","max_lora_chunk_size":16,"experts_shared_outer_loras":null,"lora_use_virtual_experts":false,"lora_strict_loading":false,"lora_drain_wait_threshold":0.0,"enable_two_batch_overlap":false,"enable_single_batch_overlap":false,"tbo_token_distribution_threshold":0.48,"cpu_offload_gb":0,"offload_group_size":-1,"offload_num_in_group":1,"offload_prefetch_step":1,"offload_mode":"cpu","enable_lmcache":false,"lmcache_config_file":null,"enable_flexkv":false,"flexkv_config_file":null,"kt_weight_path":null,"kt_method":"AMXINT4","kt_cpuinfer":null,"kt_threadpool_count":2,"kt_num_gpu_experts":null,"kt_max_deferred_experts_per_token":null,"dllm_algorithm":null,"dllm_algorithm_config":null,"dllm_fdfo":true,"disaggregation_mode":"decode","disaggregation_transfer_backend":"mooncake","disaggregation_bootstrap_port":28800,"disaggregation_ib_device":"mlx5_0,mlx5_1,mlx5_2,mlx5_3","disaggregation_decode_enable_radix_cache":false,"disaggregation_decode_enable_offload_kvcache":false,"num_reserved_decode_tokens":512,"disaggregation_decode_extra_slots":0,"disaggregation_decode_polling_interval":1,"optimistic_prefill_attempts":0,"encoder_only":false,"language_only":false,"encoder_transfer_backend":"zmq_to_tokenizer","encoder_urls":[],"encoder_bootstrap_port":8997,"encoder_register_urls":[],"enable_adaptive_dispatch_to_encoder":false,"enable_pdmux":false,"pdmux_config_path":null,"sm_group_num":8,"custom_weight_loader":[],"weight_loader_disable_mmap":false,"weight_loader_prefetch_checkpoints":false,"weight_loader_prefetch_num_threads":4,"weight_loader_drop_cache_after_load":false,"remote_instance_weight_loader_seed_instance_ip":null,"remote_instance_weight_loader_seed_instance_service_port":null,"remote_instance_weight_loader_send_weights_group_ports":null,"remote_instance_weight_loader_backend":"nccl","remote_instance_weight_loader_start_seed_via_transfer_engine":false,"engine_info_bootstrap_port":6789,"modelexpress_config":null,"download_dir":null,"model_checksum":null,"delete_ckpt_after_loading":false,"decrypted_config_file":null,"decrypted_draft_config_file":null,"checkpoint_engine_wait_weights_before_ready":false,"enable_prefill_delayer":false,"prefill_delayer_max_delay_passes":30,"prefill_delayer_token_usage_low_watermark":null,"prefill_delayer_forward_passes_buckets":null,"prefill_delayer_wait_seconds_buckets":null,"prefill_delayer_queue_min_ratio":null,"prefill_delayer_max_delay_ms":null,"min_free_slots_delay":null,"enable_deterministic_inference":false,"rl_on_policy_target":null,"kv_canary":"none","kv_canary_real_data":"none","kv_canary_sweep_interval":0,"enable_dynamic_batch_tokenizer":false,"dynamic_batch_tokenizer_batch_size":32,"dynamic_batch_tokenizer_batch_timeout":0.002,"enable_tokenizer_batch_encode":false,"disable_tokenizer_batch_decode":false,"debug_tensor_dump_output_folder":null,"debug_tensor_dump_layers":null,"debug_tensor_dump_input_file":null,"enable_memory_saver":false,"enable_weights_cpu_backup":false,"enable_draft_weights_cpu_backup":false,"enable_custom_logit_processor":false,"enable_return_hidden_states":false,"enable_return_routed_experts":false,"enable_return_indexer_topk":false,"disable_outlines_disk_cache":false,"enable_mis":false,"weight_cache_mode":"off","weight_cache_socket":null,"weight_cache_timeout":1800,"forward_hooks":null,"msprobe_dump_config":null,"_resolved_overrides":[["_sampling_backend_default",{"sampling_backend":"flashinfer"}],["_attention_backend_default",{"attention_backend":"triton"}],["_data_parallelism_defaults",{"enable_dp_attention":false,"enable_dp_lm_head":false}],["_speculative_moe_runner_default",{"speculative_moe_runner_backend":"marlin"}]],"grpc_worker_threads":4,"_quantization_explicitly_unset":false,"_cuda_graph_config_locked":[["decode","max_bs"]],"_declarations_materialized":true,"_runtime_mutations":[["model_runner.chunked_prefix_cache_gate",{"disable_chunked_prefix_cache":true}],["mamba_pool.memory_budget",{"max_mamba_cache_size":181}]],"_in_override":false,"_mx_config_cache":{},"max_speculative_num_draft_tokens":null,"last_gen_throughput":48.2592441209455,"memory_usage":{"weight":58.57,"kvcache":11.34,"token_capacity":440256,"graph":0.6},"effective_max_running_requests_per_dp":181}],"prefill":[{"model_path":"/data/hf_models/Kimi-K3","tokenizer_path":"/data/hf_models/Kimi-K3","tokenizer_mode":"auto","tokenizer_backend":"huggingface","tokenizer_worker_num":1,"detokenizer_worker_num":1,"skip_tokenizer_init":false,"load_format":"auto","model_loader_extra_config":"{}","trust_remote_code":true,"context_length":null,"is_embedding":false,"enable_multimodal":null,"revision":null,"model_impl":"auto","model_config_parser":"auto","json_model_override_args":"{}","dtype":"auto","quantization":null,"quantization_param_path":null,"kv_cache_dtype":"auto","enable_fp32_lm_head":false,"modelopt_quant":null,"modelopt_checkpoint_restore_path":null,"modelopt_checkpoint_save_path":null,"modelopt_export_path":null,"quantize_and_serve":false,"rl_quant_profile":null,"enable_tf32_matmul":false,"mem_fraction_static":0.88,"max_running_requests":null,"max_queued_requests":null,"max_total_tokens":null,"chunked_prefill_size":8192,"enable_dynamic_chunking":false,"max_prefill_tokens":16384,"prefill_max_requests":null,"schedule_policy":"fcfs","enable_priority_scheduling":false,"disable_priority_preemption":false,"default_priority_value":null,"abort_on_priority_when_disabled":false,"schedule_low_priority_values_first":false,"priority_scheduling_preemption_threshold":10,"retraction_policy":"length","schedule_conservativeness":1.0,"page_size":64,"swa_full_tokens_ratio":0.8,"disable_hybrid_swa_memory":false,"radix_eviction_policy":"lru","prefill_only_disable_kv_cache":false,"disable_radix_cache":true,"enable_page_major_kv_layout":false,"enable_unified_memory":false,"disable_chunked_prefix_cache":false,"disable_overlap_schedule":true,"num_continuous_decode_steps":1,"scheduler_recv_interval":1,"enable_mixed_chunk":false,"nccl_port":null,"dist_timeout":3600,"dist_init_addr":"174.1.60.1:20000","nnodes":4,"node_rank":0,"tp_size":4,"dcp_size":1,"pp_size":8,"pp_max_micro_batch_size":null,"pp_async_batch_depth":0,"dp_size":1,"load_balance_method":"follow_bootstrap_room","attn_cp_size":1,"moe_dp_size":1,"dcp_comm_backend":"ag_rs","dcp_replicate_q_proj":null,"dwdp_size":1,"enable_prefill_cp":false,"cp_strategy":null,"enable_dsa_cache_layer_split":false,"enable_dsa_prefill_context_parallel":false,"dsa_prefill_cp_mode":"round-robin-split","enable_prefill_context_parallel":false,"prefill_cp_mode":"in-seq-split","enable_cp_decode_attn_tp":false,"enable_dp_attention":false,"enable_dp_attention_local_control_broadcast":false,"enable_dp_lm_head":false,"enable_attn_tp_input_scattered":false,"disable_attn_tp_gather":false,"enable_p2p_check":false,"device":"cuda","base_gpu_id":0,"gpu_id_step":1,"random_seed":984607766,"watchdog_timeout":300,"soft_watchdog_timeout":null,"sleep_on_idle":false,"use_ray":false,"custom_sigquit_handler":null,"numa_node":null,"gc_threshold":null,"host":"0.0.0.0","port":30000,"fastapi_root_path":"","smg_grpc_mode":false,"grpc_mode":false,"grpc_port":null,"sidecar":null,"sidecar_args":null,"skip_server_warmup":false,"warmups":null,"enable_http2":false,"ssl_keyfile":null,"ssl_certfile":null,"ssl_ca_certs":null,"ssl_keyfile_password":null,"enable_ssl_refresh":false,"api_key":null,"admin_api_key":null,"served_model_name":"kimi-k3","weight_version":"default","chat_template":null,"hf_chat_template_name":null,"completion_template":null,"file_storage_path":"sglang_storage","enable_cache_report":false,"reasoning_parser":null,"default_chat_template_kwargs":null,"strip_thinking_cache":false,"enable_strict_thinking":false,"tool_call_parser":null,"tool_server":null,"sampling_defaults":"model","asr_max_buffer_seconds":60,"asr_max_concurrent_sessions":32,"preferred_sampling_params":null,"allow_auto_truncate":false,"stream_interval":1,"batch_notify_size":16,"stream_response_default_include_usage":false,"incremental_streaming_output":false,"enable_streaming_session":false,"enable_session_radix_cache":false,"log_level":"info","log_level_http":null,"log_requests":false,"log_requests_level":2,"log_requests_format":"text","log_requests_target":null,"uvicorn_access_log_exclude_prefixes":[],"crash_dump_folder":null,"show_time_cost":false,"enable_metrics":false,"smg_http_sidecar_port":null,"enable_mfu_metrics":false,"enable_metrics_for_all_schedulers":false,"load_snapshot_publish_interval":15,"tokenizer_metrics_custom_labels_header":"x-custom-labels","tokenizer_metrics_allowed_custom_labels":null,"extra_metric_labels":null,"bucket_time_to_first_token":null,"bucket_inter_token_latency":null,"bucket_e2e_request_latency":null,"prompt_tokens_buckets":null,"generation_tokens_buckets":null,"gc_warning_threshold_secs":0.0,"decode_log_interval":40,"enable_request_time_stats_logging":false,"kv_events_config":null,"enable_forward_pass_metrics":false,"forward_pass_metrics_worker_id":"","forward_pass_metrics_ipc_name":null,"enable_trace":false,"trace_modules":"request","otlp_traces_endpoint":"localhost:4317","export_metrics_to_file":false,"export_metrics_to_file_dir":null,"stat_loggers":null,"constrained_json_whitespace_pattern":null,"constrained_json_disable_any_whitespace":false,"attention_backend":"triton","decode_attention_backend":null,"prefill_attention_backend":null,"sampling_backend":"flashinfer","grammar_backend":"xgrammar","radix_cache_backend":null,"mm_attention_backend":null,"fp8_gemm_runner_backend":"auto","fp4_gemm_runner_backend":"auto","bf16_gemm_backend":"auto","dsa_prefill_backend":null,"dsa_decode_backend":null,"dsa_paged_mqa_logits_backend":"auto","dsa_topk_backend":"sgl-kernel","disable_flashinfer_autotune":false,"flashinfer_autotune_skip_ops":null,"mamba_backend":"triton","cuda_graph_config":{"decode":{"backend":"disabled","max_bs":16,"bs":[1,2,4,8,12,16],"tc_compiler":"eager","full_prefill_max_req":null},"prefill":{"backend":"disabled","max_bs":2048,"bs":[4,8,12,16,20,24,28,32,48,64,80,96,112,128,144,160,176,192,208,224,240,256,288,320,352,384,416,448,480,512,576,640,704,768,832,896,960,1024,1280,1536,1792,2048],"tc_compiler":"eager","full_prefill_max_req":null}},"cuda_graph_backend_decode":null,"cuda_graph_backend_prefill":null,"cuda_graph_max_bs_decode":16,"cuda_graph_max_bs_prefill":null,"cuda_graph_bs_decode":null,"cuda_graph_bs_prefill":null,"cuda_graph_tc_compiler":null,"disable_prefill_cuda_graph":false,"disable_decode_cuda_graph":false,"disable_cuda_graph":false,"disable_cuda_graph_padding":false,"enable_profile_cuda_graph":false,"enable_cudagraph_gc":false,"debug_cuda_graph":false,"enable_layerwise_nvtx_marker":false,"enable_nccl_nvls":false,"enable_symm_mem":false,"triton_attention_reduce_in_fp32":false,"triton_attention_num_kv_splits":8,"triton_attention_split_tile_size":null,"flashinfer_mla_disable_ragged":false,"enable_fused_qk_norm_rope":false,"enable_precise_embedding_interpolation":false,"enable_fused_moe_sum_all_reduce":false,"enable_deepseek_v4_fp4_indexer":false,"disable_custom_all_reduce":false,"enable_mscclpp":false,"enable_torch_symm_mem":false,"enable_scattered_sconv":false,"pre_warm_nccl":false,"enable_quant_communications":false,"enable_flashinfer_allreduce_fusion":false,"enforce_disable_flashinfer_allreduce_fusion":false,"flashinfer_allreduce_fusion_backend":null,"enable_aiter_allreduce_fusion":false,"enable_torch_compile":false,"enable_torch_compile_debug_mode":false,"torch_compile_max_bs":32,"torchao_config":"","speculative_algorithm":null,"speculative_draft_model_path":null,"speculative_draft_model_revision":null,"speculative_draft_load_format":null,"speculative_num_steps":null,"speculative_eagle_topk":null,"speculative_num_draft_tokens":null,"speculative_dflash_block_size":null,"speculative_dspark_block_size":null,"speculative_dspark_sps_table_path":null,"speculative_dspark_confidence_sts_path":null,"speculative_dspark_align_verify_tokens_to_graph_tier":false,"speculative_accept_threshold_single":1.0,"speculative_accept_threshold_acc":1.0,"speculative_use_rejection_sampling":false,"speculative_token_map":null,"speculative_attention_mode":"prefill","speculative_draft_attention_backend":null,"speculative_draft_window_size":null,"speculative_moe_runner_backend":"flashinfer_mxfp4","speculative_moe_a2a_backend":null,"speculative_draft_model_quantization":null,"speculative_skip_dp_mlp_sync":false,"enable_multi_layer_eagle":false,"speculative_adaptive":false,"speculative_adaptive_config":null,"decoupled_spec_bind_endpoint":null,"decoupled_spec_connect_endpoints":null,"decoupled_spec_rank":null,"decoupled_spec_role":"null","spec_trace_dir":null,"speculative_ngram_min_bfs_breadth":1,"speculative_ngram_max_bfs_breadth":10,"speculative_ngram_match_type":"BFS","speculative_ngram_max_trie_depth":18,"speculative_ngram_capacity":10000000,"speculative_ngram_external_corpus_path":null,"speculative_ngram_external_sam_budget":0,"speculative_ngram_external_corpus_max_tokens":10000000,"ep_size":4,"moe_a2a_backend":"none","moe_runner_backend":"flashinfer_mxfp4","flashinfer_mxfp4_moe_precision":"default","deepep_mode":"auto","fuseep_mode":2,"deepep_dispatcher_output_dtype":"auto","ep_num_redundant_experts":0,"ep_dispatch_algorithm":null,"init_expert_location":"trivial","enable_eplb":false,"eplb_algorithm":"auto","eplb_rebalance_num_iterations":1000,"eplb_rebalance_layers_per_chunk":null,"eplb_min_rebalancing_utilization_threshold":1.0,"expert_distribution_recorder_mode":null,"expert_distribution_recorder_buffer_size":1000,"enable_expert_distribution_metrics":false,"deepep_config":null,"moe_dense_tp_size":null,"elastic_ep_backend":null,"enable_elastic_expert_backup":false,"mooncake_ib_device":null,"enable_waterfill":false,"ep_join_mode":null,"ep_join_rank_offset":0,"elastic_ep_initial_size":null,"max_ep_size":null,"elastic_ep_scale_timeout":600,"elastic_ep_rejoin":false,"disable_flashinfer_cutlass_moe_fp4_allgather":false,"disable_shared_experts_fusion":false,"enforce_shared_experts_fusion":false,"max_mamba_cache_size":null,"mamba_ssm_dtype":null,"mamba_max_states_per_path":-1,"enable_mamba_cache_stochastic_rounding":false,"mamba_cache_philox_rounds":0,"mamba_full_memory_ratio":0.36,"mamba_radix_cache_strategy":"extra_buffer_lazy","uses_mamba_radix_cache":false,"mamba_track_interval":256,"enable_int8_mamba_checkpoint":false,"int8_mamba_ckpt_size":null,"linear_attn_backend":"triton","linear_attn_decode_backend":null,"linear_attn_prefill_backend":null,"linear_attn_verify_backend":null,"enable_linear_replayssm":false,"linear_replayssm_cache_len":16,"enable_linear_replayssm_spec":false,"enable_hierarchical_cache":false,"hicache_ratio":2.0,"hicache_size":0,"hicache_write_policy":"write_through","hicache_io_backend":"kernel","hicache_mem_layout":"page_first","hicache_storage_backend":null,"hicache_storage_prefetch_policy":"timeout","hicache_storage_backend_extra_config":null,"enable_hisparse":false,"hisparse_config":null,"enable_broadcast_mm_inputs_process":false,"enable_prefix_mm_cache":false,"mm_enable_dp_encoder":false,"mm_process_config":{},"mm_processor_worker_num":0,"mm_io_worker_num":0,"limit_mm_data_per_request":null,"enable_mm_global_cache":false,"disable_fast_image_processor":false,"mm_feature_transport":"cpu","keep_mm_feature_on_device":false,"enable_lora":null,"enable_lora_overlap_loading":null,"max_lora_rank":null,"lora_target_modules":null,"lora_paths":null,"max_loaded_loras":null,"max_loras_per_batch":8,"lora_eviction_policy":"lru","lora_backend":"csgmv","max_lora_chunk_size":16,"experts_shared_outer_loras":null,"lora_use_virtual_experts":false,"lora_strict_loading":false,"lora_drain_wait_threshold":0.0,"enable_two_batch_overlap":false,"enable_single_batch_overlap":false,"tbo_token_distribution_threshold":0.48,"cpu_offload_gb":0,"offload_group_size":-1,"offload_num_in_group":1,"offload_prefetch_step":1,"offload_mode":"cpu","enable_lmcache":false,"lmcache_config_file":null,"enable_flexkv":false,"flexkv_config_file":null,"kt_weight_path":null,"kt_method":"AMXINT4","kt_cpuinfer":null,"kt_threadpool_count":2,"kt_num_gpu_experts":null,"kt_max_deferred_experts_per_token":null,"dllm_algorithm":null,"dllm_algorithm_config":null,"dllm_fdfo":true,"disaggregation_mode":"prefill","disaggregation_transfer_backend":"mooncake","disaggregation_bootstrap_port":28800,"disaggregation_ib_device":"mlx5_0,mlx5_1,mlx5_2,mlx5_3","disaggregation_decode_enable_radix_cache":false,"disaggregation_decode_enable_offload_kvcache":false,"num_reserved_decode_tokens":512,"disaggregation_decode_extra_slots":null,"disaggregation_decode_polling_interval":1,"optimistic_prefill_attempts":0,"encoder_only":false,"language_only":false,"encoder_transfer_backend":"zmq_to_tokenizer","encoder_urls":[],"encoder_bootstrap_port":8997,"encoder_register_urls":[],"enable_adaptive_dispatch_to_encoder":false,"enable_pdmux":false,"pdmux_config_path":null,"sm_group_num":8,"custom_weight_loader":[],"weight_loader_disable_mmap":false,"weight_loader_prefetch_checkpoints":false,"weight_loader_prefetch_num_threads":4,"weight_loader_drop_cache_after_load":false,"remote_instance_weight_loader_seed_instance_ip":null,"remote_instance_weight_loader_seed_instance_service_port":null,"remote_instance_weight_loader_send_weights_group_ports":null,"remote_instance_weight_loader_backend":"nccl","remote_instance_weight_loader_start_seed_via_transfer_engine":false,"engine_info_bootstrap_port":6789,"modelexpress_config":null,"download_dir":null,"model_checksum":null,"delete_ckpt_after_loading":false,"decrypted_config_file":null,"decrypted_draft_config_file":null,"checkpoint_engine_wait_weights_before_ready":false,"enable_prefill_delayer":false,"prefill_delayer_max_delay_passes":30,"prefill_delayer_token_usage_low_watermark":null,"prefill_delayer_forward_passes_buckets":null,"prefill_delayer_wait_seconds_buckets":null,"prefill_delayer_queue_min_ratio":null,"prefill_delayer_max_delay_ms":null,"min_free_slots_delay":null,"enable_deterministic_inference":false,"rl_on_policy_target":null,"kv_canary":"none","kv_canary_real_data":"none","kv_canary_sweep_interval":0,"enable_dynamic_batch_tokenizer":false,"dynamic_batch_tokenizer_batch_size":32,"dynamic_batch_tokenizer_batch_timeout":0.002,"enable_tokenizer_batch_encode":false,"disable_tokenizer_batch_decode":false,"debug_tensor_dump_output_folder":null,"debug_tensor_dump_layers":null,"debug_tensor_dump_input_file":null,"enable_memory_saver":false,"enable_weights_cpu_backup":false,"enable_draft_weights_cpu_backup":false,"enable_custom_logit_processor":false,"enable_return_hidden_states":false,"enable_return_routed_experts":false,"enable_return_indexer_topk":false,"disable_outlines_disk_cache":false,"enable_mis":false,"weight_cache_mode":"off","weight_cache_socket":null,"weight_cache_timeout":1800,"forward_hooks":null,"msprobe_dump_config":null,"status":"ready","max_total_num_tokens":3801408,"max_req_input_len":1048570,"internal_states":[{"model_path":"/data/hf_models/Kimi-K3","tokenizer_path":"/data/hf_models/Kimi-K3","tokenizer_mode":"auto","tokenizer_backend":"huggingface","tokenizer_worker_num":1,"detokenizer_worker_num":1,"skip_tokenizer_init":false,"load_format":"auto","model_loader_extra_config":"{}","trust_remote_code":true,"context_length":null,"is_embedding":false,"enable_multimodal":null,"revision":null,"model_impl":"auto","model_config_parser":"auto","json_model_override_args":"{}","dtype":"auto","quantization":null,"quantization_param_path":null,"kv_cache_dtype":"auto","enable_fp32_lm_head":false,"modelopt_quant":null,"modelopt_checkpoint_restore_path":null,"modelopt_checkpoint_save_path":null,"modelopt_export_path":null,"quantize_and_serve":false,"rl_quant_profile":null,"enable_tf32_matmul":false,"mem_fraction_static":0.88,"max_running_requests":null,"max_queued_requests":null,"max_total_tokens":null,"chunked_prefill_size":8192,"enable_dynamic_chunking":false,"max_prefill_tokens":16384,"prefill_max_requests":null,"schedule_policy":"fcfs","enable_priority_scheduling":false,"disable_priority_preemption":false,"default_priority_value":null,"abort_on_priority_when_disabled":false,"schedule_low_priority_values_first":false,"priority_scheduling_preemption_threshold":10,"retraction_policy":"length","schedule_conservativeness":1.0,"page_size":64,"swa_full_tokens_ratio":0.8,"disable_hybrid_swa_memory":false,"radix_eviction_policy":"lru","prefill_only_disable_kv_cache":false,"disable_radix_cache":true,"enable_page_major_kv_layout":false,"enable_unified_memory":false,"disable_chunked_prefix_cache":true,"disable_overlap_schedule":true,"num_continuous_decode_steps":1,"scheduler_recv_interval":1,"enable_mixed_chunk":false,"nccl_port":null,"dist_timeout":3600,"dist_init_addr":"174.1.60.1:20000","nnodes":4,"node_rank":0,"tp_size":4,"dcp_size":1,"pp_size":8,"pp_max_micro_batch_size":6,"pp_async_batch_depth":0,"dp_size":1,"load_balance_method":"follow_bootstrap_room","attn_cp_size":1,"moe_dp_size":1,"dcp_comm_backend":"ag_rs","dcp_replicate_q_proj":null,"dwdp_size":1,"enable_prefill_cp":false,"cp_strategy":null,"enable_dsa_cache_layer_split":false,"enable_dsa_prefill_context_parallel":false,"dsa_prefill_cp_mode":"round-robin-split","enable_prefill_context_parallel":false,"prefill_cp_mode":"in-seq-split","enable_cp_decode_attn_tp":false,"enable_dp_attention":false,"enable_dp_attention_local_control_broadcast":false,"enable_dp_lm_head":false,"enable_attn_tp_input_scattered":false,"disable_attn_tp_gather":false,"enable_p2p_check":false,"device":"cuda","base_gpu_id":0,"gpu_id_step":1,"random_seed":984607766,"watchdog_timeout":300,"soft_watchdog_timeout":null,"sleep_on_idle":false,"use_ray":false,"numa_node":null,"gc_threshold":null,"host":"0.0.0.0","port":30000,"fastapi_root_path":"","smg_grpc_mode":false,"grpc_mode":false,"grpc_port":null,"sidecar":null,"sidecar_args":null,"skip_server_warmup":false,"warmups":null,"enable_http2":false,"ssl_keyfile":null,"ssl_certfile":null,"ssl_ca_certs":null,"ssl_keyfile_password":null,"enable_ssl_refresh":false,"api_key":null,"admin_api_key":null,"served_model_name":"kimi-k3","weight_version":"default","chat_template":null,"hf_chat_template_name":null,"completion_template":null,"file_storage_path":"sglang_storage","enable_cache_report":false,"reasoning_parser":null,"default_chat_template_kwargs":null,"strip_thinking_cache":false,"enable_strict_thinking":false,"tool_call_parser":null,"tool_server":null,"sampling_defaults":"model","asr_max_buffer_seconds":60,"asr_max_concurrent_sessions":32,"preferred_sampling_params":null,"allow_auto_truncate":false,"stream_interval":1,"batch_notify_size":16,"stream_response_default_include_usage":false,"incremental_streaming_output":false,"enable_streaming_session":false,"enable_session_radix_cache":false,"log_level":"info","log_level_http":null,"log_requests":false,"log_requests_level":2,"log_requests_format":"text","log_requests_target":null,"uvicorn_access_log_exclude_prefixes":[],"crash_dump_folder":null,"show_time_cost":false,"enable_metrics":false,"smg_http_sidecar_port":null,"enable_mfu_metrics":false,"enable_metrics_for_all_schedulers":false,"load_snapshot_publish_interval":15,"tokenizer_metrics_custom_labels_header":"x-custom-labels","tokenizer_metrics_allowed_custom_labels":null,"extra_metric_labels":null,"bucket_time_to_first_token":null,"bucket_inter_token_latency":null,"bucket_e2e_request_latency":null,"prompt_tokens_buckets":null,"generation_tokens_buckets":null,"gc_warning_threshold_secs":0.0,"decode_log_interval":40,"enable_request_time_stats_logging":false,"kv_events_config":null,"enable_forward_pass_metrics":false,"forward_pass_metrics_worker_id":"","forward_pass_metrics_ipc_name":null,"enable_trace":false,"trace_modules":"request","otlp_traces_endpoint":"localhost:4317","export_metrics_to_file":false,"export_metrics_to_file_dir":null,"stat_loggers":null,"constrained_json_whitespace_pattern":null,"constrained_json_disable_any_whitespace":false,"attention_backend":"triton","decode_attention_backend":null,"prefill_attention_backend":null,"sampling_backend":"flashinfer","grammar_backend":"xgrammar","radix_cache_backend":null,"mm_attention_backend":null,"fp8_gemm_runner_backend":"auto","fp4_gemm_runner_backend":"auto","bf16_gemm_backend":"auto","dsa_prefill_backend":null,"dsa_decode_backend":null,"dsa_paged_mqa_logits_backend":"auto","dsa_topk_backend":"sgl-kernel","disable_flashinfer_autotune":false,"flashinfer_autotune_skip_ops":null,"mamba_backend":"triton","cuda_graph_config":{"decode":{"backend":"disabled","max_bs":16,"bs":[1,2,4,8,12,16],"tc_compiler":"eager","full_prefill_max_req":null},"prefill":{"backend":"disabled","max_bs":2048,"bs":[4,8,12,16,20,24,28,32,48,64,80,96,112,128,144,160,176,192,208,224,240,256,288,320,352,384,416,448,480,512,576,640,704,768,832,896,960,1024,1280,1536,1792,2048],"tc_compiler":"eager","full_prefill_max_req":null}},"cuda_graph_backend_decode":null,"cuda_graph_backend_prefill":null,"cuda_graph_max_bs_decode":16,"cuda_graph_max_bs_prefill":null,"cuda_graph_bs_decode":null,"cuda_graph_bs_prefill":null,"cuda_graph_tc_compiler":null,"disable_prefill_cuda_graph":false,"disable_decode_cuda_graph":false,"disable_cuda_graph":false,"disable_cuda_graph_padding":false,"enable_profile_cuda_graph":false,"enable_cudagraph_gc":false,"debug_cuda_graph":false,"enable_layerwise_nvtx_marker":false,"enable_nccl_nvls":false,"enable_symm_mem":false,"triton_attention_reduce_in_fp32":false,"triton_attention_num_kv_splits":8,"triton_attention_split_tile_size":null,"flashinfer_mla_disable_ragged":false,"enable_fused_qk_norm_rope":false,"enable_precise_embedding_interpolation":false,"enable_fused_moe_sum_all_reduce":false,"enable_deepseek_v4_fp4_indexer":false,"disable_custom_all_reduce":false,"enable_mscclpp":false,"enable_torch_symm_mem":false,"enable_scattered_sconv":false,"pre_warm_nccl":false,"enable_quant_communications":false,"enable_flashinfer_allreduce_fusion":false,"enforce_disable_flashinfer_allreduce_fusion":false,"flashinfer_allreduce_fusion_backend":null,"enable_aiter_allreduce_fusion":false,"enable_torch_compile":false,"enable_torch_compile_debug_mode":false,"torch_compile_max_bs":32,"torchao_config":"","speculative_algorithm":null,"speculative_draft_model_path":null,"speculative_draft_model_revision":null,"speculative_draft_load_format":null,"speculative_num_steps":null,"speculative_eagle_topk":null,"speculative_num_draft_tokens":null,"speculative_dflash_block_size":null,"speculative_dspark_block_size":null,"speculative_dspark_sps_table_path":null,"speculative_dspark_confidence_sts_path":null,"speculative_dspark_align_verify_tokens_to_graph_tier":false,"speculative_accept_threshold_single":1.0,"speculative_accept_threshold_acc":1.0,"speculative_use_rejection_sampling":false,"speculative_token_map":null,"speculative_attention_mode":"prefill","speculative_draft_attention_backend":null,"speculative_draft_window_size":null,"speculative_moe_runner_backend":"flashinfer_mxfp4","speculative_moe_a2a_backend":null,"speculative_draft_model_quantization":null,"speculative_skip_dp_mlp_sync":false,"enable_multi_layer_eagle":false,"speculative_adaptive":false,"speculative_adaptive_config":null,"decoupled_spec_bind_endpoint":null,"decoupled_spec_connect_endpoints":null,"decoupled_spec_rank":null,"decoupled_spec_role":"null","spec_trace_dir":null,"speculative_ngram_min_bfs_breadth":1,"speculative_ngram_max_bfs_breadth":10,"speculative_ngram_match_type":"BFS","speculative_ngram_max_trie_depth":18,"speculative_ngram_capacity":10000000,"speculative_ngram_external_corpus_path":null,"speculative_ngram_external_sam_budget":0,"speculative_ngram_external_corpus_max_tokens":10000000,"ep_size":4,"moe_a2a_backend":"none","moe_runner_backend":"flashinfer_mxfp4","flashinfer_mxfp4_moe_precision":"default","deepep_mode":"auto","fuseep_mode":2,"deepep_dispatcher_output_dtype":"auto","ep_num_redundant_experts":0,"ep_dispatch_algorithm":null,"init_expert_location":"trivial","enable_eplb":false,"eplb_algorithm":"auto","eplb_rebalance_num_iterations":1000,"eplb_rebalance_layers_per_chunk":null,"eplb_min_rebalancing_utilization_threshold":1.0,"expert_distribution_recorder_mode":null,"expert_distribution_recorder_buffer_size":1000,"enable_expert_distribution_metrics":false,"deepep_config":null,"moe_dense_tp_size":null,"elastic_ep_backend":null,"enable_elastic_expert_backup":false,"mooncake_ib_device":null,"enable_waterfill":false,"ep_join_mode":null,"ep_join_rank_offset":0,"elastic_ep_initial_size":null,"max_ep_size":null,"elastic_ep_scale_timeout":600,"elastic_ep_rejoin":false,"disable_flashinfer_cutlass_moe_fp4_allgather":false,"disable_shared_experts_fusion":false,"enforce_shared_experts_fusion":false,"max_mamba_cache_size":54,"mamba_ssm_dtype":null,"mamba_max_states_per_path":-1,"enable_mamba_cache_stochastic_rounding":false,"mamba_cache_philox_rounds":0,"mamba_full_memory_ratio":0.36,"mamba_radix_cache_strategy":"extra_buffer_lazy","uses_mamba_radix_cache":false,"mamba_track_interval":256,"enable_int8_mamba_checkpoint":false,"int8_mamba_ckpt_size":null,"linear_attn_backend":"triton","linear_attn_decode_backend":null,"linear_attn_prefill_backend":null,"linear_attn_verify_backend":null,"enable_linear_replayssm":false,"linear_replayssm_cache_len":16,"enable_linear_replayssm_spec":false,"enable_hierarchical_cache":false,"hicache_ratio":2.0,"hicache_size":0,"hicache_write_policy":"write_through","hicache_io_backend":"kernel","hicache_mem_layout":"page_first","hicache_storage_backend":null,"hicache_storage_prefetch_policy":"timeout","hicache_storage_backend_extra_config":null,"enable_hisparse":false,"hisparse_config":null,"enable_broadcast_mm_inputs_process":false,"enable_prefix_mm_cache":false,"mm_enable_dp_encoder":false,"mm_process_config":{},"mm_processor_worker_num":0,"mm_io_worker_num":0,"limit_mm_data_per_request":null,"enable_mm_global_cache":false,"disable_fast_image_processor":false,"mm_feature_transport":"cpu","keep_mm_feature_on_device":false,"enable_lora":null,"enable_lora_overlap_loading":null,"max_lora_rank":null,"lora_target_modules":null,"lora_paths":null,"max_loaded_loras":null,"max_loras_per_batch":8,"lora_eviction_policy":"lru","lora_backend":"csgmv","max_lora_chunk_size":16,"experts_shared_outer_loras":null,"lora_use_virtual_experts":false,"lora_strict_loading":false,"lora_drain_wait_threshold":0.0,"enable_two_batch_overlap":false,"enable_single_batch_overlap":false,"tbo_token_distribution_threshold":0.48,"cpu_offload_gb":0,"offload_group_size":-1,"offload_num_in_group":1,"offload_prefetch_step":1,"offload_mode":"cpu","enable_lmcache":false,"lmcache_config_file":null,"enable_flexkv":false,"flexkv_config_file":null,"kt_weight_path":null,"kt_method":"AMXINT4","kt_cpuinfer":null,"kt_threadpool_count":2,"kt_num_gpu_experts":null,"kt_max_deferred_experts_per_token":null,"dllm_algorithm":null,"dllm_algorithm_config":null,"dllm_fdfo":true,"disaggregation_mode":"prefill","disaggregation_transfer_backend":"mooncake","disaggregation_bootstrap_port":28800,"disaggregation_ib_device":"mlx5_0,mlx5_1,mlx5_2,mlx5_3","disaggregation_decode_enable_radix_cache":false,"disaggregation_decode_enable_offload_kvcache":false,"num_reserved_decode_tokens":512,"disaggregation_decode_extra_slots":null,"disaggregation_decode_polling_interval":1,"optimistic_prefill_attempts":0,"encoder_only":false,"language_only":false,"encoder_transfer_backend":"zmq_to_tokenizer","encoder_urls":[],"encoder_bootstrap_port":8997,"encoder_register_urls":[],"enable_adaptive_dispatch_to_encoder":false,"enable_pdmux":false,"pdmux_config_path":null,"sm_group_num":8,"custom_weight_loader":[],"weight_loader_disable_mmap":false,"weight_loader_prefetch_checkpoints":false,"weight_loader_prefetch_num_threads":4,"weight_loader_drop_cache_after_load":false,"remote_instance_weight_loader_seed_instance_ip":null,"remote_instance_weight_loader_seed_instance_service_port":null,"remote_instance_weight_loader_send_weights_group_ports":null,"remote_instance_weight_loader_backend":"nccl","remote_instance_weight_loader_start_seed_via_transfer_engine":false,"engine_info_bootstrap_port":6789,"modelexpress_config":null,"download_dir":null,"model_checksum":null,"delete_ckpt_after_loading":false,"decrypted_config_file":null,"decrypted_draft_config_file":null,"checkpoint_engine_wait_weights_before_ready":false,"enable_prefill_delayer":false,"prefill_delayer_max_delay_passes":30,"prefill_delayer_token_usage_low_watermark":null,"prefill_delayer_forward_passes_buckets":null,"prefill_delayer_wait_seconds_buckets":null,"prefill_delayer_queue_min_ratio":null,"prefill_delayer_max_delay_ms":null,"min_free_slots_delay":null,"enable_deterministic_inference":false,"rl_on_policy_target":null,"kv_canary":"none","kv_canary_real_data":"none","kv_canary_sweep_interval":0,"enable_dynamic_batch_tokenizer":false,"dynamic_batch_tokenizer_batch_size":32,"dynamic_batch_tokenizer_batch_timeout":0.002,"enable_tokenizer_batch_encode":false,"disable_tokenizer_batch_decode":false,"debug_tensor_dump_output_folder":null,"debug_tensor_dump_layers":null,"debug_tensor_dump_input_file":null,"enable_memory_saver":false,"enable_weights_cpu_backup":false,"enable_draft_weights_cpu_backup":false,"enable_custom_logit_processor":false,"enable_return_hidden_states":false,"enable_return_routed_experts":false,"enable_return_indexer_topk":false,"disable_outlines_disk_cache":false,"enable_mis":false,"weight_cache_mode":"off","weight_cache_socket":null,"weight_cache_timeout":1800,"forward_hooks":null,"msprobe_dump_config":null,"_resolved_overrides":[["_sampling_backend_default",{"sampling_backend":"flashinfer"}],["_attention_backend_default",{"attention_backend":"triton"}],["_data_parallelism_defaults",{"enable_dp_attention":false,"enable_dp_lm_head":false}],["_pipeline_parallel_overlap_disable",{"disable_overlap_schedule":true}],["_speculative_moe_runner_default",{"speculative_moe_runner_backend":"flashinfer_mxfp4"}]],"grpc_worker_threads":4,"_quantization_explicitly_unset":false,"_cuda_graph_config_locked":[["decode","max_bs"]],"_declarations_materialized":true,"_runtime_mutations":[["model_runner.chunked_prefix_cache_gate",{"disable_chunked_prefix_cache":true}],["mamba_pool.memory_budget",{"max_mamba_cache_size":54}]],"_in_override":false,"_mx_config_cache":{},"max_speculative_num_draft_tokens":null,"last_gen_throughput":0.0,"memory_usage":{"weight":43.09,"kvcache":8.16,"token_capacity":3801408,"graph":0},"effective_max_running_requests_per_dp":54}],"version":"0.5.16","kv_events":null}],"decode":[{"model_path":"/data/hf_models/Kimi-K3","tokenizer_path":"/data/hf_models/Kimi-K3","tokenizer_mode":"auto","tokenizer_backend":"huggingface","tokenizer_worker_num":1,"detokenizer_worker_num":1,"skip_tokenizer_init":false,"load_format":"auto","model_loader_extra_config":"{}","trust_remote_code":true,"context_length":null,"is_embedding":false,"enable_multimodal":null,"revision":null,"model_impl":"auto","model_config_parser":"auto","json_model_override_args":"{}","dtype":"auto","quantization":null,"quantization_param_path":null,"kv_cache_dtype":"auto","enable_fp32_lm_head":false,"modelopt_quant":null,"modelopt_checkpoint_restore_path":null,"modelopt_checkpoint_save_path":null,"modelopt_export_path":null,"quantize_and_serve":false,"rl_quant_profile":null,"enable_tf32_matmul":false,"mem_fraction_static":0.88,"max_running_requests":null,"max_queued_requests":null,"max_total_tokens":null,"chunked_prefill_size":8192,"enable_dynamic_chunking":false,"max_prefill_tokens":16384,"prefill_max_requests":null,"schedule_policy":"fcfs","enable_priority_scheduling":false,"disable_priority_preemption":false,"default_priority_value":null,"abort_on_priority_when_disabled":false,"schedule_low_priority_values_first":false,"priority_scheduling_preemption_threshold":10,"retraction_policy":"length","schedule_conservativeness":1.0,"page_size":64,"swa_full_tokens_ratio":0.8,"disable_hybrid_swa_memory":false,"radix_eviction_policy":"lru","prefill_only_disable_kv_cache":false,"disable_radix_cache":true,"enable_page_major_kv_layout":false,"enable_unified_memory":false,"disable_chunked_prefix_cache":false,"disable_overlap_schedule":false,"num_continuous_decode_steps":1,"scheduler_recv_interval":1,"enable_mixed_chunk":false,"nccl_port":null,"dist_timeout":3600,"dist_init_addr":"174.1.60.5:20000","nnodes":4,"node_rank":0,"tp_size":32,"dcp_size":1,"pp_size":1,"pp_max_micro_batch_size":null,"pp_async_batch_depth":0,"dp_size":1,"load_balance_method":"round_robin","attn_cp_size":1,"moe_dp_size":1,"dcp_comm_backend":"ag_rs","dcp_replicate_q_proj":null,"dwdp_size":1,"enable_prefill_cp":false,"cp_strategy":null,"enable_dsa_cache_layer_split":false,"enable_dsa_prefill_context_parallel":false,"dsa_prefill_cp_mode":"round-robin-split","enable_prefill_context_parallel":false,"prefill_cp_mode":"in-seq-split","enable_cp_decode_attn_tp":false,"enable_dp_attention":false,"enable_dp_attention_local_control_broadcast":false,"enable_dp_lm_head":false,"enable_attn_tp_input_scattered":false,"disable_attn_tp_gather":false,"enable_p2p_check":false,"device":"cuda","base_gpu_id":0,"gpu_id_step":1,"random_seed":369235288,"watchdog_timeout":300,"soft_watchdog_timeout":null,"sleep_on_idle":false,"use_ray":false,"custom_sigquit_handler":null,"numa_node":null,"gc_threshold":null,"host":"0.0.0.0","port":30000,"fastapi_root_path":"","smg_grpc_mode":false,"grpc_mode":false,"grpc_port":null,"sidecar":null,"sidecar_args":null,"skip_server_warmup":false,"warmups":null,"enable_http2":false,"ssl_keyfile":null,"ssl_certfile":null,"ssl_ca_certs":null,"ssl_keyfile_password":null,"enable_ssl_refresh":false,"api_key":null,"admin_api_key":null,"served_model_name":"kimi-k3","weight_version":"default","chat_template":null,"hf_chat_template_name":null,"completion_template":null,"file_storage_path":"sglang_storage","enable_cache_report":false,"reasoning_parser":null,"default_chat_template_kwargs":null,"strip_thinking_cache":false,"enable_strict_thinking":false,"tool_call_parser":null,"tool_server":null,"sampling_defaults":"model","asr_max_buffer_seconds":60,"asr_max_concurrent_sessions":32,"preferred_sampling_params":null,"allow_auto_truncate":false,"stream_interval":1,"batch_notify_size":16,"stream_response_default_include_usage":false,"incremental_streaming_output":false,"enable_streaming_session":false,"enable_session_radix_cache":false,"log_level":"info","log_level_http":null,"log_requests":false,"log_requests_level":2,"log_requests_format":"text","log_requests_target":null,"uvicorn_access_log_exclude_prefixes":[],"crash_dump_folder":null,"show_time_cost":false,"enable_metrics":false,"smg_http_sidecar_port":null,"enable_mfu_metrics":false,"enable_metrics_for_all_schedulers":false,"load_snapshot_publish_interval":15,"tokenizer_metrics_custom_labels_header":"x-custom-labels","tokenizer_metrics_allowed_custom_labels":null,"extra_metric_labels":null,"bucket_time_to_first_token":null,"bucket_inter_token_latency":null,"bucket_e2e_request_latency":null,"prompt_tokens_buckets":null,"generation_tokens_buckets":null,"gc_warning_threshold_secs":0.0,"decode_log_interval":40,"enable_request_time_stats_logging":false,"kv_events_config":null,"enable_forward_pass_metrics":false,"forward_pass_metrics_worker_id":"","forward_pass_metrics_ipc_name":null,"enable_trace":false,"trace_modules":"request","otlp_traces_endpoint":"localhost:4317","export_metrics_to_file":false,"export_metrics_to_file_dir":null,"stat_loggers":null,"constrained_json_whitespace_pattern":null,"constrained_json_disable_any_whitespace":false,"attention_backend":"triton","decode_attention_backend":null,"prefill_attention_backend":null,"sampling_backend":"flashinfer","grammar_backend":"xgrammar","radix_cache_backend":null,"mm_attention_backend":null,"fp8_gemm_runner_backend":"auto","fp4_gemm_runner_backend":"auto","bf16_gemm_backend":"auto","dsa_prefill_backend":null,"dsa_decode_backend":null,"dsa_paged_mqa_logits_backend":"auto","dsa_topk_backend":"sgl-kernel","disable_flashinfer_autotune":false,"flashinfer_autotune_skip_ops":null,"mamba_backend":"triton","cuda_graph_config":{"decode":{"backend":"full","max_bs":16,"bs":[1,2,4,8,12,16],"tc_compiler":"eager","full_prefill_max_req":null},"prefill":{"backend":"disabled","max_bs":2048,"bs":[4,8,12,16,20,24,28,32,48,64,80,96,112,128,144,160,176,192,208,224,240,256,288,320,352,384,416,448,480,512,576,640,704,768,832,896,960,1024,1280,1536,1792,2048],"tc_compiler":"eager","full_prefill_max_req":null}},"cuda_graph_backend_decode":null,"cuda_graph_backend_prefill":null,"cuda_graph_max_bs_decode":16,"cuda_graph_max_bs_prefill":null,"cuda_graph_bs_decode":null,"cuda_graph_bs_prefill":null,"cuda_graph_tc_compiler":null,"disable_prefill_cuda_graph":false,"disable_decode_cuda_graph":false,"disable_cuda_graph":false,"disable_cuda_graph_padding":false,"enable_profile_cuda_graph":false,"enable_cudagraph_gc":false,"debug_cuda_graph":false,"enable_layerwise_nvtx_marker":false,"enable_nccl_nvls":false,"enable_symm_mem":false,"triton_attention_reduce_in_fp32":false,"triton_attention_num_kv_splits":8,"triton_attention_split_tile_size":null,"flashinfer_mla_disable_ragged":false,"enable_fused_qk_norm_rope":false,"enable_precise_embedding_interpolation":false,"enable_fused_moe_sum_all_reduce":false,"enable_deepseek_v4_fp4_indexer":false,"disable_custom_all_reduce":false,"enable_mscclpp":false,"enable_torch_symm_mem":false,"enable_scattered_sconv":false,"pre_warm_nccl":false,"enable_quant_communications":false,"enable_flashinfer_allreduce_fusion":false,"enforce_disable_flashinfer_allreduce_fusion":false,"flashinfer_allreduce_fusion_backend":null,"enable_aiter_allreduce_fusion":false,"enable_torch_compile":false,"enable_torch_compile_debug_mode":false,"torch_compile_max_bs":32,"torchao_config":"","speculative_algorithm":null,"speculative_draft_model_path":null,"speculative_draft_model_revision":null,"speculative_draft_load_format":null,"speculative_num_steps":null,"speculative_eagle_topk":null,"speculative_num_draft_tokens":null,"speculative_dflash_block_size":null,"speculative_dspark_block_size":null,"speculative_dspark_sps_table_path":null,"speculative_dspark_confidence_sts_path":null,"speculative_dspark_align_verify_tokens_to_graph_tier":false,"speculative_accept_threshold_single":1.0,"speculative_accept_threshold_acc":1.0,"speculative_use_rejection_sampling":false,"speculative_token_map":null,"speculative_attention_mode":"prefill","speculative_draft_attention_backend":null,"speculative_draft_window_size":null,"speculative_moe_runner_backend":"marlin","speculative_moe_a2a_backend":null,"speculative_draft_model_quantization":null,"speculative_skip_dp_mlp_sync":false,"enable_multi_layer_eagle":false,"speculative_adaptive":false,"speculative_adaptive_config":null,"decoupled_spec_bind_endpoint":null,"decoupled_spec_connect_endpoints":null,"decoupled_spec_rank":null,"decoupled_spec_role":"null","spec_trace_dir":null,"speculative_ngram_min_bfs_breadth":1,"speculative_ngram_max_bfs_breadth":10,"speculative_ngram_match_type":"BFS","speculative_ngram_max_trie_depth":18,"speculative_ngram_capacity":10000000,"speculative_ngram_external_corpus_path":null,"speculative_ngram_external_sam_budget":0,"speculative_ngram_external_corpus_max_tokens":10000000,"ep_size":32,"moe_a2a_backend":"none","moe_runner_backend":"marlin","flashinfer_mxfp4_moe_precision":"default","deepep_mode":"auto","fuseep_mode":2,"deepep_dispatcher_output_dtype":"auto","ep_num_redundant_experts":0,"ep_dispatch_algorithm":null,"init_expert_location":"trivial","enable_eplb":false,"eplb_algorithm":"auto","eplb_rebalance_num_iterations":1000,"eplb_rebalance_layers_per_chunk":null,"eplb_min_rebalancing_utilization_threshold":1.0,"expert_distribution_recorder_mode":null,"expert_distribution_recorder_buffer_size":1000,"enable_expert_distribution_metrics":false,"deepep_config":null,"moe_dense_tp_size":null,"elastic_ep_backend":null,"enable_elastic_expert_backup":false,"mooncake_ib_device":null,"enable_waterfill":false,"ep_join_mode":null,"ep_join_rank_offset":0,"elastic_ep_initial_size":null,"max_ep_size":null,"elastic_ep_scale_timeout":600,"elastic_ep_rejoin":false,"disable_flashinfer_cutlass_moe_fp4_allgather":false,"disable_shared_experts_fusion":false,"enforce_shared_experts_fusion":false,"max_mamba_cache_size":null,"mamba_ssm_dtype":null,"mamba_max_states_per_path":-1,"enable_mamba_cache_stochastic_rounding":false,"mamba_cache_philox_rounds":0,"mamba_full_memory_ratio":0.21,"mamba_radix_cache_strategy":"extra_buffer_lazy","uses_mamba_radix_cache":false,"mamba_track_interval":256,"enable_int8_mamba_checkpoint":false,"int8_mamba_ckpt_size":null,"linear_attn_backend":"triton","linear_attn_decode_backend":null,"linear_attn_prefill_backend":null,"linear_attn_verify_backend":null,"enable_linear_replayssm":false,"linear_replayssm_cache_len":16,"enable_linear_replayssm_spec":false,"enable_hierarchical_cache":false,"hicache_ratio":2.0,"hicache_size":0,"hicache_write_policy":"write_through","hicache_io_backend":"kernel","hicache_mem_layout":"page_first","hicache_storage_backend":null,"hicache_storage_prefetch_policy":"timeout","hicache_storage_backend_extra_config":null,"enable_hisparse":false,"hisparse_config":null,"enable_broadcast_mm_inputs_process":false,"enable_prefix_mm_cache":false,"mm_enable_dp_encoder":false,"mm_process_config":{},"mm_processor_worker_num":0,"mm_io_worker_num":0,"limit_mm_data_per_request":null,"enable_mm_global_cache":false,"disable_fast_image_processor":false,"mm_feature_transport":"cpu","keep_mm_feature_on_device":false,"enable_lora":null,"enable_lora_overlap_loading":null,"max_lora_rank":null,"lora_target_modules":null,"lora_paths":null,"max_loaded_loras":null,"max_loras_per_batch":8,"lora_eviction_policy":"lru","lora_backend":"csgmv","max_lora_chunk_size":16,"experts_shared_outer_loras":null,"lora_use_virtual_experts":false,"lora_strict_loading":false,"lora_drain_wait_threshold":0.0,"enable_two_batch_overlap":false,"enable_single_batch_overlap":false,"tbo_token_distribution_threshold":0.48,"cpu_offload_gb":0,"offload_group_size":-1,"offload_num_in_group":1,"offload_prefetch_step":1,"offload_mode":"cpu","enable_lmcache":false,"lmcache_config_file":null,"enable_flexkv":false,"flexkv_config_file":null,"kt_weight_path":null,"kt_method":"AMXINT4","kt_cpuinfer":null,"kt_threadpool_count":2,"kt_num_gpu_experts":null,"kt_max_deferred_experts_per_token":null,"dllm_algorithm":null,"dllm_algorithm_config":null,"dllm_fdfo":true,"disaggregation_mode":"decode","disaggregation_transfer_backend":"mooncake","disaggregation_bootstrap_port":28800,"disaggregation_ib_device":"mlx5_0,mlx5_1,mlx5_2,mlx5_3","disaggregation_decode_enable_radix_cache":false,"disaggregation_decode_enable_offload_kvcache":false,"num_reserved_decode_tokens":512,"disaggregation_decode_extra_slots":0,"disaggregation_decode_polling_interval":1,"optimistic_prefill_attempts":0,"encoder_only":false,"language_only":false,"encoder_transfer_backend":"zmq_to_tokenizer","encoder_urls":[],"encoder_bootstrap_port":8997,"encoder_register_urls":[],"enable_adaptive_dispatch_to_encoder":false,"enable_pdmux":false,"pdmux_config_path":null,"sm_group_num":8,"custom_weight_loader":[],"weight_loader_disable_mmap":false,"weight_loader_prefetch_checkpoints":false,"weight_loader_prefetch_num_threads":4,"weight_loader_drop_cache_after_load":false,"remote_instance_weight_loader_seed_instance_ip":null,"remote_instance_weight_loader_seed_instance_service_port":null,"remote_instance_weight_loader_send_weights_group_ports":null,"remote_instance_weight_loader_backend":"nccl","remote_instance_weight_loader_start_seed_via_transfer_engine":false,"engine_info_bootstrap_port":6789,"modelexpress_config":null,"download_dir":null,"model_checksum":null,"delete_ckpt_after_loading":false,"decrypted_config_file":null,"decrypted_draft_config_file":null,"checkpoint_engine_wait_weights_before_ready":false,"enable_prefill_delayer":false,"prefill_delayer_max_delay_passes":30,"prefill_delayer_token_usage_low_watermark":null,"prefill_delayer_forward_passes_buckets":null,"prefill_delayer_wait_seconds_buckets":null,"prefill_delayer_queue_min_ratio":null,"prefill_delayer_max_delay_ms":null,"min_free_slots_delay":null,"enable_deterministic_inference":false,"rl_on_policy_target":null,"kv_canary":"none","kv_canary_real_data":"none","kv_canary_sweep_interval":0,"enable_dynamic_batch_tokenizer":false,"dynamic_batch_tokenizer_batch_size":32,"dynamic_batch_tokenizer_batch_timeout":0.002,"enable_tokenizer_batch_encode":false,"disable_tokenizer_batch_decode":false,"debug_tensor_dump_output_folder":null,"debug_tensor_dump_layers":null,"debug_tensor_dump_input_file":null,"enable_memory_saver":false,"enable_weights_cpu_backup":false,"enable_draft_weights_cpu_backup":false,"enable_custom_logit_processor":false,"enable_return_hidden_states":false,"enable_return_routed_experts":false,"enable_return_indexer_topk":false,"disable_outlines_disk_cache":false,"enable_mis":false,"weight_cache_mode":"off","weight_cache_socket":null,"weight_cache_timeout":1800,"forward_hooks":null,"msprobe_dump_config":null,"status":"ready","max_total_num_tokens":440256,"max_req_input_len":440250,"internal_states":[{"model_path":"/data/hf_models/Kimi-K3","tokenizer_path":"/data/hf_models/Kimi-K3","tokenizer_mode":"auto","tokenizer_backend":"huggingface","tokenizer_worker_num":1,"detokenizer_worker_num":1,"skip_tokenizer_init":false,"load_format":"auto","model_loader_extra_config":"{}","trust_remote_code":true,"context_length":null,"is_embedding":false,"enable_multimodal":null,"revision":null,"model_impl":"auto","model_config_parser":"auto","json_model_override_args":"{}","dtype":"auto","quantization":null,"quantization_param_path":null,"kv_cache_dtype":"auto","enable_fp32_lm_head":false,"modelopt_quant":null,"modelopt_checkpoint_restore_path":null,"modelopt_checkpoint_save_path":null,"modelopt_export_path":null,"quantize_and_serve":false,"rl_quant_profile":null,"enable_tf32_matmul":false,"mem_fraction_static":0.88,"max_running_requests":null,"max_queued_requests":null,"max_total_tokens":null,"chunked_prefill_size":8192,"enable_dynamic_chunking":false,"max_prefill_tokens":16384,"prefill_max_requests":null,"schedule_policy":"fcfs","enable_priority_scheduling":false,"disable_priority_preemption":false,"default_priority_value":null,"abort_on_priority_when_disabled":false,"schedule_low_priority_values_first":false,"priority_scheduling_preemption_threshold":10,"retraction_policy":"length","schedule_conservativeness":1.0,"page_size":64,"swa_full_tokens_ratio":0.8,"disable_hybrid_swa_memory":false,"radix_eviction_policy":"lru","prefill_only_disable_kv_cache":false,"disable_radix_cache":true,"enable_page_major_kv_layout":false,"enable_unified_memory":false,"disable_chunked_prefix_cache":true,"disable_overlap_schedule":false,"num_continuous_decode_steps":1,"scheduler_recv_interval":1,"enable_mixed_chunk":false,"nccl_port":null,"dist_timeout":3600,"dist_init_addr":"174.1.60.5:20000","nnodes":4,"node_rank":0,"tp_size":32,"dcp_size":1,"pp_size":1,"pp_max_micro_batch_size":181,"pp_async_batch_depth":0,"dp_size":1,"load_balance_method":"round_robin","attn_cp_size":1,"moe_dp_size":1,"dcp_comm_backend":"ag_rs","dcp_replicate_q_proj":null,"dwdp_size":1,"enable_prefill_cp":false,"cp_strategy":null,"enable_dsa_cache_layer_split":false,"enable_dsa_prefill_context_parallel":false,"dsa_prefill_cp_mode":"round-robin-split","enable_prefill_context_parallel":false,"prefill_cp_mode":"in-seq-split","enable_cp_decode_attn_tp":false,"enable_dp_attention":false,"enable_dp_attention_local_control_broadcast":false,"enable_dp_lm_head":false,"enable_attn_tp_input_scattered":false,"disable_attn_tp_gather":false,"enable_p2p_check":false,"device":"cuda","base_gpu_id":0,"gpu_id_step":1,"random_seed":369235288,"watchdog_timeout":300,"soft_watchdog_timeout":null,"sleep_on_idle":false,"use_ray":false,"numa_node":null,"gc_threshold":null,"host":"0.0.0.0","port":30000,"fastapi_root_path":"","smg_grpc_mode":false,"grpc_mode":false,"grpc_port":null,"sidecar":null,"sidecar_args":null,"skip_server_warmup":false,"warmups":null,"enable_http2":false,"ssl_keyfile":null,"ssl_certfile":null,"ssl_ca_certs":null,"ssl_keyfile_password":null,"enable_ssl_refresh":false,"api_key":null,"admin_api_key":null,"served_model_name":"kimi-k3","weight_version":"default","chat_template":null,"hf_chat_template_name":null,"completion_template":null,"file_storage_path":"sglang_storage","enable_cache_report":false,"reasoning_parser":null,"default_chat_template_kwargs":null,"strip_thinking_cache":false,"enable_strict_thinking":false,"tool_call_parser":null,"tool_server":null,"sampling_defaults":"model","asr_max_buffer_seconds":60,"asr_max_concurrent_sessions":32,"preferred_sampling_params":null,"allow_auto_truncate":false,"stream_interval":1,"batch_notify_size":16,"stream_response_default_include_usage":false,"incremental_streaming_output":false,"enable_streaming_session":false,"enable_session_radix_cache":false,"log_level":"info","log_level_http":null,"log_requests":false,"log_requests_level":2,"log_requests_format":"text","log_requests_target":null,"uvicorn_access_log_exclude_prefixes":[],"crash_dump_folder":null,"show_time_cost":false,"enable_metrics":false,"smg_http_sidecar_port":null,"enable_mfu_metrics":false,"enable_metrics_for_all_schedulers":false,"load_snapshot_publish_interval":15,"tokenizer_metrics_custom_labels_header":"x-custom-labels","tokenizer_metrics_allowed_custom_labels":null,"extra_metric_labels":null,"bucket_time_to_first_token":null,"bucket_inter_token_latency":null,"bucket_e2e_request_latency":null,"prompt_tokens_buckets":null,"generation_tokens_buckets":null,"gc_warning_threshold_secs":0.0,"decode_log_interval":40,"enable_request_time_stats_logging":false,"kv_events_config":null,"enable_forward_pass_metrics":false,"forward_pass_metrics_worker_id":"","forward_pass_metrics_ipc_name":null,"enable_trace":false,"trace_modules":"request","otlp_traces_endpoint":"localhost:4317","export_metrics_to_file":false,"export_metrics_to_file_dir":null,"stat_loggers":null,"constrained_json_whitespace_pattern":null,"constrained_json_disable_any_whitespace":false,"attention_backend":"triton","decode_attention_backend":null,"prefill_attention_backend":null,"sampling_backend":"flashinfer","grammar_backend":"xgrammar","radix_cache_backend":null,"mm_attention_backend":null,"fp8_gemm_runner_backend":"auto","fp4_gemm_runner_backend":"auto","bf16_gemm_backend":"auto","dsa_prefill_backend":null,"dsa_decode_backend":null,"dsa_paged_mqa_logits_backend":"auto","dsa_topk_backend":"sgl-kernel","disable_flashinfer_autotune":false,"flashinfer_autotune_skip_ops":null,"mamba_backend":"triton","cuda_graph_config":{"decode":{"backend":"full","max_bs":16,"bs":[1,2,4,8,12,16],"tc_compiler":"eager","full_prefill_max_req":null},"prefill":{"backend":"disabled","max_bs":2048,"bs":[4,8,12,16,20,24,28,32,48,64,80,96,112,128,144,160,176,192,208,224,240,256,288,320,352,384,416,448,480,512,576,640,704,768,832,896,960,1024,1280,1536,1792,2048],"tc_compiler":"eager","full_prefill_max_req":null}},"cuda_graph_backend_decode":null,"cuda_graph_backend_prefill":null,"cuda_graph_max_bs_decode":16,"cuda_graph_max_bs_prefill":null,"cuda_graph_bs_decode":null,"cuda_graph_bs_prefill":null,"cuda_graph_tc_compiler":null,"disable_prefill_cuda_graph":false,"disable_decode_cuda_graph":false,"disable_cuda_graph":false,"disable_cuda_graph_padding":false,"enable_profile_cuda_graph":false,"enable_cudagraph_gc":false,"debug_cuda_graph":false,"enable_layerwise_nvtx_marker":false,"enable_nccl_nvls":false,"enable_symm_mem":false,"triton_attention_reduce_in_fp32":false,"triton_attention_num_kv_splits":8,"triton_attention_split_tile_size":null,"flashinfer_mla_disable_ragged":false,"enable_fused_qk_norm_rope":false,"enable_precise_embedding_interpolation":false,"enable_fused_moe_sum_all_reduce":false,"enable_deepseek_v4_fp4_indexer":false,"disable_custom_all_reduce":false,"enable_mscclpp":false,"enable_torch_symm_mem":false,"enable_scattered_sconv":false,"pre_warm_nccl":false,"enable_quant_communications":false,"enable_flashinfer_allreduce_fusion":false,"enforce_disable_flashinfer_allreduce_fusion":false,"flashinfer_allreduce_fusion_backend":null,"enable_aiter_allreduce_fusion":false,"enable_torch_compile":false,"enable_torch_compile_debug_mode":false,"torch_compile_max_bs":32,"torchao_config":"","speculative_algorithm":null,"speculative_draft_model_path":null,"speculative_draft_model_revision":null,"speculative_draft_load_format":null,"speculative_num_steps":null,"speculative_eagle_topk":null,"speculative_num_draft_tokens":null,"speculative_dflash_block_size":null,"speculative_dspark_block_size":null,"speculative_dspark_sps_table_path":null,"speculative_dspark_confidence_sts_path":null,"speculative_dspark_align_verify_tokens_to_graph_tier":false,"speculative_accept_threshold_single":1.0,"speculative_accept_threshold_acc":1.0,"speculative_use_rejection_sampling":false,"speculative_token_map":null,"speculative_attention_mode":"prefill","speculative_draft_attention_backend":null,"speculative_draft_window_size":null,"speculative_moe_runner_backend":"marlin","speculative_moe_a2a_backend":null,"speculative_draft_model_quantization":null,"speculative_skip_dp_mlp_sync":false,"enable_multi_layer_eagle":false,"speculative_adaptive":false,"speculative_adaptive_config":null,"decoupled_spec_bind_endpoint":null,"decoupled_spec_connect_endpoints":null,"decoupled_spec_rank":null,"decoupled_spec_role":"null","spec_trace_dir":null,"speculative_ngram_min_bfs_breadth":1,"speculative_ngram_max_bfs_breadth":10,"speculative_ngram_match_type":"BFS","speculative_ngram_max_trie_depth":18,"speculative_ngram_capacity":10000000,"speculative_ngram_external_corpus_path":null,"speculative_ngram_external_sam_budget":0,"speculative_ngram_external_corpus_max_tokens":10000000,"ep_size":32,"moe_a2a_backend":"none","moe_runner_backend":"marlin","flashinfer_mxfp4_moe_precision":"default","deepep_mode":"auto","fuseep_mode":2,"deepep_dispatcher_output_dtype":"auto","ep_num_redundant_experts":0,"ep_dispatch_algorithm":null,"init_expert_location":"trivial","enable_eplb":false,"eplb_algorithm":"auto","eplb_rebalance_num_iterations":1000,"eplb_rebalance_layers_per_chunk":null,"eplb_min_rebalancing_utilization_threshold":1.0,"expert_distribution_recorder_mode":null,"expert_distribution_recorder_buffer_size":1000,"enable_expert_distribution_metrics":false,"deepep_config":null,"moe_dense_tp_size":null,"elastic_ep_backend":null,"enable_elastic_expert_backup":false,"mooncake_ib_device":null,"enable_waterfill":false,"ep_join_mode":null,"ep_join_rank_offset":0,"elastic_ep_initial_size":null,"max_ep_size":null,"elastic_ep_scale_timeout":600,"elastic_ep_rejoin":false,"disable_flashinfer_cutlass_moe_fp4_allgather":false,"disable_shared_experts_fusion":false,"enforce_shared_experts_fusion":false,"max_mamba_cache_size":181,"mamba_ssm_dtype":null,"mamba_max_states_per_path":-1,"enable_mamba_cache_stochastic_rounding":false,"mamba_cache_philox_rounds":0,"mamba_full_memory_ratio":0.21,"mamba_radix_cache_strategy":"extra_buffer_lazy","uses_mamba_radix_cache":false,"mamba_track_interval":256,"enable_int8_mamba_checkpoint":false,"int8_mamba_ckpt_size":null,"linear_attn_backend":"triton","linear_attn_decode_backend":null,"linear_attn_prefill_backend":null,"linear_attn_verify_backend":null,"enable_linear_replayssm":false,"linear_replayssm_cache_len":16,"enable_linear_replayssm_spec":false,"enable_hierarchical_cache":false,"hicache_ratio":2.0,"hicache_size":0,"hicache_write_policy":"write_through","hicache_io_backend":"kernel","hicache_mem_layout":"page_first","hicache_storage_backend":null,"hicache_storage_prefetch_policy":"timeout","hicache_storage_backend_extra_config":null,"enable_hisparse":false,"hisparse_config":null,"enable_broadcast_mm_inputs_process":false,"enable_prefix_mm_cache":false,"mm_enable_dp_encoder":false,"mm_process_config":{},"mm_processor_worker_num":0,"mm_io_worker_num":0,"limit_mm_data_per_request":null,"enable_mm_global_cache":false,"disable_fast_image_processor":false,"mm_feature_transport":"cpu","keep_mm_feature_on_device":false,"enable_lora":null,"enable_lora_overlap_loading":null,"max_lora_rank":null,"lora_target_modules":null,"lora_paths":null,"max_loaded_loras":null,"max_loras_per_batch":8,"lora_eviction_policy":"lru","lora_backend":"csgmv","max_lora_chunk_size":16,"experts_shared_outer_loras":null,"lora_use_virtual_experts":false,"lora_strict_loading":false,"lora_drain_wait_threshold":0.0,"enable_two_batch_overlap":false,"enable_single_batch_overlap":false,"tbo_token_distribution_threshold":0.48,"cpu_offload_gb":0,"offload_group_size":-1,"offload_num_in_group":1,"offload_prefetch_step":1,"offload_mode":"cpu","enable_lmcache":false,"lmcache_config_file":null,"enable_flexkv":false,"flexkv_config_file":null,"kt_weight_path":null,"kt_method":"AMXINT4","kt_cpuinfer":null,"kt_threadpool_count":2,"kt_num_gpu_experts":null,"kt_max_deferred_experts_per_token":null,"dllm_algorithm":null,"dllm_algorithm_config":null,"dllm_fdfo":true,"disaggregation_mode":"decode","disaggregation_transfer_backend":"mooncake","disaggregation_bootstrap_port":28800,"disaggregation_ib_device":"mlx5_0,mlx5_1,mlx5_2,mlx5_3","disaggregation_decode_enable_radix_cache":false,"disaggregation_decode_enable_offload_kvcache":false,"num_reserved_decode_tokens":512,"disaggregation_decode_extra_slots":0,"disaggregation_decode_polling_interval":1,"optimistic_prefill_attempts":0,"encoder_only":false,"language_only":false,"encoder_transfer_backend":"zmq_to_tokenizer","encoder_urls":[],"encoder_bootstrap_port":8997,"encoder_register_urls":[],"enable_adaptive_dispatch_to_encoder":false,"enable_pdmux":false,"pdmux_config_path":null,"sm_group_num":8,"custom_weight_loader":[],"weight_loader_disable_mmap":false,"weight_loader_prefetch_checkpoints":false,"weight_loader_prefetch_num_threads":4,"weight_loader_drop_cache_after_load":false,"remote_instance_weight_loader_seed_instance_ip":null,"remote_instance_weight_loader_seed_instance_service_port":null,"remote_instance_weight_loader_send_weights_group_ports":null,"remote_instance_weight_loader_backend":"nccl","remote_instance_weight_loader_start_seed_via_transfer_engine":false,"engine_info_bootstrap_port":6789,"modelexpress_config":null,"download_dir":null,"model_checksum":null,"delete_ckpt_after_loading":false,"decrypted_config_file":null,"decrypted_draft_config_file":null,"checkpoint_engine_wait_weights_before_ready":false,"enable_prefill_delayer":false,"prefill_delayer_max_delay_passes":30,"prefill_delayer_token_usage_low_watermark":null,"prefill_delayer_forward_passes_buckets":null,"prefill_delayer_wait_seconds_buckets":null,"prefill_delayer_queue_min_ratio":null,"prefill_delayer_max_delay_ms":null,"min_free_slots_delay":null,"enable_deterministic_inference":false,"rl_on_policy_target":null,"kv_canary":"none","kv_canary_real_data":"none","kv_canary_sweep_interval":0,"enable_dynamic_batch_tokenizer":false,"dynamic_batch_tokenizer_batch_size":32,"dynamic_batch_tokenizer_batch_timeout":0.002,"enable_tokenizer_batch_encode":false,"disable_tokenizer_batch_decode":false,"debug_tensor_dump_output_folder":null,"debug_tensor_dump_layers":null,"debug_tensor_dump_input_file":null,"enable_memory_saver":false,"enable_weights_cpu_backup":false,"enable_draft_weights_cpu_backup":false,"enable_custom_logit_processor":false,"enable_return_hidden_states":false,"enable_return_routed_experts":false,"enable_return_indexer_topk":false,"disable_outlines_disk_cache":false,"enable_mis":false,"weight_cache_mode":"off","weight_cache_socket":null,"weight_cache_timeout":1800,"forward_hooks":null,"msprobe_dump_config":null,"_resolved_overrides":[["_sampling_backend_default",{"sampling_backend":"flashinfer"}],["_attention_backend_default",{"attention_backend":"triton"}],["_data_parallelism_defaults",{"enable_dp_attention":false,"enable_dp_lm_head":false}],["_speculative_moe_runner_default",{"speculative_moe_runner_backend":"marlin"}]],"grpc_worker_threads":4,"_quantization_explicitly_unset":false,"_cuda_graph_config_locked":[["decode","max_bs"]],"_declarations_materialized":true,"_runtime_mutations":[["model_runner.chunked_prefix_cache_gate",{"disable_chunked_prefix_cache":true}],["mamba_pool.memory_budget",{"max_mamba_cache_size":181}]],"_in_override":false,"_mx_config_cache":{},"max_speculative_num_draft_tokens":null,"last_gen_throughput":48.2592441209455,"memory_usage":{"weight":58.57,"kvcache":11.34,"token_capacity":440256,"graph":0.6},"effective_max_running_requests_per_dp":181}],"version":"0.5.16","kv_events":null}]}
\ No newline at end of file
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-c8-20260827-1340/metadata/status_after_validation.txt b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-c8-20260827-1340/metadata/status_after_validation.txt
new file mode 100644
index 0000000..5e4883e
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-c8-20260827-1340/metadata/status_after_validation.txt
@@ -0,0 +1,85 @@
+===== node 1 =====
+kimi3_pro6000_pd_pp8_standard_prefill_node0|local/sglang:kimi-k3-sm120-flashinfer-mxfp4-pd1|Up About an hour
+kimi3_pro6000_pd_pp8_standard_mc_master|local/sglang:kimi-k3-sm120-flashinfer-mxfp4-pd1|Up About an hour
+0, 59711 MiB, 0 %
+1, 59711 MiB, 0 %
+2, 59711 MiB, 0 %
+3, 59711 MiB, 0 %
+4, 67385 MiB, 0 %
+5, 67389 MiB, 0 %
+6, 67383 MiB, 0 %
+7, 67395 MiB, 0 %
+===== node 2 =====
+kimi3_pro6000_pd_pp8_standard_prefill_node1|local/sglang:kimi-k3-sm120-flashinfer-mxfp4-pd1|Up About an hour
+0, 70433 MiB, 0 %
+1, 70439 MiB, 0 %
+2, 70439 MiB, 0 %
+3, 70439 MiB, 0 %
+4, 74813 MiB, 0 %
+5, 74813 MiB, 0 %
+6, 74813 MiB, 0 %
+7, 74813 MiB, 0 %
+===== node 3 =====
+kimi3_pro6000_pd_pp8_standard_prefill_node2|local/sglang:kimi-k3-sm120-flashinfer-mxfp4-pd1|Up About an hour
+0, 75679 MiB, 0 %
+1, 75675 MiB, 0 %
+2, 75675 MiB, 0 %
+3, 75675 MiB, 0 %
+4, 76689 MiB, 0 %
+5, 76689 MiB, 0 %
+6, 76685 MiB, 0 %
+7, 76689 MiB, 0 %
+===== node 4 =====
+kimi3_pro6000_pd_pp8_standard_prefill_node3|local/sglang:kimi-k3-sm120-flashinfer-mxfp4-pd1|Up About an hour
+0, 77185 MiB, 0 %
+1, 77185 MiB, 0 %
+2, 77185 MiB, 0 %
+3, 77189 MiB, 0 %
+4, 79665 MiB, 0 %
+5, 79665 MiB, 0 %
+6, 79665 MiB, 0 %
+7, 79665 MiB, 0 %
+===== node 5 =====
+kimi3_pro6000_pd_pp8_standard_router|local/sglang:kimi-k3-sm120-flashinfer-mxfp4-pd1|Up About an hour
+kimi3_pro6000_pd_pp8_standard_decode_node0|local/sglang:kimi-k3-sm120-flashinfer-mxfp4-pd1|Up About an hour
+0, 76277 MiB, 0 %
+1, 76261 MiB, 0 %
+2, 76245 MiB, 0 %
+3, 76293 MiB, 0 %
+4, 76317 MiB, 0 %
+5, 76261 MiB, 0 %
+6, 76245 MiB, 0 %
+7, 76333 MiB, 0 %
+===== node 6 =====
+kimi3_pro6000_pd_pp8_standard_decode_node1|local/sglang:kimi-k3-sm120-flashinfer-mxfp4-pd1|Up About an hour
+0, 76317 MiB, 0 %
+1, 76261 MiB, 0 %
+2, 76245 MiB, 0 %
+3, 76333 MiB, 0 %
+4, 76317 MiB, 0 %
+5, 76261 MiB, 0 %
+6, 76245 MiB, 0 %
+7, 76333 MiB, 0 %
+===== node 7 =====
+kimi3_pro6000_pd_pp8_standard_decode_node2|local/sglang:kimi-k3-sm120-flashinfer-mxfp4-pd1|Up About an hour
+0, 76317 MiB, 0 %
+1, 76261 MiB, 0 %
+2, 76245 MiB, 0 %
+3, 76333 MiB, 0 %
+4, 76317 MiB, 0 %
+5, 76261 MiB, 0 %
+6, 76245 MiB, 0 %
+7, 76333 MiB, 0 %
+===== node 8 =====
+kimi3_pro6000_pd_pp8_standard_decode_node3|local/sglang:kimi-k3-sm120-flashinfer-mxfp4-pd1|Up About an hour
+0, 76317 MiB, 0 %
+1, 76261 MiB, 0 %
+2, 76245 MiB, 0 %
+3, 76333 MiB, 0 %
+4, 76277 MiB, 0 %
+5, 76261 MiB, 0 %
+6, 76245 MiB, 0 %
+7, 76293 MiB, 0 %
+p=healthy
+d=healthy
+router=healthy
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-c8-20260827-1340/orchestrator.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-c8-20260827-1340/orchestrator.log
new file mode 100644
index 0000000..b1a5d60
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-c8-20260827-1340/orchestrator.log
@@ -0,0 +1 @@
+[2026-08-27 13:39:32] case=decode_16k_512_c8 isl=16384 osl=512 c=8 prompts=40
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-c8-20260827-1340/summary.csv b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-c8-20260827-1340/summary.csv
new file mode 100644
index 0000000..3ba36b2
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-decode512-c8-20260827-1340/summary.csv
@@ -0,0 +1,2 @@
+case,completed,input_throughput,request_throughput,median_ttft_ms,p95_ttft_ms,median_tpot_ms,p95_tpot_ms,median_e2e_latency_ms,p95_e2e_latency_ms
+decode_16k_512_c8,40,2544.2701870772244,0.15528992841047512,18754.675634438172,23181.66090096347,63.195876082325256,66.5476718326072,51368.52010944858,56342.666492424905
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/commands/d_node0.cmd.txt b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/commands/d_node0.cmd.txt
new file mode 100644
index 0000000..1c83e68
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/commands/d_node0.cmd.txt
@@ -0,0 +1 @@
+docker run -d --name kimi3_pro6000_pd_pp8_standard_decode_node0 --gpus all --network host --ipc=host --ulimit memlock=-1 --device /dev/infiniband --shm-size 32g --entrypoint bash -v /data/hf_models/Kimi-K3:/data/hf_models/Kimi-K3:ro -e CUDA_VISIBLE_DEVICES=0\,1\,2\,3\,4\,5\,6\,7 -e SGLANG_HOST_IP=174.1.60.5 -e GLOO_SOCKET_IFNAME=bond0 -e NCCL_SOCKET_IFNAME=bond1 -e NCCL_IB_HCA=mlx5_0\,mlx5_1\,mlx5_2\,mlx5_3 -e NCCL_IB_GID_INDEX=3 -e NCCL_IB_TIMEOUT=22 -e NCCL_IB_RETRY_CNT=7 -e NCCL_CUMEM_ENABLE=1 -e MOONCAKE_MASTER=174.1.60.1:50051 -e MOONCAKE_PROTOCOL=rdma -e SGLANG_ENABLE_TP_MEMORY_INBALANCE_CHECK=0 -e SGLANG_MOE_FUSED_GATE_RADIX=1 -e FLASHINFER_DISABLE_JIT=1 -e FLASHINFER_DISABLE_VERSION_CHECK=1 local/sglang:kimi-k3-sm120-flashinfer-mxfp4-pd1 -lc exec\ python3\ -m\ sglang.launch_server\ --model-path\ /data/hf_models/Kimi-K3\ --served-model-name\ kimi-k3\ --tp-size\ 32\ --pp-size\ 1\ --ep-size\ 32\ --nnodes\ 4\ --node-rank\ 0\ --dist-init-addr\ 174.1.60.5:20000\ --trust-remote-code\ --moe-runner-backend\ marlin\ --mem-fraction-static\ 0.88\ --cuda-graph-max-bs-decode\ 16\ --mamba-radix-cache-strategy\ extra_buffer_lazy\ --disable-radix-cache\ --dist-timeout\ 3600\ --mamba-full-memory-ratio\ 0.21\ --page-size\ 64\ --disaggregation-transfer-backend\ mooncake\ --disaggregation-bootstrap-port\ 28800\ --disaggregation-ib-device\ mlx5_0\\\,mlx5_1\\\,mlx5_2\\\,mlx5_3\ --disaggregation-mode\ decode\ --host\ 0.0.0.0\ --port\ 30000\  
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/commands/d_node1.cmd.txt b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/commands/d_node1.cmd.txt
new file mode 100644
index 0000000..953c919
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/commands/d_node1.cmd.txt
@@ -0,0 +1 @@
+docker run -d --name kimi3_pro6000_pd_pp8_standard_decode_node1 --gpus all --network host --ipc=host --ulimit memlock=-1 --device /dev/infiniband --shm-size 32g --entrypoint bash -v /data/hf_models/Kimi-K3:/data/hf_models/Kimi-K3:ro -e CUDA_VISIBLE_DEVICES=0\,1\,2\,3\,4\,5\,6\,7 -e SGLANG_HOST_IP=174.1.60.6 -e GLOO_SOCKET_IFNAME=bond0 -e NCCL_SOCKET_IFNAME=bond1 -e NCCL_IB_HCA=mlx5_0\,mlx5_1\,mlx5_2\,mlx5_3 -e NCCL_IB_GID_INDEX=3 -e NCCL_IB_TIMEOUT=22 -e NCCL_IB_RETRY_CNT=7 -e NCCL_CUMEM_ENABLE=1 -e MOONCAKE_MASTER=174.1.60.1:50051 -e MOONCAKE_PROTOCOL=rdma -e SGLANG_ENABLE_TP_MEMORY_INBALANCE_CHECK=0 -e SGLANG_MOE_FUSED_GATE_RADIX=1 -e FLASHINFER_DISABLE_JIT=1 -e FLASHINFER_DISABLE_VERSION_CHECK=1 local/sglang:kimi-k3-sm120-flashinfer-mxfp4-pd1 -lc exec\ python3\ -m\ sglang.launch_server\ --model-path\ /data/hf_models/Kimi-K3\ --served-model-name\ kimi-k3\ --tp-size\ 32\ --pp-size\ 1\ --ep-size\ 32\ --nnodes\ 4\ --node-rank\ 1\ --dist-init-addr\ 174.1.60.5:20000\ --trust-remote-code\ --moe-runner-backend\ marlin\ --mem-fraction-static\ 0.88\ --cuda-graph-max-bs-decode\ 16\ --mamba-radix-cache-strategy\ extra_buffer_lazy\ --disable-radix-cache\ --dist-timeout\ 3600\ --mamba-full-memory-ratio\ 0.21\ --page-size\ 64\ --disaggregation-transfer-backend\ mooncake\ --disaggregation-bootstrap-port\ 28800\ --disaggregation-ib-device\ mlx5_0\\\,mlx5_1\\\,mlx5_2\\\,mlx5_3\ --disaggregation-mode\ decode\ --host\ 0.0.0.0\ --port\ 30000\  
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/commands/d_node2.cmd.txt b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/commands/d_node2.cmd.txt
new file mode 100644
index 0000000..eebd226
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/commands/d_node2.cmd.txt
@@ -0,0 +1 @@
+docker run -d --name kimi3_pro6000_pd_pp8_standard_decode_node2 --gpus all --network host --ipc=host --ulimit memlock=-1 --device /dev/infiniband --shm-size 32g --entrypoint bash -v /data/hf_models/Kimi-K3:/data/hf_models/Kimi-K3:ro -e CUDA_VISIBLE_DEVICES=0\,1\,2\,3\,4\,5\,6\,7 -e SGLANG_HOST_IP=174.1.60.7 -e GLOO_SOCKET_IFNAME=bond0 -e NCCL_SOCKET_IFNAME=bond1 -e NCCL_IB_HCA=mlx5_0\,mlx5_1\,mlx5_2\,mlx5_3 -e NCCL_IB_GID_INDEX=3 -e NCCL_IB_TIMEOUT=22 -e NCCL_IB_RETRY_CNT=7 -e NCCL_CUMEM_ENABLE=1 -e MOONCAKE_MASTER=174.1.60.1:50051 -e MOONCAKE_PROTOCOL=rdma -e SGLANG_ENABLE_TP_MEMORY_INBALANCE_CHECK=0 -e SGLANG_MOE_FUSED_GATE_RADIX=1 -e FLASHINFER_DISABLE_JIT=1 -e FLASHINFER_DISABLE_VERSION_CHECK=1 local/sglang:kimi-k3-sm120-flashinfer-mxfp4-pd1 -lc exec\ python3\ -m\ sglang.launch_server\ --model-path\ /data/hf_models/Kimi-K3\ --served-model-name\ kimi-k3\ --tp-size\ 32\ --pp-size\ 1\ --ep-size\ 32\ --nnodes\ 4\ --node-rank\ 2\ --dist-init-addr\ 174.1.60.5:20000\ --trust-remote-code\ --moe-runner-backend\ marlin\ --mem-fraction-static\ 0.88\ --cuda-graph-max-bs-decode\ 16\ --mamba-radix-cache-strategy\ extra_buffer_lazy\ --disable-radix-cache\ --dist-timeout\ 3600\ --mamba-full-memory-ratio\ 0.21\ --page-size\ 64\ --disaggregation-transfer-backend\ mooncake\ --disaggregation-bootstrap-port\ 28800\ --disaggregation-ib-device\ mlx5_0\\\,mlx5_1\\\,mlx5_2\\\,mlx5_3\ --disaggregation-mode\ decode\ --host\ 0.0.0.0\ --port\ 30000\  
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/commands/d_node3.cmd.txt b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/commands/d_node3.cmd.txt
new file mode 100644
index 0000000..7006c2c
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/commands/d_node3.cmd.txt
@@ -0,0 +1 @@
+docker run -d --name kimi3_pro6000_pd_pp8_standard_decode_node3 --gpus all --network host --ipc=host --ulimit memlock=-1 --device /dev/infiniband --shm-size 32g --entrypoint bash -v /data/hf_models/Kimi-K3:/data/hf_models/Kimi-K3:ro -e CUDA_VISIBLE_DEVICES=0\,1\,2\,3\,4\,5\,6\,7 -e SGLANG_HOST_IP=174.1.60.8 -e GLOO_SOCKET_IFNAME=bond0 -e NCCL_SOCKET_IFNAME=bond1 -e NCCL_IB_HCA=mlx5_0\,mlx5_1\,mlx5_2\,mlx5_3 -e NCCL_IB_GID_INDEX=3 -e NCCL_IB_TIMEOUT=22 -e NCCL_IB_RETRY_CNT=7 -e NCCL_CUMEM_ENABLE=1 -e MOONCAKE_MASTER=174.1.60.1:50051 -e MOONCAKE_PROTOCOL=rdma -e SGLANG_ENABLE_TP_MEMORY_INBALANCE_CHECK=0 -e SGLANG_MOE_FUSED_GATE_RADIX=1 -e FLASHINFER_DISABLE_JIT=1 -e FLASHINFER_DISABLE_VERSION_CHECK=1 local/sglang:kimi-k3-sm120-flashinfer-mxfp4-pd1 -lc exec\ python3\ -m\ sglang.launch_server\ --model-path\ /data/hf_models/Kimi-K3\ --served-model-name\ kimi-k3\ --tp-size\ 32\ --pp-size\ 1\ --ep-size\ 32\ --nnodes\ 4\ --node-rank\ 3\ --dist-init-addr\ 174.1.60.5:20000\ --trust-remote-code\ --moe-runner-backend\ marlin\ --mem-fraction-static\ 0.88\ --cuda-graph-max-bs-decode\ 16\ --mamba-radix-cache-strategy\ extra_buffer_lazy\ --disable-radix-cache\ --dist-timeout\ 3600\ --mamba-full-memory-ratio\ 0.21\ --page-size\ 64\ --disaggregation-transfer-backend\ mooncake\ --disaggregation-bootstrap-port\ 28800\ --disaggregation-ib-device\ mlx5_0\\\,mlx5_1\\\,mlx5_2\\\,mlx5_3\ --disaggregation-mode\ decode\ --host\ 0.0.0.0\ --port\ 30000\  
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/commands/p_node0.cmd.txt b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/commands/p_node0.cmd.txt
new file mode 100644
index 0000000..a30b4dc
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/commands/p_node0.cmd.txt
@@ -0,0 +1 @@
+docker run -d --name kimi3_pro6000_pd_pp8_standard_prefill_node0 --gpus all --network host --ipc=host --ulimit memlock=-1 --device /dev/infiniband --shm-size 32g --entrypoint bash -v /data/hf_models/Kimi-K3:/data/hf_models/Kimi-K3:ro -e CUDA_VISIBLE_DEVICES=0\,1\,2\,3\,4\,5\,6\,7 -e SGLANG_HOST_IP=174.1.60.1 -e GLOO_SOCKET_IFNAME=bond0 -e NCCL_SOCKET_IFNAME=bond1 -e NCCL_IB_HCA=mlx5_0\,mlx5_1\,mlx5_2\,mlx5_3 -e NCCL_IB_GID_INDEX=3 -e NCCL_IB_TIMEOUT=22 -e NCCL_IB_RETRY_CNT=7 -e NCCL_CUMEM_ENABLE=1 -e MOONCAKE_MASTER=174.1.60.1:50051 -e MOONCAKE_PROTOCOL=rdma -e SGLANG_ENABLE_TP_MEMORY_INBALANCE_CHECK=0 -e SGLANG_MOE_FUSED_GATE_RADIX=1 -e FLASHINFER_DISABLE_JIT=1 -e FLASHINFER_DISABLE_VERSION_CHECK=1 local/sglang:kimi-k3-sm120-flashinfer-mxfp4-pd1 -lc exec\ python3\ -m\ sglang.launch_server\ --model-path\ /data/hf_models/Kimi-K3\ --served-model-name\ kimi-k3\ --tp-size\ 4\ --pp-size\ 8\ --ep-size\ 4\ --nnodes\ 4\ --node-rank\ 0\ --dist-init-addr\ 174.1.60.1:20000\ --trust-remote-code\ --moe-runner-backend\ flashinfer_mxfp4\ --chunked-prefill-size\ 8192\ --page-size\ 64\ --mem-fraction-static\ 0.88\ --cuda-graph-max-bs-decode\ 16\ --mamba-radix-cache-strategy\ extra_buffer_lazy\ --disable-radix-cache\ --dist-timeout\ 3600\ --mamba-full-memory-ratio\ 0.36\ --disaggregation-transfer-backend\ mooncake\ --disaggregation-bootstrap-port\ 28800\ --disaggregation-ib-device\ mlx5_0\\\,mlx5_1\\\,mlx5_2\\\,mlx5_3\ --disaggregation-mode\ prefill\ --host\ 0.0.0.0\ --port\ 30000\  
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/commands/p_node1.cmd.txt b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/commands/p_node1.cmd.txt
new file mode 100644
index 0000000..dfc04c7
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/commands/p_node1.cmd.txt
@@ -0,0 +1 @@
+docker run -d --name kimi3_pro6000_pd_pp8_standard_prefill_node1 --gpus all --network host --ipc=host --ulimit memlock=-1 --device /dev/infiniband --shm-size 32g --entrypoint bash -v /data/hf_models/Kimi-K3:/data/hf_models/Kimi-K3:ro -e CUDA_VISIBLE_DEVICES=0\,1\,2\,3\,4\,5\,6\,7 -e SGLANG_HOST_IP=174.1.60.2 -e GLOO_SOCKET_IFNAME=bond0 -e NCCL_SOCKET_IFNAME=bond1 -e NCCL_IB_HCA=mlx5_0\,mlx5_1\,mlx5_2\,mlx5_3 -e NCCL_IB_GID_INDEX=3 -e NCCL_IB_TIMEOUT=22 -e NCCL_IB_RETRY_CNT=7 -e NCCL_CUMEM_ENABLE=1 -e MOONCAKE_MASTER=174.1.60.1:50051 -e MOONCAKE_PROTOCOL=rdma -e SGLANG_ENABLE_TP_MEMORY_INBALANCE_CHECK=0 -e SGLANG_MOE_FUSED_GATE_RADIX=1 -e FLASHINFER_DISABLE_JIT=1 -e FLASHINFER_DISABLE_VERSION_CHECK=1 local/sglang:kimi-k3-sm120-flashinfer-mxfp4-pd1 -lc exec\ python3\ -m\ sglang.launch_server\ --model-path\ /data/hf_models/Kimi-K3\ --served-model-name\ kimi-k3\ --tp-size\ 4\ --pp-size\ 8\ --ep-size\ 4\ --nnodes\ 4\ --node-rank\ 1\ --dist-init-addr\ 174.1.60.1:20000\ --trust-remote-code\ --moe-runner-backend\ flashinfer_mxfp4\ --chunked-prefill-size\ 8192\ --page-size\ 64\ --mem-fraction-static\ 0.88\ --cuda-graph-max-bs-decode\ 16\ --mamba-radix-cache-strategy\ extra_buffer_lazy\ --disable-radix-cache\ --dist-timeout\ 3600\ --mamba-full-memory-ratio\ 0.36\ --disaggregation-transfer-backend\ mooncake\ --disaggregation-bootstrap-port\ 28800\ --disaggregation-ib-device\ mlx5_0\\\,mlx5_1\\\,mlx5_2\\\,mlx5_3\ --disaggregation-mode\ prefill\ --host\ 0.0.0.0\ --port\ 30000\  
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/commands/p_node2.cmd.txt b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/commands/p_node2.cmd.txt
new file mode 100644
index 0000000..d1f0006
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/commands/p_node2.cmd.txt
@@ -0,0 +1 @@
+docker run -d --name kimi3_pro6000_pd_pp8_standard_prefill_node2 --gpus all --network host --ipc=host --ulimit memlock=-1 --device /dev/infiniband --shm-size 32g --entrypoint bash -v /data/hf_models/Kimi-K3:/data/hf_models/Kimi-K3:ro -e CUDA_VISIBLE_DEVICES=0\,1\,2\,3\,4\,5\,6\,7 -e SGLANG_HOST_IP=174.1.60.3 -e GLOO_SOCKET_IFNAME=bond0 -e NCCL_SOCKET_IFNAME=bond1 -e NCCL_IB_HCA=mlx5_0\,mlx5_1\,mlx5_2\,mlx5_3 -e NCCL_IB_GID_INDEX=3 -e NCCL_IB_TIMEOUT=22 -e NCCL_IB_RETRY_CNT=7 -e NCCL_CUMEM_ENABLE=1 -e MOONCAKE_MASTER=174.1.60.1:50051 -e MOONCAKE_PROTOCOL=rdma -e SGLANG_ENABLE_TP_MEMORY_INBALANCE_CHECK=0 -e SGLANG_MOE_FUSED_GATE_RADIX=1 -e FLASHINFER_DISABLE_JIT=1 -e FLASHINFER_DISABLE_VERSION_CHECK=1 local/sglang:kimi-k3-sm120-flashinfer-mxfp4-pd1 -lc exec\ python3\ -m\ sglang.launch_server\ --model-path\ /data/hf_models/Kimi-K3\ --served-model-name\ kimi-k3\ --tp-size\ 4\ --pp-size\ 8\ --ep-size\ 4\ --nnodes\ 4\ --node-rank\ 2\ --dist-init-addr\ 174.1.60.1:20000\ --trust-remote-code\ --moe-runner-backend\ flashinfer_mxfp4\ --chunked-prefill-size\ 8192\ --page-size\ 64\ --mem-fraction-static\ 0.88\ --cuda-graph-max-bs-decode\ 16\ --mamba-radix-cache-strategy\ extra_buffer_lazy\ --disable-radix-cache\ --dist-timeout\ 3600\ --mamba-full-memory-ratio\ 0.36\ --disaggregation-transfer-backend\ mooncake\ --disaggregation-bootstrap-port\ 28800\ --disaggregation-ib-device\ mlx5_0\\\,mlx5_1\\\,mlx5_2\\\,mlx5_3\ --disaggregation-mode\ prefill\ --host\ 0.0.0.0\ --port\ 30000\  
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/commands/p_node3.cmd.txt b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/commands/p_node3.cmd.txt
new file mode 100644
index 0000000..70df297
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/commands/p_node3.cmd.txt
@@ -0,0 +1 @@
+docker run -d --name kimi3_pro6000_pd_pp8_standard_prefill_node3 --gpus all --network host --ipc=host --ulimit memlock=-1 --device /dev/infiniband --shm-size 32g --entrypoint bash -v /data/hf_models/Kimi-K3:/data/hf_models/Kimi-K3:ro -e CUDA_VISIBLE_DEVICES=0\,1\,2\,3\,4\,5\,6\,7 -e SGLANG_HOST_IP=174.1.60.4 -e GLOO_SOCKET_IFNAME=bond0 -e NCCL_SOCKET_IFNAME=bond1 -e NCCL_IB_HCA=mlx5_0\,mlx5_1\,mlx5_2\,mlx5_3 -e NCCL_IB_GID_INDEX=3 -e NCCL_IB_TIMEOUT=22 -e NCCL_IB_RETRY_CNT=7 -e NCCL_CUMEM_ENABLE=1 -e MOONCAKE_MASTER=174.1.60.1:50051 -e MOONCAKE_PROTOCOL=rdma -e SGLANG_ENABLE_TP_MEMORY_INBALANCE_CHECK=0 -e SGLANG_MOE_FUSED_GATE_RADIX=1 -e FLASHINFER_DISABLE_JIT=1 -e FLASHINFER_DISABLE_VERSION_CHECK=1 local/sglang:kimi-k3-sm120-flashinfer-mxfp4-pd1 -lc exec\ python3\ -m\ sglang.launch_server\ --model-path\ /data/hf_models/Kimi-K3\ --served-model-name\ kimi-k3\ --tp-size\ 4\ --pp-size\ 8\ --ep-size\ 4\ --nnodes\ 4\ --node-rank\ 3\ --dist-init-addr\ 174.1.60.1:20000\ --trust-remote-code\ --moe-runner-backend\ flashinfer_mxfp4\ --chunked-prefill-size\ 8192\ --page-size\ 64\ --mem-fraction-static\ 0.88\ --cuda-graph-max-bs-decode\ 16\ --mamba-radix-cache-strategy\ extra_buffer_lazy\ --disable-radix-cache\ --dist-timeout\ 3600\ --mamba-full-memory-ratio\ 0.36\ --disaggregation-transfer-backend\ mooncake\ --disaggregation-bootstrap-port\ 28800\ --disaggregation-ib-device\ mlx5_0\\\,mlx5_1\\\,mlx5_2\\\,mlx5_3\ --disaggregation-mode\ prefill\ --host\ 0.0.0.0\ --port\ 30000\  
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/commands/router.cmd.txt b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/commands/router.cmd.txt
new file mode 100644
index 0000000..ded65e3
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/commands/router.cmd.txt
@@ -0,0 +1 @@
+docker run -d --name kimi3_pro6000_pd_pp8_standard_router --network host --entrypoint python3 local/sglang:kimi-k3-sm120-flashinfer-mxfp4-pd1 -m sglang_router.launch_router --pd-disaggregation --mini-lb --prefill http://174.1.60.1:30000 28800 --decode http://174.1.60.5:30000 --host 0.0.0.0 --port 31000 
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/d_healthy_d_node0.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/d_healthy_d_node0.log
new file mode 100644
index 0000000..91d61b8
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/d_healthy_d_node0.log
@@ -0,0 +1,600 @@
+/usr/local/lib/python3.12/dist-packages/torchao/quantization/quant_api.py:1731: SyntaxWarning: invalid escape sequence '\.'
+  """Configuration class for applying different quantization configs to modules or parameters based on their fully qualified names (FQNs).
+/sgl-workspace/sglang/python/sglang/launch_server.py:56: UserWarning: 'python -m sglang.launch_server' is still supported, but 'sglang serve' is the recommended entrypoint.
+  Example: sglang serve --model-path <model> [options]
+  warnings.warn(
+[2026-08-27 04:36:08] KV cache is forced as chunk cache for decode server
+[2026-08-27 04:36:11] Breakable CUDA graph is incompatible with MLA attention; disabling prefill CUDA graph.
+[2026-08-27 04:36:12] Attention backend not specified. Use triton backend by default.
+[2026-08-27 04:36:14] server_args=ServerArgs(model_path='/data/hf_models/Kimi-K3', tokenizer_path='/data/hf_models/Kimi-K3', tokenizer_mode='auto', tokenizer_backend='huggingface', tokenizer_worker_num=1, detokenizer_worker_num=1, skip_tokenizer_init=False, load_format='auto', model_loader_extra_config='{}', trust_remote_code=True, context_length=None, is_embedding=False, enable_multimodal=None, revision=None, model_impl='auto', model_config_parser='auto', json_model_override_args='{}', dtype='auto', quantization=None, quantization_param_path=None, kv_cache_dtype='auto', enable_fp32_lm_head=False, modelopt_quant=None, modelopt_checkpoint_restore_path=None, modelopt_checkpoint_save_path=None, modelopt_export_path=None, quantize_and_serve=False, rl_quant_profile=None, enable_tf32_matmul=False, mem_fraction_static=0.88, max_running_requests=None, max_queued_requests=None, max_total_tokens=None, chunked_prefill_size=8192, enable_dynamic_chunking=False, max_prefill_tokens=16384, prefill_max_requests=None, schedule_policy='fcfs', enable_priority_scheduling=False, disable_priority_preemption=False, default_priority_value=None, abort_on_priority_when_disabled=False, schedule_low_priority_values_first=False, priority_scheduling_preemption_threshold=10, retraction_policy='length', schedule_conservativeness=1.0, page_size=64, swa_full_tokens_ratio=0.8, disable_hybrid_swa_memory=False, radix_eviction_policy='lru', prefill_only_disable_kv_cache=False, disable_radix_cache=True, enable_page_major_kv_layout=False, enable_unified_memory=False, disable_chunked_prefix_cache=False, disable_overlap_schedule=False, num_continuous_decode_steps=1, scheduler_recv_interval=1, enable_mixed_chunk=False, nccl_port=None, dist_timeout=3600, dist_init_addr='174.1.60.5:20000', nnodes=4, node_rank=0, tp_size=32, dcp_size=1, pp_size=1, pp_max_micro_batch_size=None, pp_async_batch_depth=0, dp_size=1, load_balance_method='round_robin', attn_cp_size=1, moe_dp_size=1, dcp_comm_backend='ag_rs', dcp_replicate_q_proj=None, dwdp_size=1, enable_prefill_cp=False, cp_strategy=None, enable_dsa_cache_layer_split=False, enable_dsa_prefill_context_parallel=False, dsa_prefill_cp_mode='round-robin-split', enable_prefill_context_parallel=False, prefill_cp_mode='in-seq-split', enable_cp_decode_attn_tp=False, enable_dp_attention=False, enable_dp_attention_local_control_broadcast=False, enable_dp_lm_head=False, enable_attn_tp_input_scattered=False, disable_attn_tp_gather=False, enable_p2p_check=False, device='cuda', base_gpu_id=0, gpu_id_step=1, random_seed=369235288, watchdog_timeout=300, soft_watchdog_timeout=None, sleep_on_idle=False, use_ray=False, custom_sigquit_handler=None, numa_node=None, gc_threshold=None, host='0.0.0.0', port=30000, fastapi_root_path='', smg_grpc_mode=False, grpc_mode=False, grpc_port=None, sidecar=None, sidecar_args=None, skip_server_warmup=False, warmups=None, enable_http2=False, ssl_keyfile=None, ssl_certfile=None, ssl_ca_certs=None, ssl_keyfile_password=None, enable_ssl_refresh=False, api_key=None, admin_api_key=None, served_model_name='kimi-k3', weight_version='default', chat_template=None, hf_chat_template_name=None, completion_template=None, file_storage_path='sglang_storage', enable_cache_report=False, reasoning_parser=None, default_chat_template_kwargs=None, strip_thinking_cache=False, enable_strict_thinking=False, tool_call_parser=None, tool_server=None, sampling_defaults='model', asr_max_buffer_seconds=60, asr_max_concurrent_sessions=32, preferred_sampling_params=None, allow_auto_truncate=False, stream_interval=1, batch_notify_size=16, stream_response_default_include_usage=False, incremental_streaming_output=False, enable_streaming_session=False, enable_session_radix_cache=False, log_level='info', log_level_http=None, log_requests=False, log_requests_level=2, log_requests_format='text', log_requests_target=None, uvicorn_access_log_exclude_prefixes=[], crash_dump_folder=None, show_time_cost=False, enable_metrics=False, smg_http_sidecar_port=None, enable_mfu_metrics=False, enable_metrics_for_all_schedulers=False, load_snapshot_publish_interval=15, tokenizer_metrics_custom_labels_header='x-custom-labels', tokenizer_metrics_allowed_custom_labels=None, extra_metric_labels=None, bucket_time_to_first_token=None, bucket_inter_token_latency=None, bucket_e2e_request_latency=None, prompt_tokens_buckets=None, generation_tokens_buckets=None, gc_warning_threshold_secs=0.0, decode_log_interval=40, enable_request_time_stats_logging=False, kv_events_config=None, enable_forward_pass_metrics=False, forward_pass_metrics_worker_id='', forward_pass_metrics_ipc_name=None, enable_trace=False, trace_modules='request', otlp_traces_endpoint='localhost:4317', export_metrics_to_file=False, export_metrics_to_file_dir=None, stat_loggers=None, constrained_json_whitespace_pattern=None, constrained_json_disable_any_whitespace=False, attention_backend='triton', decode_attention_backend=None, prefill_attention_backend=None, sampling_backend='flashinfer', grammar_backend='xgrammar', radix_cache_backend=None, mm_attention_backend=None, fp8_gemm_runner_backend='auto', fp4_gemm_runner_backend='auto', bf16_gemm_backend='auto', dsa_prefill_backend=None, dsa_decode_backend=None, dsa_paged_mqa_logits_backend='auto', dsa_topk_backend='sgl-kernel', disable_flashinfer_autotune=False, flashinfer_autotune_skip_ops=None, mamba_backend='triton', cuda_graph_config=CudaGraphConfig(decode=PhaseConfig(backend='full', max_bs=16, bs=[1, 2, 4, 8, 12, 16], tc_compiler='eager', full_prefill_max_req=None), prefill=PhaseConfig(backend='disabled', max_bs=2048, bs=[4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], tc_compiler='eager', full_prefill_max_req=None)), cuda_graph_backend_decode=None, cuda_graph_backend_prefill=None, cuda_graph_max_bs_decode=16, cuda_graph_max_bs_prefill=None, cuda_graph_bs_decode=None, cuda_graph_bs_prefill=None, cuda_graph_tc_compiler=None, disable_prefill_cuda_graph=False, disable_decode_cuda_graph=False, disable_cuda_graph=False, disable_cuda_graph_padding=False, enable_profile_cuda_graph=False, enable_cudagraph_gc=False, debug_cuda_graph=False, enable_layerwise_nvtx_marker=False, enable_nccl_nvls=False, enable_symm_mem=False, triton_attention_reduce_in_fp32=False, triton_attention_num_kv_splits=8, triton_attention_split_tile_size=None, flashinfer_mla_disable_ragged=False, enable_fused_qk_norm_rope=False, enable_precise_embedding_interpolation=False, enable_fused_moe_sum_all_reduce=False, enable_deepseek_v4_fp4_indexer=False, disable_custom_all_reduce=False, enable_mscclpp=False, enable_torch_symm_mem=False, enable_scattered_sconv=False, pre_warm_nccl=False, enable_quant_communications=False, enable_flashinfer_allreduce_fusion=False, enforce_disable_flashinfer_allreduce_fusion=False, flashinfer_allreduce_fusion_backend=None, enable_aiter_allreduce_fusion=False, enable_torch_compile=False, enable_torch_compile_debug_mode=False, torch_compile_max_bs=32, torchao_config='', speculative_algorithm=None, speculative_draft_model_path=None, speculative_draft_model_revision=None, speculative_draft_load_format=None, speculative_num_steps=None, speculative_eagle_topk=None, speculative_num_draft_tokens=None, speculative_dflash_block_size=None, speculative_dspark_block_size=None, speculative_dspark_sps_table_path=None, speculative_dspark_confidence_sts_path=None, speculative_dspark_align_verify_tokens_to_graph_tier=False, speculative_accept_threshold_single=1.0, speculative_accept_threshold_acc=1.0, speculative_use_rejection_sampling=False, speculative_token_map=None, speculative_attention_mode='prefill', speculative_draft_attention_backend=None, speculative_draft_window_size=None, speculative_moe_runner_backend='marlin', speculative_moe_a2a_backend=None, speculative_draft_model_quantization=None, speculative_skip_dp_mlp_sync=False, enable_multi_layer_eagle=False, speculative_adaptive=False, speculative_adaptive_config=None, decoupled_spec_bind_endpoint=None, decoupled_spec_connect_endpoints=None, decoupled_spec_rank=None, decoupled_spec_role='null', spec_trace_dir=None, speculative_ngram_min_bfs_breadth=1, speculative_ngram_max_bfs_breadth=10, speculative_ngram_match_type='BFS', speculative_ngram_max_trie_depth=18, speculative_ngram_capacity=10000000, speculative_ngram_external_corpus_path=None, speculative_ngram_external_sam_budget=0, speculative_ngram_external_corpus_max_tokens=10000000, ep_size=32, moe_a2a_backend='none', moe_runner_backend='marlin', flashinfer_mxfp4_moe_precision='default', deepep_mode='auto', fuseep_mode=2, deepep_dispatcher_output_dtype='auto', ep_num_redundant_experts=0, ep_dispatch_algorithm=None, init_expert_location='trivial', enable_eplb=False, eplb_algorithm='auto', eplb_rebalance_num_iterations=1000, eplb_rebalance_layers_per_chunk=None, eplb_min_rebalancing_utilization_threshold=1.0, expert_distribution_recorder_mode=None, expert_distribution_recorder_buffer_size=1000, enable_expert_distribution_metrics=False, deepep_config=None, moe_dense_tp_size=None, elastic_ep_backend=None, enable_elastic_expert_backup=False, mooncake_ib_device=None, enable_waterfill=False, ep_join_mode=None, ep_join_rank_offset=0, elastic_ep_initial_size=None, max_ep_size=None, elastic_ep_scale_timeout=600, elastic_ep_rejoin=False, disable_flashinfer_cutlass_moe_fp4_allgather=False, disable_shared_experts_fusion=False, enforce_shared_experts_fusion=False, max_mamba_cache_size=None, mamba_ssm_dtype=None, mamba_max_states_per_path=-1, enable_mamba_cache_stochastic_rounding=False, mamba_cache_philox_rounds=0, mamba_full_memory_ratio=0.21, mamba_radix_cache_strategy='extra_buffer_lazy', uses_mamba_radix_cache=False, mamba_track_interval=256, enable_int8_mamba_checkpoint=False, int8_mamba_ckpt_size=None, linear_attn_backend='triton', linear_attn_decode_backend=None, linear_attn_prefill_backend=None, linear_attn_verify_backend=None, enable_linear_replayssm=False, linear_replayssm_cache_len=16, enable_linear_replayssm_spec=False, enable_hierarchical_cache=False, hicache_ratio=2.0, hicache_size=0, hicache_write_policy='write_through', hicache_io_backend='kernel', hicache_mem_layout='page_first', hicache_storage_backend=None, hicache_storage_prefetch_policy='timeout', hicache_storage_backend_extra_config=None, enable_hisparse=False, hisparse_config=None, enable_broadcast_mm_inputs_process=False, enable_prefix_mm_cache=False, mm_enable_dp_encoder=False, mm_process_config={}, mm_processor_worker_num=0, mm_io_worker_num=0, limit_mm_data_per_request=None, enable_mm_global_cache=False, disable_fast_image_processor=False, mm_feature_transport='cpu', keep_mm_feature_on_device=False, enable_lora=None, enable_lora_overlap_loading=None, max_lora_rank=None, lora_target_modules=None, lora_paths=None, max_loaded_loras=None, max_loras_per_batch=8, lora_eviction_policy='lru', lora_backend='csgmv', max_lora_chunk_size=16, experts_shared_outer_loras=None, lora_use_virtual_experts=False, lora_strict_loading=False, lora_drain_wait_threshold=0.0, enable_two_batch_overlap=False, enable_single_batch_overlap=False, tbo_token_distribution_threshold=0.48, cpu_offload_gb=0, offload_group_size=-1, offload_num_in_group=1, offload_prefetch_step=1, offload_mode='cpu', enable_lmcache=False, lmcache_config_file=None, enable_flexkv=False, flexkv_config_file=None, kt_weight_path=None, kt_method='AMXINT4', kt_cpuinfer=None, kt_threadpool_count=2, kt_num_gpu_experts=None, kt_max_deferred_experts_per_token=None, dllm_algorithm=None, dllm_algorithm_config=None, dllm_fdfo=True, disaggregation_mode='decode', disaggregation_transfer_backend='mooncake', disaggregation_bootstrap_port=28800, disaggregation_ib_device='mlx5_0,mlx5_1,mlx5_2,mlx5_3', disaggregation_decode_enable_radix_cache=False, disaggregation_decode_enable_offload_kvcache=False, num_reserved_decode_tokens=512, disaggregation_decode_extra_slots=0, disaggregation_decode_polling_interval=1, optimistic_prefill_attempts=0, encoder_only=False, language_only=False, encoder_transfer_backend='zmq_to_tokenizer', encoder_urls=[], encoder_bootstrap_port=8997, encoder_register_urls=[], enable_adaptive_dispatch_to_encoder=False, enable_pdmux=False, pdmux_config_path=None, sm_group_num=8, custom_weight_loader=[], weight_loader_disable_mmap=False, weight_loader_prefetch_checkpoints=False, weight_loader_prefetch_num_threads=4, weight_loader_drop_cache_after_load=False, remote_instance_weight_loader_seed_instance_ip=None, remote_instance_weight_loader_seed_instance_service_port=None, remote_instance_weight_loader_send_weights_group_ports=None, remote_instance_weight_loader_backend='nccl', remote_instance_weight_loader_start_seed_via_transfer_engine=False, engine_info_bootstrap_port=6789, modelexpress_config=None, download_dir=None, model_checksum=None, delete_ckpt_after_loading=False, decrypted_config_file=None, decrypted_draft_config_file=None, checkpoint_engine_wait_weights_before_ready=False, enable_prefill_delayer=False, prefill_delayer_max_delay_passes=30, prefill_delayer_token_usage_low_watermark=None, prefill_delayer_forward_passes_buckets=None, prefill_delayer_wait_seconds_buckets=None, prefill_delayer_queue_min_ratio=None, prefill_delayer_max_delay_ms=None, min_free_slots_delay=None, enable_deterministic_inference=False, rl_on_policy_target=None, kv_canary='none', kv_canary_real_data='none', kv_canary_sweep_interval=0, enable_dynamic_batch_tokenizer=False, dynamic_batch_tokenizer_batch_size=32, dynamic_batch_tokenizer_batch_timeout=0.002, enable_tokenizer_batch_encode=False, disable_tokenizer_batch_decode=False, debug_tensor_dump_output_folder=None, debug_tensor_dump_layers=None, debug_tensor_dump_input_file=None, enable_memory_saver=False, enable_weights_cpu_backup=False, enable_draft_weights_cpu_backup=False, enable_custom_logit_processor=False, enable_return_hidden_states=False, enable_return_routed_experts=False, enable_return_indexer_topk=False, disable_outlines_disk_cache=False, enable_mis=False, weight_cache_mode='off', weight_cache_socket=None, weight_cache_timeout=1800, forward_hooks=None, msprobe_dump_config=None)
+[2026-08-27 04:36:14] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:14] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:14] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:14] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:14] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:14] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:14] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:14] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:17] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:17] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:18] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:19] Multimodal processor concurrency enabled with 2 isolated worker threads (auto).
+[2026-08-27 04:36:19] No HuggingFace chat template found
+[2026-08-27 04:36:19] No chat template found, defaulting to 'string' content format
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:34 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:34 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:34 TP4 EP4] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:34 TP4 EP4] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:34 TP6 EP6] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:34 TP6 EP6] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:34] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:34] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:34 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:34 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+/sgl-workspace/sglang/python/sglang/srt/utils/hf_transformers/tokenizer.py:423: UserWarning: Using a slow tokenizer. This might cause a significant slowdown. Consider using a fast tokenizer instead.
+  warnings.warn(
+[2026-08-27 04:36:34] Applying special tokens cache patch for Kimi tokenizer: <class 'transformers_modules.Kimi_hyphen_K3.caf40b2773f9029e.tokenization_kimi.TikTokenTokenizer'>
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:34 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:34 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:34 TP7 EP7] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:34 TP7 EP7] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:34 TP5 EP5] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:34 TP5 EP5] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:35 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:35 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:36 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:36 TP6 EP6] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:36 TP4 EP4] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:36 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:36 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:36 TP7 EP7] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:36 TP5 EP5] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:37 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:37.190651   236 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:37.190778   236 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:37.190796   236 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.5 port: 12001
+I0827 04:36:37.190856   236 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.5:16500
+I0827 04:36:37.191038   236 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:37.194572   241 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:37.194692   241 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:37.194712   241 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.5 port: 12001
+I0827 04:36:37.194767   241 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.5:15678
+I0827 04:36:37.194965   241 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:37.219420   236 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:37.223201   241 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:37.227941   236 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:37.231628   241 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:37.236233   236 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:37.239850   241 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:37.244647   236 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:37.248145   241 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:37.248418   236 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:37.248464   236 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:37.248525   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.250921   241 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:37.250947   241 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:37.251004   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.255973   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:37.257668   236 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:05
+I0827 04:36:37.257705   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.283210   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:37.283238   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:37.284873   236 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:05
+I0827 04:36:37.284902   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.284904   241 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:05
+I0827 04:36:37.284934   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.293713   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:37.295406   241 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:05
+I0827 04:36:37.295433   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.311285   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:37.312896   236 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:05
+I0827 04:36:37.312925   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.327399   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:37.328944   241 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:05
+I0827 04:36:37.328972   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.339579   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:37.341301   236 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:05
+I0827 04:36:37.341434   236 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:37 TP1 EP1] Init torch distributed begin.
+I0827 04:36:37.359290   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:37.360814   241 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:05
+I0827 04:36:37.360936   241 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:37 TP6 EP6] Init torch distributed begin.
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:37.366210   239 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:37.366365   239 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:37.366390   239 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.5 port: 12001
+I0827 04:36:37.366467   239 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.5:15787
+I0827 04:36:37.366704   239 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:37.395485   239 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:37.402652   239 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:37.409775   239 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:37.434130   239 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:37.436257   239 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:37.436282   239 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:37.436331   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.442425   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:37.443729   239 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:05
+I0827 04:36:37.443749   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.448421   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:37.449750   239 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:05
+I0827 04:36:37.449769   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.454622   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:37.455952   239 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:05
+I0827 04:36:37.455972   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.463356   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:37.464902   239 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:05
+I0827 04:36:37.465021   239 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:37 TP4 EP4] Init torch distributed begin.
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:37.510618   238 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:37.510744   238 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:37.510763   238 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.5 port: 12001
+I0827 04:36:37.510824   238 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.5:16205
+I0827 04:36:37.511003   238 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:37.524268   242 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:37.524391   242 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:37.524411   242 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.5 port: 12001
+I0827 04:36:37.524524   242 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.5:16605
+I0827 04:36:37.524709   242 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:37.528693   235 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:37.528817   235 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:37.528836   235 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.5 port: 12001
+I0827 04:36:37.528893   235 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.5:16672
+I0827 04:36:37.529069   235 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:37.535192   238 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:37.535425   235 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:37.541376   235 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:37.542625   238 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:37.547586   235 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:37.549854   238 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:37.551366   242 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:37.553905   235 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:37.556406   235 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:37.556468   235 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:37.556523   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.558794   242 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:37.563313   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:37.566102   242 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:37.573271   242 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:37.574895   238 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:37.576385   242 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:37.576412   242 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:37.576478   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.577947   238 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:37.577976   238 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:37.578033   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.583841   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:37.585472   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:37.585623   242 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:05
+I0827 04:36:37.585656   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.587072   238 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:05
+I0827 04:36:37.587105   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.594069   235 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:05
+I0827 04:36:37.594128   235 rdma_context.cpp:205] Using SIEVE endpoint store
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:37.598918   240 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:37.599033   240 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:37.599050   240 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.5 port: 12001
+I0827 04:36:37.599107   240 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.5:15428
+I0827 04:36:37.599270   240 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:37.605341   240 topology.cpp:127] Device mlx5_0 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:37.606741   237 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:37.606841   237 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:37.606855   237 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.5 port: 12001
+I0827 04:36:37.606901   237 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.5:16531
+I0827 04:36:37.607028   237 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:37.611215   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:37.611989   237 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:37.612797   242 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:05
+I0827 04:36:37.612828   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.615110   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:37.616685   238 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:05
+I0827 04:36:37.616714   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.619258   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:37.620560   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:37.620851   235 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:05
+I0827 04:36:37.620885   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.622239   242 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:05
+I0827 04:36:37.622268   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.624459   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:37.625998   238 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:05
+I0827 04:36:37.626025   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.630085   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:37.630686   240 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:37.631680   242 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:05
+I0827 04:36:37.631806   242 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:37 TP7 EP7] Init torch distributed begin.
+I0827 04:36:37.633723   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:37.635268   238 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:05
+I0827 04:36:37.635390   238 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:37 TP3 EP3] Init torch distributed begin.
+I0827 04:36:37.642561   237 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:37.649834   237 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:37.651238   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:37.652786   235 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:05
+I0827 04:36:37.652813   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.657148   237 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:37.658787   240 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:37.659991   237 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:37.660017   237 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:37.660071   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.667523   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:37.669195   240 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:37.669226   237 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:05
+I0827 04:36:37.669257   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.672511   240 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:37.672547   240 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:37.672606   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.677067   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:37.678661   237 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:05
+I0827 04:36:37.678689   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.682178   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:37.683274   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:37.683740   240 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:05
+I0827 04:36:37.683773   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.684834   235 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:05
+I0827 04:36:37.684954   235 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:37 TP0 EP0] Init torch distributed begin.
+I0827 04:36:37.703369   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:37.704975   237 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:05
+I0827 04:36:37.705003   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.711591   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:37.713003   240 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:05
+I0827 04:36:37.713030   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.735574   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:37.737313   237 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:05
+I0827 04:36:37.737439   237 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:37 TP2 EP2] Init torch distributed begin.
+I0827 04:36:37.740485   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:37.741588   240 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:05
+I0827 04:36:37.741605   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.766494   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:37.767982   240 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:05
+I0827 04:36:37.768137   240 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:37 TP5 EP5] Init torch distributed begin.
+[2026-08-27 04:36:46 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:36:48 TP0 EP0] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP1 EP1] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP2 EP2] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP5 EP5] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP4 EP4] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP3 EP3] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP6 EP6] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP7 EP7] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP0 EP0] Init torch distributed ends. elapsed=10.55 s, mem usage=0.40 GB
+[2026-08-27 04:36:48 TP7 EP7] Init torch distributed ends. elapsed=10.61 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP6 EP6] Init torch distributed ends. elapsed=10.88 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP4 EP4] Init torch distributed ends. elapsed=10.77 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP3 EP3] Init torch distributed ends. elapsed=10.60 s, mem usage=0.40 GB
+[2026-08-27 04:36:48 TP1 EP1] Init torch distributed ends. elapsed=10.90 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP5 EP5] Init torch distributed ends. elapsed=10.47 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP2 EP2] Init torch distributed ends. elapsed=10.50 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP7 EP7] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP4 EP4] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP1 EP1] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP6 EP6] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP7 EP7] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP4 EP4] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP5 EP5] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP6 EP6] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP3 EP3] Load weight begin. avail mem=82.19 GB
+[2026-08-27 04:36:48 TP0 EP0] Load weight begin. avail mem=82.19 GB
+[2026-08-27 04:36:48 TP5 EP5] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP2 EP2] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP4 EP4] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP7 EP7] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP6 EP6] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP4 EP4] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP7 EP7] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP6 EP6] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP5 EP5] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP5 EP5] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:50 TP1 EP1] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP0 EP0] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP2 EP2] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP3 EP3] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP5 EP5] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP4 EP4] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP6 EP6] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP7 EP7] multimem all-gather disabled because the TP group spans across nodes.
+
Multi-thread loading shards:   0% Completed | 0/96 [00:00<?, ?it/s]
Multi-thread loading shards:   1% Completed | 1/96 [00:00<00:24,  3.85it/s]
Multi-thread loading shards:   2% Completed | 2/96 [00:07<07:13,  4.61s/it]
Multi-thread loading shards:   3% Completed | 3/96 [00:09<04:44,  3.06s/it]
Multi-thread loading shards:   4% Completed | 4/96 [00:10<03:33,  2.32s/it]
Multi-thread loading shards:   5% Completed | 5/96 [00:11<02:54,  1.92s/it]
Multi-thread loading shards:   6% Completed | 6/96 [00:12<02:29,  1.66s/it]
Multi-thread loading shards:   7% Completed | 7/96 [00:13<02:13,  1.50s/it]
Multi-thread loading shards:   8% Completed | 8/96 [00:15<02:04,  1.41s/it]
Multi-thread loading shards:   9% Completed | 9/96 [00:16<01:56,  1.34s/it]
Multi-thread loading shards:  10% Completed | 10/96 [00:17<01:47,  1.25s/it]
Multi-thread loading shards:  11% Completed | 11/96 [00:18<01:44,  1.23s/it]
Multi-thread loading shards:  12% Completed | 12/96 [00:19<01:43,  1.23s/it]
Multi-thread loading shards:  14% Completed | 13/96 [00:21<01:59,  1.44s/it]
Multi-thread loading shards:  15% Completed | 14/96 [00:22<01:51,  1.36s/it]
Multi-thread loading shards:  16% Completed | 15/96 [00:24<01:45,  1.30s/it]
Multi-thread loading shards:  17% Completed | 16/96 [00:25<01:42,  1.28s/it]
Multi-thread loading shards:  18% Completed | 17/96 [00:26<01:39,  1.26s/it]
Multi-thread loading shards:  19% Completed | 18/96 [00:27<01:36,  1.24s/it]
Multi-thread loading shards:  20% Completed | 19/96 [00:28<01:32,  1.20s/it]
Multi-thread loading shards:  21% Completed | 20/96 [00:29<01:30,  1.19s/it]
Multi-thread loading shards:  22% Completed | 21/96 [00:31<01:29,  1.19s/it]
Multi-thread loading shards:  23% Completed | 22/96 [00:32<01:27,  1.18s/it]
Multi-thread loading shards:  24% Completed | 23/96 [00:33<01:26,  1.18s/it]
Multi-thread loading shards:  25% Completed | 24/96 [00:34<01:23,  1.16s/it]
Multi-thread loading shards:  26% Completed | 25/96 [00:35<01:20,  1.14s/it]
Multi-thread loading shards:  27% Completed | 26/96 [00:36<01:18,  1.13s/it]
Multi-thread loading shards:  28% Completed | 27/96 [00:37<01:19,  1.15s/it]
Multi-thread loading shards:  29% Completed | 28/96 [00:39<01:18,  1.16s/it]
Multi-thread loading shards:  30% Completed | 29/96 [00:40<01:17,  1.16s/it]
Multi-thread loading shards:  31% Completed | 30/96 [00:41<01:17,  1.18s/it]
Multi-thread loading shards:  32% Completed | 31/96 [00:42<01:16,  1.18s/it]
Multi-thread loading shards:  33% Completed | 32/96 [00:44<01:30,  1.42s/it]
Multi-thread loading shards:  34% Completed | 33/96 [00:45<01:23,  1.33s/it]
Multi-thread loading shards:  35% Completed | 34/96 [00:46<01:17,  1.26s/it]
Multi-thread loading shards:  36% Completed | 35/96 [00:48<01:14,  1.23s/it]
Multi-thread loading shards:  38% Completed | 36/96 [00:49<01:13,  1.22s/it]
Multi-thread loading shards:  39% Completed | 37/96 [00:50<01:11,  1.21s/it]
Multi-thread loading shards:  40% Completed | 38/96 [00:51<01:09,  1.20s/it]
Multi-thread loading shards:  41% Completed | 39/96 [00:52<01:07,  1.18s/it]
Multi-thread loading shards:  42% Completed | 40/96 [00:53<01:05,  1.18s/it]
Multi-thread loading shards:  43% Completed | 41/96 [00:55<01:04,  1.17s/it]
Multi-thread loading shards:  44% Completed | 42/96 [00:56<01:02,  1.17s/it]
Multi-thread loading shards:  45% Completed | 43/96 [00:57<01:03,  1.19s/it]
Multi-thread loading shards:  46% Completed | 44/96 [00:58<01:01,  1.18s/it]
Multi-thread loading shards:  47% Completed | 45/96 [00:59<00:57,  1.13s/it]
Multi-thread loading shards:  48% Completed | 46/96 [01:00<00:57,  1.15s/it]
Multi-thread loading shards:  49% Completed | 47/96 [01:02<00:57,  1.17s/it]
Multi-thread loading shards:  50% Completed | 48/96 [01:03<00:55,  1.15s/it]
Multi-thread loading shards:  51% Completed | 49/96 [01:04<00:53,  1.13s/it]
Multi-thread loading shards:  52% Completed | 50/96 [01:06<01:01,  1.34s/it]
Multi-thread loading shards:  53% Completed | 51/96 [01:07<00:58,  1.29s/it]
Multi-thread loading shards:  54% Completed | 52/96 [01:08<00:55,  1.27s/it]
Multi-thread loading shards:  55% Completed | 53/96 [01:09<00:53,  1.24s/it]
Multi-thread loading shards:  56% Completed | 54/96 [01:10<00:50,  1.21s/it]
Multi-thread loading shards:  57% Completed | 55/96 [01:12<00:49,  1.22s/it]
Multi-thread loading shards:  58% Completed | 56/96 [01:13<00:48,  1.22s/it]
Multi-thread loading shards:  59% Completed | 57/96 [01:14<00:47,  1.21s/it]
Multi-thread loading shards:  60% Completed | 58/96 [01:15<00:45,  1.20s/it]
Multi-thread loading shards:  61% Completed | 59/96 [01:16<00:44,  1.19s/it]
Multi-thread loading shards:  62% Completed | 60/96 [01:17<00:43,  1.20s/it]
Multi-thread loading shards:  64% Completed | 61/96 [01:19<00:42,  1.21s/it]
Multi-thread loading shards:  65% Completed | 62/96 [01:20<00:40,  1.19s/it]
Multi-thread loading shards:  66% Completed | 63/96 [01:21<00:40,  1.21s/it]
Multi-thread loading shards:  67% Completed | 64/96 [01:22<00:38,  1.20s/it]
Multi-thread loading shards:  68% Completed | 65/96 [01:24<00:37,  1.20s/it]
Multi-thread loading shards:  69% Completed | 66/96 [01:25<00:36,  1.20s/it]
Multi-thread loading shards:  70% Completed | 67/96 [01:26<00:35,  1.22s/it]
Multi-thread loading shards:  71% Completed | 68/96 [01:28<00:41,  1.48s/it]
Multi-thread loading shards:  72% Completed | 69/96 [01:29<00:37,  1.39s/it]
Multi-thread loading shards:  73% Completed | 70/96 [01:30<00:34,  1.34s/it]
Multi-thread loading shards:  74% Completed | 71/96 [01:32<00:32,  1.29s/it]
Multi-thread loading shards:  75% Completed | 72/96 [01:33<00:31,  1.30s/it]
Multi-thread loading shards:  76% Completed | 73/96 [01:34<00:29,  1.29s/it]
Multi-thread loading shards:  77% Completed | 74/96 [01:35<00:27,  1.27s/it]
Multi-thread loading shards:  78% Completed | 75/96 [01:37<00:26,  1.26s/it]
Multi-thread loading shards:  79% Completed | 76/96 [01:38<00:24,  1.24s/it]
Multi-thread loading shards:  80% Completed | 77/96 [01:39<00:23,  1.23s/it]
Multi-thread loading shards:  81% Completed | 78/96 [01:40<00:21,  1.19s/it]
Multi-thread loading shards:  82% Completed | 79/96 [01:41<00:20,  1.20s/it]
Multi-thread loading shards:  83% Completed | 80/96 [01:43<00:19,  1.20s/it]
Multi-thread loading shards:  84% Completed | 81/96 [01:44<00:18,  1.21s/it]
Multi-thread loading shards:  85% Completed | 82/96 [01:45<00:16,  1.20s/it]
Multi-thread loading shards:  86% Completed | 83/96 [01:46<00:15,  1.17s/it]
Multi-thread loading shards:  88% Completed | 84/96 [01:47<00:14,  1.17s/it]
Multi-thread loading shards:  89% Completed | 85/96 [01:48<00:12,  1.10s/it]
Multi-thread loading shards:  90% Completed | 86/96 [01:49<00:10,  1.05s/it]
Multi-thread loading shards:  91% Completed | 87/96 [01:50<00:09,  1.02s/it]
Multi-thread loading shards:  92% Completed | 88/96 [01:51<00:07,  1.01it/s]
Multi-thread loading shards:  93% Completed | 89/96 [01:52<00:06,  1.03it/s]
Multi-thread loading shards:  94% Completed | 90/96 [01:53<00:05,  1.04it/s]
Multi-thread loading shards:  95% Completed | 91/96 [01:54<00:04,  1.06it/s]
Multi-thread loading shards:  96% Completed | 92/96 [01:55<00:03,  1.06it/s]
Multi-thread loading shards:  97% Completed | 93/96 [01:56<00:02,  1.08it/s]
Multi-thread loading shards:  98% Completed | 94/96 [01:56<00:01,  1.46it/s]
Multi-thread loading shards: 100% Completed | 96/96 [01:56<00:00,  2.28it/s]
Multi-thread loading shards: 100% Completed | 96/96 [01:56<00:00,  1.21s/it]
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:58 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP6 EP6] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP0 EP0] Load weight end. elapsed=130.28 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP6 EP6] Load weight end. elapsed=130.29 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP3 EP3] Load weight end. elapsed=130.28 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP2 EP2] Load weight end. elapsed=130.32 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP5 EP5] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP5 EP5] Load weight end. elapsed=130.38 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP4 EP4] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP4 EP4] Load weight end. elapsed=130.48 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP7 EP7] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP7 EP7] Load weight end. elapsed=130.48 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:39:01 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:39:01 TP1 EP1] Load weight end. elapsed=132.81 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:39:01 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP6 EP6] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP6 EP6] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP5 EP5] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP5 EP5] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP7 EP7] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP7 EP7] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP4 EP4] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP4 EP4] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:02 TP0 EP0] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP7 EP7] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP6 EP6] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP5 EP5] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP1 EP1] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP4 EP4] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP3 EP3] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP2 EP2] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP7 EP7] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP4 EP4] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP6 EP6] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP5 EP5] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP0 EP0] Memory pool end. avail mem=9.18 GB
+[2026-08-27 04:39:02 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP1 EP1] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP6 EP6] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP5 EP5] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP7 EP7] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP6 EP6] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP5 EP5] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP7 EP7] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP4 EP4] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP4 EP4] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP2 EP2] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP3 EP3] Memory pool end. avail mem=9.18 GB
+[2026-08-27 04:39:02 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:39:02 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:39:02 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP0 EP0] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.17 GB
+[2026-08-27 04:39:02 TP6 EP6] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP6 EP6] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP1 EP1] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP5 EP5] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP7 EP7] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP5 EP5] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP7 EP7] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:39:02 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP3 EP3] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.17 GB
+[2026-08-27 04:39:02 TP4 EP4] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP4 EP4] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:39:02 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP2 EP2] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+
  0%|          | 0/6 [00:00<?, ?it/s]
Capturing batches (bs=16 avail_mem=9.00 GB):   0%|          | 0/6 [00:00<?, ?it/s]
Capturing batches (bs=16 avail_mem=9.00 GB):  17%|█▋        | 1/6 [03:43<18:38, 223.66s/it]
Capturing batches (bs=12 avail_mem=8.84 GB):  17%|█▋        | 1/6 [03:43<18:38, 223.66s/it]
Capturing batches (bs=12 avail_mem=8.84 GB):  33%|███▎      | 2/6 [03:44<06:11, 92.84s/it] 
Capturing batches (bs=8 avail_mem=8.78 GB):  33%|███▎      | 2/6 [03:44<06:11, 92.84s/it] 
Capturing batches (bs=8 avail_mem=8.78 GB):  50%|█████     | 3/6 [03:49<02:36, 52.32s/it]
Capturing batches (bs=4 avail_mem=8.73 GB):  50%|█████     | 3/6 [03:49<02:36, 52.32s/it]
Capturing batches (bs=4 avail_mem=8.73 GB):  67%|██████▋   | 4/6 [03:49<01:03, 32.00s/it]
Capturing batches (bs=2 avail_mem=8.68 GB):  67%|██████▋   | 4/6 [03:49<01:03, 32.00s/it]
Capturing batches (bs=2 avail_mem=8.68 GB):  83%|████████▎ | 5/6 [03:50<00:20, 20.77s/it]
Capturing batches (bs=1 avail_mem=8.62 GB):  83%|████████▎ | 5/6 [03:50<00:20, 20.77s/it]
Capturing batches (bs=1 avail_mem=8.62 GB): 100%|██████████| 6/6 [03:52<00:00, 14.36s/it]
Capturing batches (bs=1 avail_mem=8.62 GB): 100%|██████████| 6/6 [03:52<00:00, 38.78s/it]
+[2026-08-27 04:42:56 TP4 EP4] Capture target decode CUDA graph end. elapsed=233.75 s, mem usage=0.60 GB, avail mem=8.52 GB.
+[2026-08-27 04:42:56 TP4 EP4] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP3 EP3] Capture target decode CUDA graph end. elapsed=233.75 s, mem usage=0.62 GB, avail mem=8.55 GB.
+[2026-08-27 04:42:56 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP0 EP0] Capture target decode CUDA graph end. elapsed=233.77 s, mem usage=0.60 GB, avail mem=8.56 GB.
+[2026-08-27 04:42:56 TP0 EP0] max_total_num_tokens=440256, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=181, context_len=1048576, available_gpu_mem=8.56 GB
+[2026-08-27 04:42:56 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP7 EP7] Capture target decode CUDA graph end. elapsed=233.77 s, mem usage=0.62 GB, avail mem=8.51 GB.
+[2026-08-27 04:42:56 TP7 EP7] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP6 EP6] Capture target decode CUDA graph end. elapsed=233.86 s, mem usage=0.56 GB, avail mem=8.59 GB.
+[2026-08-27 04:42:56 TP6 EP6] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP1 EP1] Capture target decode CUDA graph end. elapsed=233.86 s, mem usage=0.58 GB, avail mem=8.58 GB.
+[2026-08-27 04:42:56 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP2 EP2] Capture target decode CUDA graph end. elapsed=233.86 s, mem usage=0.56 GB, avail mem=8.59 GB.
+[2026-08-27 04:42:56 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP5 EP5] Capture target decode CUDA graph end. elapsed=233.90 s, mem usage=0.58 GB, avail mem=8.58 GB.
+[2026-08-27 04:42:56 TP5 EP5] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+W0827 04:42:57.550421  2569 memory_location.cpp:73] Failed to get NUMA node, addr: 0x599a0800, len: 23168: Operation not permitted [1]
+W0827 04:42:57.550521  2570 memory_location.cpp:73] Failed to get NUMA node, addr: 0x599cb380, len: 23168: Operation not permitted [1]
+W0827 04:42:57.550542  2571 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59e1b540, len: 23168: Operation not permitted [1]
+W0827 04:42:57.550628  2572 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59c8a2c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.550725  2573 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5be7a240, len: 185344: Operation not permitted [1]
+W0827 04:42:57.550769  2574 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5bea7680, len: 185344: Operation not permitted [1]
+W0827 04:42:57.550900  2575 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59cc48c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.551011  2576 memory_location.cpp:73] Failed to get NUMA node, addr: 0x45f39fc0, len: 46336: Operation not permitted [1]
+W0827 04:42:57.551070  2577 memory_location.cpp:73] Failed to get NUMA node, addr: 0x599bfbc0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.551105  2578 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59c69c80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.580384  2676 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59c988c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.580493  2678 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59a30c40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.580636  2677 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59d06340, len: 23168: Operation not permitted [1]
+W0827 04:42:57.580636  2679 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5bd91640, len: 185344: Operation not permitted [1]
+W0827 04:42:57.580685  2675 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59e30d80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.580811  2681 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59f53440, len: 23168: Operation not permitted [1]
+W0827 04:42:57.580950  2680 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5bdbea80, len: 185344: Operation not permitted [1]
+W0827 04:42:57.581368  2686 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59dca000, len: 23168: Operation not permitted [1]
+W0827 04:42:57.581421  2682 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5987fc00, len: 46336: Operation not permitted [1]
+W0827 04:42:57.581741  2684 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59e10480, len: 23168: Operation not permitted [1]
+W0827 04:42:57.597819  2845 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59224900, len: 23168: Operation not permitted [1]
+W0827 04:42:57.598558  2839 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5917e740, len: 23168: Operation not permitted [1]
+W0827 04:42:57.598562  2849 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58e6ce00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.598762  2838 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58e982c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.598834  2841 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5b32ec00, len: 185344: Operation not permitted [1]
+W0827 04:42:57.598897  2834 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5927b680, len: 23168: Operation not permitted [1]
+W0827 04:42:57.598922  2843 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5b35c040, len: 185344: Operation not permitted [1]
+W0827 04:42:57.598953  2836 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5939fc40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.599097  2850 memory_location.cpp:73] Failed to get NUMA node, addr: 0x592d3fc0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.599287  2847 memory_location.cpp:73] Failed to get NUMA node, addr: 0x453f4c40, len: 46336: Operation not permitted [1]
+W0827 04:42:57.612057  2918 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59f4a140, len: 23168: Operation not permitted [1]
+W0827 04:42:57.612175  2919 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59f098c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.612243  2920 memory_location.cpp:73] Failed to get NUMA node, addr: 0x46045dc0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.612710  2925 memory_location.cpp:73] Failed to get NUMA node, addr: 0x46106480, len: 46336: Operation not permitted [1]
+W0827 04:42:57.612715  2924 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59a7ab40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.612880  2922 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5bf25300, len: 185344: Operation not permitted [1]
+W0827 04:42:57.612898  2927 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59e8a9c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.612931  2921 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59fec840, len: 23168: Operation not permitted [1]
+W0827 04:42:57.612948  2923 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5bf52740, len: 185344: Operation not permitted [1]
+W0827 04:42:57.612996  2926 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59edf040, len: 23168: Operation not permitted [1]
+W0827 04:42:57.618498  2975 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a4cda80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.618582  2974 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a3878c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.618582  2973 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a3abc80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.618669  2976 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5c40da00, len: 185344: Operation not permitted [1]
+W0827 04:42:57.618669  2978 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59f66a40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.618741  2977 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5c43ae40, len: 185344: Operation not permitted [1]
+W0827 04:42:57.618854  2979 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59ce6c00, len: 46336: Operation not permitted [1]
+W0827 04:42:57.618898  2980 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a2ebe80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.619125  2982 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a37d0c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.632592  2972 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a371340, len: 23168: Operation not permitted [1]
+W0827 04:42:57.776260  3310 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58ee7080, len: 23168: Operation not permitted [1]
+W0827 04:42:57.776290  3311 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5854f480, len: 23168: Operation not permitted [1]
+W0827 04:42:57.776294  3312 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58edd900, len: 23168: Operation not permitted [1]
+W0827 04:42:57.776240  3309 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58f70e40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.776324  3313 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a63db40, len: 185344: Operation not permitted [1]
+W0827 04:42:57.776337  3314 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a66af80, len: 185344: Operation not permitted [1]
+W0827 04:42:57.776417  3315 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58e5ec40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.776444  3316 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58109800, len: 46336: Operation not permitted [1]
+W0827 04:42:57.776451  3317 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58ed2cc0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.776649  3318 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58316800, len: 23168: Operation not permitted [1]
+W0827 04:42:57.812515  3457 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58919a80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.812533  3458 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5892c840, len: 23168: Operation not permitted [1]
+W0827 04:42:57.812597  3460 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58063380, len: 23168: Operation not permitted [1]
+W0827 04:42:57.812603  3459 memory_location.cpp:73] Failed to get NUMA node, addr: 0x588ed280, len: 23168: Operation not permitted [1]
+W0827 04:42:57.812654  3461 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59fc0080, len: 185344: Operation not permitted [1]
+W0827 04:42:57.812734  3462 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59fed4c0, len: 185344: Operation not permitted [1]
+W0827 04:42:57.812816  3463 memory_location.cpp:73] Failed to get NUMA node, addr: 0x57ebf300, len: 23168: Operation not permitted [1]
+W0827 04:42:57.812877  3464 memory_location.cpp:73] Failed to get NUMA node, addr: 0x44a36bc0, len: 46336: Operation not permitted [1]
+W0827 04:42:57.812991  3466 memory_location.cpp:73] Failed to get NUMA node, addr: 0x44982040, len: 23168: Operation not permitted [1]
+W0827 04:42:57.812996  3465 memory_location.cpp:73] Failed to get NUMA node, addr: 0x588ba900, len: 23168: Operation not permitted [1]
+W0827 04:42:57.851508  3605 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5888adc0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.851550  3606 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58775700, len: 23168: Operation not permitted [1]
+W0827 04:42:57.851619  3607 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58aca800, len: 23168: Operation not permitted [1]
+W0827 04:42:57.851691  3608 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5886ea80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.851799  3609 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5aaacb40, len: 185344: Operation not permitted [1]
+W0827 04:42:57.851871  3611 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58ad7b00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.851959  3610 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5aad9f80, len: 185344: Operation not permitted [1]
+W0827 04:42:57.851996  3612 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58aa8840, len: 46336: Operation not permitted [1]
+W0827 04:42:57.852005  3614 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58ca01c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.852105  3613 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5879a9c0, len: 23168: Operation not permitted [1]
+[2026-08-27 04:43:00] INFO:     Started server process [1]
+[2026-08-27 04:43:00] INFO:     Waiting for application startup.
+[2026-08-27 04:43:00] Calling super().encode with {'add_special_tokens': False}
+[2026-08-27 04:43:00] Calling super().encode with {'add_special_tokens': False}
+[2026-08-27 04:43:00] INFO:     Application startup complete.
+[2026-08-27 04:43:00] INFO:     Uvicorn running on http://0.0.0.0:30000 (Press CTRL+C to quit)
+[2026-08-27 04:43:01] INFO:     174.1.60.1:55082 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:43:01] INFO:     127.0.0.1:38530 - "GET /model_info HTTP/1.1" 200 OK
+[2026-08-27 04:43:02] INFO:     174.1.60.1:55098 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:43:03] INFO:     127.0.0.1:38536 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 04:43:04] INFO:     174.1.60.1:55110 - "GET /health HTTP/1.1" 200 OK
+[2026-08-27 04:43:01] Start of pd disaggregation warmup ...
+[2026-08-27 04:43:03] Disaggregation warmup requests completed for all 1 DP ranks
+[2026-08-27 04:43:03] End of disaggregation warmup
+[2026-08-27 04:43:03] The server is fired up and ready to roll!
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/d_healthy_d_node1.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/d_healthy_d_node1.log
new file mode 100644
index 0000000..2e1c104
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/d_healthy_d_node1.log
@@ -0,0 +1,498 @@
+/usr/local/lib/python3.12/dist-packages/torchao/quantization/quant_api.py:1731: SyntaxWarning: invalid escape sequence '\.'
+  """Configuration class for applying different quantization configs to modules or parameters based on their fully qualified names (FQNs).
+/sgl-workspace/sglang/python/sglang/launch_server.py:56: UserWarning: 'python -m sglang.launch_server' is still supported, but 'sglang serve' is the recommended entrypoint.
+  Example: sglang serve --model-path <model> [options]
+  warnings.warn(
+[2026-08-27 04:35:58] KV cache is forced as chunk cache for decode server
+[2026-08-27 04:36:02] Breakable CUDA graph is incompatible with MLA attention; disabling prefill CUDA graph.
+[2026-08-27 04:36:02] Attention backend not specified. Use triton backend by default.
+[2026-08-27 04:36:05] server_args=ServerArgs(model_path='/data/hf_models/Kimi-K3', tokenizer_path='/data/hf_models/Kimi-K3', tokenizer_mode='auto', tokenizer_backend='huggingface', tokenizer_worker_num=1, detokenizer_worker_num=1, skip_tokenizer_init=False, load_format='auto', model_loader_extra_config='{}', trust_remote_code=True, context_length=None, is_embedding=False, enable_multimodal=None, revision=None, model_impl='auto', model_config_parser='auto', json_model_override_args='{}', dtype='auto', quantization=None, quantization_param_path=None, kv_cache_dtype='auto', enable_fp32_lm_head=False, modelopt_quant=None, modelopt_checkpoint_restore_path=None, modelopt_checkpoint_save_path=None, modelopt_export_path=None, quantize_and_serve=False, rl_quant_profile=None, enable_tf32_matmul=False, mem_fraction_static=0.88, max_running_requests=None, max_queued_requests=None, max_total_tokens=None, chunked_prefill_size=8192, enable_dynamic_chunking=False, max_prefill_tokens=16384, prefill_max_requests=None, schedule_policy='fcfs', enable_priority_scheduling=False, disable_priority_preemption=False, default_priority_value=None, abort_on_priority_when_disabled=False, schedule_low_priority_values_first=False, priority_scheduling_preemption_threshold=10, retraction_policy='length', schedule_conservativeness=1.0, page_size=64, swa_full_tokens_ratio=0.8, disable_hybrid_swa_memory=False, radix_eviction_policy='lru', prefill_only_disable_kv_cache=False, disable_radix_cache=True, enable_page_major_kv_layout=False, enable_unified_memory=False, disable_chunked_prefix_cache=False, disable_overlap_schedule=False, num_continuous_decode_steps=1, scheduler_recv_interval=1, enable_mixed_chunk=False, nccl_port=None, dist_timeout=3600, dist_init_addr='174.1.60.5:20000', nnodes=4, node_rank=1, tp_size=32, dcp_size=1, pp_size=1, pp_max_micro_batch_size=None, pp_async_batch_depth=0, dp_size=1, load_balance_method='round_robin', attn_cp_size=1, moe_dp_size=1, dcp_comm_backend='ag_rs', dcp_replicate_q_proj=None, dwdp_size=1, enable_prefill_cp=False, cp_strategy=None, enable_dsa_cache_layer_split=False, enable_dsa_prefill_context_parallel=False, dsa_prefill_cp_mode='round-robin-split', enable_prefill_context_parallel=False, prefill_cp_mode='in-seq-split', enable_cp_decode_attn_tp=False, enable_dp_attention=False, enable_dp_attention_local_control_broadcast=False, enable_dp_lm_head=False, enable_attn_tp_input_scattered=False, disable_attn_tp_gather=False, enable_p2p_check=False, device='cuda', base_gpu_id=0, gpu_id_step=1, random_seed=434112381, watchdog_timeout=300, soft_watchdog_timeout=None, sleep_on_idle=False, use_ray=False, custom_sigquit_handler=None, numa_node=None, gc_threshold=None, host='0.0.0.0', port=30000, fastapi_root_path='', smg_grpc_mode=False, grpc_mode=False, grpc_port=None, sidecar=None, sidecar_args=None, skip_server_warmup=False, warmups=None, enable_http2=False, ssl_keyfile=None, ssl_certfile=None, ssl_ca_certs=None, ssl_keyfile_password=None, enable_ssl_refresh=False, api_key=None, admin_api_key=None, served_model_name='kimi-k3', weight_version='default', chat_template=None, hf_chat_template_name=None, completion_template=None, file_storage_path='sglang_storage', enable_cache_report=False, reasoning_parser=None, default_chat_template_kwargs=None, strip_thinking_cache=False, enable_strict_thinking=False, tool_call_parser=None, tool_server=None, sampling_defaults='model', asr_max_buffer_seconds=60, asr_max_concurrent_sessions=32, preferred_sampling_params=None, allow_auto_truncate=False, stream_interval=1, batch_notify_size=16, stream_response_default_include_usage=False, incremental_streaming_output=False, enable_streaming_session=False, enable_session_radix_cache=False, log_level='info', log_level_http=None, log_requests=False, log_requests_level=2, log_requests_format='text', log_requests_target=None, uvicorn_access_log_exclude_prefixes=[], crash_dump_folder=None, show_time_cost=False, enable_metrics=False, smg_http_sidecar_port=None, enable_mfu_metrics=False, enable_metrics_for_all_schedulers=False, load_snapshot_publish_interval=15, tokenizer_metrics_custom_labels_header='x-custom-labels', tokenizer_metrics_allowed_custom_labels=None, extra_metric_labels=None, bucket_time_to_first_token=None, bucket_inter_token_latency=None, bucket_e2e_request_latency=None, prompt_tokens_buckets=None, generation_tokens_buckets=None, gc_warning_threshold_secs=0.0, decode_log_interval=40, enable_request_time_stats_logging=False, kv_events_config=None, enable_forward_pass_metrics=False, forward_pass_metrics_worker_id='', forward_pass_metrics_ipc_name=None, enable_trace=False, trace_modules='request', otlp_traces_endpoint='localhost:4317', export_metrics_to_file=False, export_metrics_to_file_dir=None, stat_loggers=None, constrained_json_whitespace_pattern=None, constrained_json_disable_any_whitespace=False, attention_backend='triton', decode_attention_backend=None, prefill_attention_backend=None, sampling_backend='flashinfer', grammar_backend='xgrammar', radix_cache_backend=None, mm_attention_backend=None, fp8_gemm_runner_backend='auto', fp4_gemm_runner_backend='auto', bf16_gemm_backend='auto', dsa_prefill_backend=None, dsa_decode_backend=None, dsa_paged_mqa_logits_backend='auto', dsa_topk_backend='sgl-kernel', disable_flashinfer_autotune=False, flashinfer_autotune_skip_ops=None, mamba_backend='triton', cuda_graph_config=CudaGraphConfig(decode=PhaseConfig(backend='full', max_bs=16, bs=[1, 2, 4, 8, 12, 16], tc_compiler='eager', full_prefill_max_req=None), prefill=PhaseConfig(backend='disabled', max_bs=2048, bs=[4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], tc_compiler='eager', full_prefill_max_req=None)), cuda_graph_backend_decode=None, cuda_graph_backend_prefill=None, cuda_graph_max_bs_decode=16, cuda_graph_max_bs_prefill=None, cuda_graph_bs_decode=None, cuda_graph_bs_prefill=None, cuda_graph_tc_compiler=None, disable_prefill_cuda_graph=False, disable_decode_cuda_graph=False, disable_cuda_graph=False, disable_cuda_graph_padding=False, enable_profile_cuda_graph=False, enable_cudagraph_gc=False, debug_cuda_graph=False, enable_layerwise_nvtx_marker=False, enable_nccl_nvls=False, enable_symm_mem=False, triton_attention_reduce_in_fp32=False, triton_attention_num_kv_splits=8, triton_attention_split_tile_size=None, flashinfer_mla_disable_ragged=False, enable_fused_qk_norm_rope=False, enable_precise_embedding_interpolation=False, enable_fused_moe_sum_all_reduce=False, enable_deepseek_v4_fp4_indexer=False, disable_custom_all_reduce=False, enable_mscclpp=False, enable_torch_symm_mem=False, enable_scattered_sconv=False, pre_warm_nccl=False, enable_quant_communications=False, enable_flashinfer_allreduce_fusion=False, enforce_disable_flashinfer_allreduce_fusion=False, flashinfer_allreduce_fusion_backend=None, enable_aiter_allreduce_fusion=False, enable_torch_compile=False, enable_torch_compile_debug_mode=False, torch_compile_max_bs=32, torchao_config='', speculative_algorithm=None, speculative_draft_model_path=None, speculative_draft_model_revision=None, speculative_draft_load_format=None, speculative_num_steps=None, speculative_eagle_topk=None, speculative_num_draft_tokens=None, speculative_dflash_block_size=None, speculative_dspark_block_size=None, speculative_dspark_sps_table_path=None, speculative_dspark_confidence_sts_path=None, speculative_dspark_align_verify_tokens_to_graph_tier=False, speculative_accept_threshold_single=1.0, speculative_accept_threshold_acc=1.0, speculative_use_rejection_sampling=False, speculative_token_map=None, speculative_attention_mode='prefill', speculative_draft_attention_backend=None, speculative_draft_window_size=None, speculative_moe_runner_backend='marlin', speculative_moe_a2a_backend=None, speculative_draft_model_quantization=None, speculative_skip_dp_mlp_sync=False, enable_multi_layer_eagle=False, speculative_adaptive=False, speculative_adaptive_config=None, decoupled_spec_bind_endpoint=None, decoupled_spec_connect_endpoints=None, decoupled_spec_rank=None, decoupled_spec_role='null', spec_trace_dir=None, speculative_ngram_min_bfs_breadth=1, speculative_ngram_max_bfs_breadth=10, speculative_ngram_match_type='BFS', speculative_ngram_max_trie_depth=18, speculative_ngram_capacity=10000000, speculative_ngram_external_corpus_path=None, speculative_ngram_external_sam_budget=0, speculative_ngram_external_corpus_max_tokens=10000000, ep_size=32, moe_a2a_backend='none', moe_runner_backend='marlin', flashinfer_mxfp4_moe_precision='default', deepep_mode='auto', fuseep_mode=2, deepep_dispatcher_output_dtype='auto', ep_num_redundant_experts=0, ep_dispatch_algorithm=None, init_expert_location='trivial', enable_eplb=False, eplb_algorithm='auto', eplb_rebalance_num_iterations=1000, eplb_rebalance_layers_per_chunk=None, eplb_min_rebalancing_utilization_threshold=1.0, expert_distribution_recorder_mode=None, expert_distribution_recorder_buffer_size=1000, enable_expert_distribution_metrics=False, deepep_config=None, moe_dense_tp_size=None, elastic_ep_backend=None, enable_elastic_expert_backup=False, mooncake_ib_device=None, enable_waterfill=False, ep_join_mode=None, ep_join_rank_offset=0, elastic_ep_initial_size=None, max_ep_size=None, elastic_ep_scale_timeout=600, elastic_ep_rejoin=False, disable_flashinfer_cutlass_moe_fp4_allgather=False, disable_shared_experts_fusion=False, enforce_shared_experts_fusion=False, max_mamba_cache_size=None, mamba_ssm_dtype=None, mamba_max_states_per_path=-1, enable_mamba_cache_stochastic_rounding=False, mamba_cache_philox_rounds=0, mamba_full_memory_ratio=0.21, mamba_radix_cache_strategy='extra_buffer_lazy', uses_mamba_radix_cache=False, mamba_track_interval=256, enable_int8_mamba_checkpoint=False, int8_mamba_ckpt_size=None, linear_attn_backend='triton', linear_attn_decode_backend=None, linear_attn_prefill_backend=None, linear_attn_verify_backend=None, enable_linear_replayssm=False, linear_replayssm_cache_len=16, enable_linear_replayssm_spec=False, enable_hierarchical_cache=False, hicache_ratio=2.0, hicache_size=0, hicache_write_policy='write_through', hicache_io_backend='kernel', hicache_mem_layout='page_first', hicache_storage_backend=None, hicache_storage_prefetch_policy='timeout', hicache_storage_backend_extra_config=None, enable_hisparse=False, hisparse_config=None, enable_broadcast_mm_inputs_process=False, enable_prefix_mm_cache=False, mm_enable_dp_encoder=False, mm_process_config={}, mm_processor_worker_num=0, mm_io_worker_num=0, limit_mm_data_per_request=None, enable_mm_global_cache=False, disable_fast_image_processor=False, mm_feature_transport='cpu', keep_mm_feature_on_device=False, enable_lora=None, enable_lora_overlap_loading=None, max_lora_rank=None, lora_target_modules=None, lora_paths=None, max_loaded_loras=None, max_loras_per_batch=8, lora_eviction_policy='lru', lora_backend='csgmv', max_lora_chunk_size=16, experts_shared_outer_loras=None, lora_use_virtual_experts=False, lora_strict_loading=False, lora_drain_wait_threshold=0.0, enable_two_batch_overlap=False, enable_single_batch_overlap=False, tbo_token_distribution_threshold=0.48, cpu_offload_gb=0, offload_group_size=-1, offload_num_in_group=1, offload_prefetch_step=1, offload_mode='cpu', enable_lmcache=False, lmcache_config_file=None, enable_flexkv=False, flexkv_config_file=None, kt_weight_path=None, kt_method='AMXINT4', kt_cpuinfer=None, kt_threadpool_count=2, kt_num_gpu_experts=None, kt_max_deferred_experts_per_token=None, dllm_algorithm=None, dllm_algorithm_config=None, dllm_fdfo=True, disaggregation_mode='decode', disaggregation_transfer_backend='mooncake', disaggregation_bootstrap_port=28800, disaggregation_ib_device='mlx5_0,mlx5_1,mlx5_2,mlx5_3', disaggregation_decode_enable_radix_cache=False, disaggregation_decode_enable_offload_kvcache=False, num_reserved_decode_tokens=512, disaggregation_decode_extra_slots=0, disaggregation_decode_polling_interval=1, optimistic_prefill_attempts=0, encoder_only=False, language_only=False, encoder_transfer_backend='zmq_to_tokenizer', encoder_urls=[], encoder_bootstrap_port=8997, encoder_register_urls=[], enable_adaptive_dispatch_to_encoder=False, enable_pdmux=False, pdmux_config_path=None, sm_group_num=8, custom_weight_loader=[], weight_loader_disable_mmap=False, weight_loader_prefetch_checkpoints=False, weight_loader_prefetch_num_threads=4, weight_loader_drop_cache_after_load=False, remote_instance_weight_loader_seed_instance_ip=None, remote_instance_weight_loader_seed_instance_service_port=None, remote_instance_weight_loader_send_weights_group_ports=None, remote_instance_weight_loader_backend='nccl', remote_instance_weight_loader_start_seed_via_transfer_engine=False, engine_info_bootstrap_port=6789, modelexpress_config=None, download_dir=None, model_checksum=None, delete_ckpt_after_loading=False, decrypted_config_file=None, decrypted_draft_config_file=None, checkpoint_engine_wait_weights_before_ready=False, enable_prefill_delayer=False, prefill_delayer_max_delay_passes=30, prefill_delayer_token_usage_low_watermark=None, prefill_delayer_forward_passes_buckets=None, prefill_delayer_wait_seconds_buckets=None, prefill_delayer_queue_min_ratio=None, prefill_delayer_max_delay_ms=None, min_free_slots_delay=None, enable_deterministic_inference=False, rl_on_policy_target=None, kv_canary='none', kv_canary_real_data='none', kv_canary_sweep_interval=0, enable_dynamic_batch_tokenizer=False, dynamic_batch_tokenizer_batch_size=32, dynamic_batch_tokenizer_batch_timeout=0.002, enable_tokenizer_batch_encode=False, disable_tokenizer_batch_decode=False, debug_tensor_dump_output_folder=None, debug_tensor_dump_layers=None, debug_tensor_dump_input_file=None, enable_memory_saver=False, enable_weights_cpu_backup=False, enable_draft_weights_cpu_backup=False, enable_custom_logit_processor=False, enable_return_hidden_states=False, enable_return_routed_experts=False, enable_return_indexer_topk=False, disable_outlines_disk_cache=False, enable_mis=False, weight_cache_mode='off', weight_cache_socket=None, weight_cache_timeout=1800, forward_hooks=None, msprobe_dump_config=None)
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:22 TP15 EP15] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:22 TP15 EP15] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:23 TP9 EP9] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:23 TP9 EP9] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:23 TP12 EP12] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:23 TP12 EP12] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:23 TP13 EP13] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:23 TP13 EP13] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:23 TP14 EP14] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:23 TP14 EP14] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:23 TP8 EP8] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:23 TP8 EP8] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:23 TP10 EP10] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:23 TP10 EP10] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:23 TP11 EP11] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:23 TP11 EP11] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:24 TP15 EP15] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:24 TP9 EP9] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:24 TP13 EP13] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:24 TP12 EP12] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:24 TP14 EP14] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:25 TP10 EP10] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:25 TP8 EP8] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:25 TP11 EP11] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:26.087226   242 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:26.087333   242 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:26.087350   242 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.6 port: 12001
+I0827 04:36:26.087409   242 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.6:16298
+I0827 04:36:26.087569   242 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:26.095995   236 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:26.096091   236 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:26.096107   236 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.6 port: 12001
+I0827 04:36:26.096150   236 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.6:15958
+I0827 04:36:26.096269   236 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:26.096457   240 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:26.096560   240 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:26.096577   240 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.6 port: 12001
+I0827 04:36:26.096626   240 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.6:16447
+I0827 04:36:26.096773   240 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:26.096876   241 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:26.096975   241 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:26.096992   241 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.6 port: 12001
+I0827 04:36:26.097035   241 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.6:16035
+I0827 04:36:26.097170   241 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:26.103180   241 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:26.107820   240 topology.cpp:127] Device mlx5_0 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:26.109326   239 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:26.109416   239 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:26.109428   239 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.6 port: 12001
+I0827 04:36:26.109464   239 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.6:16843
+I0827 04:36:26.109562   239 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:26.114297   242 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:26.114653   239 topology.cpp:127] Device mlx5_0 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:26.115540   237 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:26.115630   237 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:26.115645   237 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.6 port: 12001
+I0827 04:36:26.115686   237 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.6:15544
+I0827 04:36:26.115813   237 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:26.118220   240 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:26.121481   242 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:26.122418   237 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:26.126497   236 topology.cpp:127] Device mlx5_0 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:26.126574   235 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:26.126684   235 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:26.126703   235 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.6 port: 12001
+I0827 04:36:26.126758   235 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.6:15006
+I0827 04:36:26.126933   235 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:26.127116   240 topology.cpp:127] Device mlx5_2 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:26.127162   238 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:26.127270   238 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:26.127285   238 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.6 port: 12001
+I0827 04:36:26.127342   238 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.6:16140
+I0827 04:36:26.127489   238 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:26.128777   237 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:26.129559   242 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:26.130432   241 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:26.133821   238 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:26.134943   235 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:26.135799   240 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:26.136080   237 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:26.138710   241 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:26.139173   242 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:26.139978   240 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:26.140033   240 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:26.140090   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.140504   235 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:26.142936   242 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:26.142961   242 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:26.143014   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.144655   237 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:26.146834   235 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:26.147805   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:26.147867   241 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:26.148357   239 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:26.148370   237 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:26.148394   237 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:26.148444   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.149503   240 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:06
+I0827 04:36:26.149536   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.150290   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:26.150381   236 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:26.151773   242 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:06
+I0827 04:36:26.151803   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.152055   241 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:26.152087   241 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:26.152148   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.153594   235 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:26.156044   235 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:26.156075   235 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:26.156124   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.157910   238 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:26.159448   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:26.159547   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:26.160912   242 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:06
+I0827 04:36:26.160940   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.161144   241 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:06
+I0827 04:36:26.161176   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.162882   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:26.164348   235 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:06
+I0827 04:36:26.164378   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.165159   238 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:26.168772   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:26.170400   241 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:06
+I0827 04:36:26.170428   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.172459   238 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:26.174376   239 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:26.175243   238 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:26.175272   238 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:26.175325   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.178375   236 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:26.178763   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:26.178785   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:26.180411   237 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:06
+I0827 04:36:26.180462   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.180604   240 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:06
+I0827 04:36:26.180672   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.181643   239 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:26.182818   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:26.184404   238 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:06
+I0827 04:36:26.184437   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.184495   239 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:26.184537   239 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:26.184595   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.185601   236 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:26.186753   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:26.188311   242 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:06
+I0827 04:36:26.188344   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.188390   236 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:26.188416   236 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:26.188470   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.190752   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:26.192051   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:26.192406   235 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:06
+I0827 04:36:26.192444   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.193856   240 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:06
+I0827 04:36:26.193897   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.194197   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:26.195964   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:26.195997   238 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:06
+I0827 04:36:26.196039   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.196040   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:26.197574   242 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:06
+I0827 04:36:26.197599   236 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:06
+I0827 04:36:26.197629   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.197691   242 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:26 TP15 EP15] Init torch distributed begin.
+I0827 04:36:26.198544   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:26.200238   241 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:06
+I0827 04:36:26.200268   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.204000   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:26.205737   238 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:06
+I0827 04:36:26.205766   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.210784   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:26.212373   237 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:06
+I0827 04:36:26.212417   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.218751   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:26.220386   235 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:06
+I0827 04:36:26.220422   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.222854   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:26.223088   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:26.223217   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:26.223345   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:26.224500   241 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:06
+I0827 04:36:26.224536   240 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:06
+I0827 04:36:26.224612   241 transfer_engine_impl.cpp:388] installTransport, type=rdma
+I0827 04:36:26.224682   240 transfer_engine_impl.cpp:388] installTransport, type=rdma
+I0827 04:36:26.224973   236 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:06
+I0827 04:36:26.225018   236 rdma_context.cpp:205] Using SIEVE endpoint store
+[2026-08-27 04:36:26 TP14 EP14] Init torch distributed begin.
+[2026-08-27 04:36:26 TP13 EP13] Init torch distributed begin.
+I0827 04:36:26.225459   239 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:06
+I0827 04:36:26.225523   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.230858   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:26.232468   238 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:06
+I0827 04:36:26.232614   238 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:26 TP11 EP11] Init torch distributed begin.
+I0827 04:36:26.233065   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:26.234685   236 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:06
+I0827 04:36:26.234727   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.238961   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:26.240710   237 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:06
+I0827 04:36:26.240758   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.242511   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:26.244089   236 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:06
+I0827 04:36:26.244236   236 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:26 TP9 EP9] Init torch distributed begin.
+I0827 04:36:26.247059   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:26.248764   235 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:06
+I0827 04:36:26.248920   235 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:26 TP8 EP8] Init torch distributed begin.
+I0827 04:36:26.250787   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:26.252344   239 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:06
+I0827 04:36:26.252393   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.266752   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:26.268360   237 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:06
+I0827 04:36:26.268492   237 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:26 TP10 EP10] Init torch distributed begin.
+I0827 04:36:26.278649   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:26.280236   239 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:06
+I0827 04:36:26.280287   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.306850   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:26.308480   239 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:06
+I0827 04:36:26.308605   239 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:26 TP12 EP12] Init torch distributed begin.
+[2026-08-27 04:36:48 TP14 EP14] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP15 EP15] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP8 EP8] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP11 EP11] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP9 EP9] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP13 EP13] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP10 EP10] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP12 EP12] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP15 EP15] Init torch distributed ends. elapsed=22.03 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP14 EP14] Init torch distributed ends. elapsed=22.01 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP13 EP13] Init torch distributed ends. elapsed=22.01 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP12 EP12] Init torch distributed ends. elapsed=21.92 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP11 EP11] Init torch distributed ends. elapsed=22.00 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP9 EP9] Init torch distributed ends. elapsed=21.99 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP10 EP10] Init torch distributed ends. elapsed=21.96 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP8 EP8] Init torch distributed ends. elapsed=21.98 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP14 EP14] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP15 EP15] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP9 EP9] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP13 EP13] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP8 EP8] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP10 EP10] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP12 EP12] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP14 EP14] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP15 EP15] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP9 EP9] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP10 EP10] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP13 EP13] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP8 EP8] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP12 EP12] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP11 EP11] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP11 EP11] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP13 EP13] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP14 EP14] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP15 EP15] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP12 EP12] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP10 EP10] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP8 EP8] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP9 EP9] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP13 EP13] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP12 EP12] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP14 EP14] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP15 EP15] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP10 EP10] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP8 EP8] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP9 EP9] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP11 EP11] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP11 EP11] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:50 TP8 EP8] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP9 EP9] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP11 EP11] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP12 EP12] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP13 EP13] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP14 EP14] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP10 EP10] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP15 EP15] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:38:57 TP14 EP14] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:57 TP14 EP14] Load weight end. elapsed=129.27 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP9 EP9] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP9 EP9] Load weight end. elapsed=129.52 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP10 EP10] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP10 EP10] Load weight end. elapsed=129.56 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP8 EP8] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP8 EP8] Load weight end. elapsed=129.57 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP13 EP13] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP13 EP13] Load weight end. elapsed=129.58 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP11 EP11] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP11 EP11] Load weight end. elapsed=129.64 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP15 EP15] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP15 EP15] Load weight end. elapsed=129.77 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP12 EP12] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP12 EP12] Load weight end. elapsed=129.90 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:39:01 TP14 EP14] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP14 EP14] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP11 EP11] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP11 EP11] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP8 EP8] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP8 EP8] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP13 EP13] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP13 EP13] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP9 EP9] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP9 EP9] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP15 EP15] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP15 EP15] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP12 EP12] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP12 EP12] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP10 EP10] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP10 EP10] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:02 TP14 EP14] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP15 EP15] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP12 EP12] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP13 EP13] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP10 EP10] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP9 EP9] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP11 EP11] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP8 EP8] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP15 EP15] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP12 EP12] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP10 EP10] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP8 EP8] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP11 EP11] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP13 EP13] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP9 EP9] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP14 EP14] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP9 EP9] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP12 EP12] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP9 EP9] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP10 EP10] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP12 EP12] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP13 EP13] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP15 EP15] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP10 EP10] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP11 EP11] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP13 EP13] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP14 EP14] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP15 EP15] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP11 EP11] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP8 EP8] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP14 EP14] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP8 EP8] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP14 EP14] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP14 EP14] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP9 EP9] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP9 EP9] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP12 EP12] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP12 EP12] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:39:02 TP10 EP10] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP10 EP10] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP8 EP8] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP8 EP8] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:39:02 TP11 EP11] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP11 EP11] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:39:02 TP13 EP13] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP13 EP13] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP15 EP15] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP15 EP15] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:42:56 TP8 EP8] Capture target decode CUDA graph end. elapsed=233.78 s, mem usage=0.60 GB, avail mem=8.52 GB.
+[2026-08-27 04:42:56 TP13 EP13] Capture target decode CUDA graph end. elapsed=233.78 s, mem usage=0.58 GB, avail mem=8.58 GB.
+[2026-08-27 04:42:56 TP8 EP8] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP13 EP13] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP14 EP14] Capture target decode CUDA graph end. elapsed=233.81 s, mem usage=0.56 GB, avail mem=8.59 GB.
+[2026-08-27 04:42:56 TP10 EP10] Capture target decode CUDA graph end. elapsed=233.79 s, mem usage=0.56 GB, avail mem=8.59 GB.
+[2026-08-27 04:42:56 TP14 EP14] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP12 EP12] Capture target decode CUDA graph end. elapsed=233.80 s, mem usage=0.60 GB, avail mem=8.52 GB.
+[2026-08-27 04:42:56 TP10 EP10] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP12 EP12] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP11 EP11] Capture target decode CUDA graph end. elapsed=233.80 s, mem usage=0.62 GB, avail mem=8.51 GB.
+[2026-08-27 04:42:56 TP11 EP11] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP9 EP9] Capture target decode CUDA graph end. elapsed=233.91 s, mem usage=0.58 GB, avail mem=8.58 GB.
+[2026-08-27 04:42:56 TP9 EP9] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP15 EP15] Capture target decode CUDA graph end. elapsed=233.94 s, mem usage=0.62 GB, avail mem=8.51 GB.
+[2026-08-27 04:42:56 TP15 EP15] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+W0827 04:42:57.337543  2432 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5972aec0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.337602  2433 memory_location.cpp:73] Failed to get NUMA node, addr: 0x594a1440, len: 23168: Operation not permitted [1]
+W0827 04:42:57.337656  2434 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59f51680, len: 23168: Operation not permitted [1]
+W0827 04:42:57.337674  2435 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59ebed40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.337788  2436 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5b797200, len: 185344: Operation not permitted [1]
+W0827 04:42:57.337911  2437 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5b7c4640, len: 185344: Operation not permitted [1]
+W0827 04:42:57.337891  2438 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59639800, len: 23168: Operation not permitted [1]
+W0827 04:42:57.337962  2439 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59435700, len: 23168: Operation not permitted [1]
+W0827 04:42:57.338008  2440 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59660c80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.338057  2441 memory_location.cpp:73] Failed to get NUMA node, addr: 0x594c9400, len: 46336: Operation not permitted [1]
+W0827 04:42:57.338071  2442 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59bb1f00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.338107  2443 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59f5fc80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.338161  2445 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5953b700, len: 23168: Operation not permitted [1]
+W0827 04:42:57.338165  2444 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59812080, len: 23168: Operation not permitted [1]
+W0827 04:42:57.338204  2446 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5baddac0, len: 185344: Operation not permitted [1]
+W0827 04:42:57.338325  2448 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59994940, len: 23168: Operation not permitted [1]
+W0827 04:42:57.338340  2447 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5bb0af00, len: 185344: Operation not permitted [1]
+W0827 04:42:57.338538  2449 memory_location.cpp:73] Failed to get NUMA node, addr: 0x45ba2780, len: 46336: Operation not permitted [1]
+W0827 04:42:57.338596  2450 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59ac5f00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.338649  2451 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59984000, len: 23168: Operation not permitted [1]
+W0827 04:42:57.381100  2729 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59997640, len: 23168: Operation not permitted [1]
+W0827 04:42:57.381177  2728 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59a855c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.381280  2730 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59953e40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.381347  2732 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5b87eac0, len: 185344: Operation not permitted [1]
+W0827 04:42:57.381466  2731 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59971000, len: 23168: Operation not permitted [1]
+W0827 04:42:57.381542  2733 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5b8abf00, len: 185344: Operation not permitted [1]
+W0827 04:42:57.381620  2734 memory_location.cpp:73] Failed to get NUMA node, addr: 0x599b8540, len: 23168: Operation not permitted [1]
+W0827 04:42:57.381628  2735 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5927e3c0, len: 46336: Operation not permitted [1]
+W0827 04:42:57.381863  2736 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59900b40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.381878  2737 memory_location.cpp:73] Failed to get NUMA node, addr: 0x598b4f80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.421048  2876 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58bb5680, len: 23168: Operation not permitted [1]
+W0827 04:42:57.421196  2878 memory_location.cpp:73] Failed to get NUMA node, addr: 0x592f7a00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.421193  2877 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59321cc0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.421262  2881 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ab93340, len: 185344: Operation not permitted [1]
+W0827 04:42:57.421321  2879 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5929c2c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.421382  2880 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ab65f00, len: 185344: Operation not permitted [1]
+W0827 04:42:57.421423  2882 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58aa4dc0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.421440  2883 memory_location.cpp:73] Failed to get NUMA node, addr: 0x585179c0, len: 46336: Operation not permitted [1]
+W0827 04:42:57.421434  2884 memory_location.cpp:73] Failed to get NUMA node, addr: 0x588b0240, len: 23168: Operation not permitted [1]
+W0827 04:42:57.421594  2885 memory_location.cpp:73] Failed to get NUMA node, addr: 0x588d0800, len: 23168: Operation not permitted [1]
+W0827 04:42:57.423333  2900 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58bc9040, len: 23168: Operation not permitted [1]
+W0827 04:42:57.423418  2903 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58a57e40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.423974  2913 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ae07a80, len: 185344: Operation not permitted [1]
+W0827 04:42:57.424072  2911 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5adda640, len: 185344: Operation not permitted [1]
+W0827 04:42:57.424093  2908 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58f8c900, len: 23168: Operation not permitted [1]
+W0827 04:42:57.424189  2924 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58d795c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.424269  2906 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58e657c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.424296  2915 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58c46ac0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.424311  2922 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58e39680, len: 23168: Operation not permitted [1]
+W0827 04:42:57.424809  2930 memory_location.cpp:73] Failed to get NUMA node, addr: 0x591a7f40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.424866  2932 memory_location.cpp:73] Failed to get NUMA node, addr: 0x593dee00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.425096  2934 memory_location.cpp:73] Failed to get NUMA node, addr: 0x596af5c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.425186  2935 memory_location.cpp:73] Failed to get NUMA node, addr: 0x595b2ac0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.425271  2938 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5b4f5f80, len: 185344: Operation not permitted [1]
+W0827 04:42:57.425292  2937 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5b4c8b40, len: 185344: Operation not permitted [1]
+W0827 04:42:57.425374  2939 memory_location.cpp:73] Failed to get NUMA node, addr: 0x594e07c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.425433  2942 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58ec6dc0, len: 46336: Operation not permitted [1]
+W0827 04:42:57.425568  2944 memory_location.cpp:73] Failed to get NUMA node, addr: 0x596a8180, len: 23168: Operation not permitted [1]
+W0827 04:42:57.425750  2919 memory_location.cpp:73] Failed to get NUMA node, addr: 0x45850900, len: 46336: Operation not permitted [1]
+W0827 04:42:57.425791  2946 memory_location.cpp:73] Failed to get NUMA node, addr: 0x592c6f00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.427150  2998 memory_location.cpp:73] Failed to get NUMA node, addr: 0x599c15c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.427410  2989 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59750440, len: 23168: Operation not permitted [1]
+W0827 04:42:57.427572  3010 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5bad0cc0, len: 185344: Operation not permitted [1]
+W0827 04:42:57.427641  3014 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a39cd80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.427647  3024 memory_location.cpp:73] Failed to get NUMA node, addr: 0x464ff140, len: 23168: Operation not permitted [1]
+W0827 04:42:57.427655  3002 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a377380, len: 23168: Operation not permitted [1]
+W0827 04:42:57.427663  2993 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59876080, len: 23168: Operation not permitted [1]
+W0827 04:42:57.427681  3007 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5baa3880, len: 185344: Operation not permitted [1]
+W0827 04:42:57.427703  3025 memory_location.cpp:73] Failed to get NUMA node, addr: 0x463cd480, len: 23168: Operation not permitted [1]
+W0827 04:42:57.427706  3018 memory_location.cpp:73] Failed to get NUMA node, addr: 0x464e6b00, len: 46336: Operation not permitted [1]
+W0827 04:42:57.586287  3470 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a4fe500, len: 23168: Operation not permitted [1]
+W0827 04:42:57.586316  3471 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a8129c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.586287  3468 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a530a00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.586329  3469 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a72a280, len: 23168: Operation not permitted [1]
+W0827 04:42:57.586364  3472 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5c756140, len: 185344: Operation not permitted [1]
+W0827 04:42:57.586422  3474 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a287d00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.586429  3473 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5c783580, len: 185344: Operation not permitted [1]
+W0827 04:42:57.586477  3476 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a731080, len: 23168: Operation not permitted [1]
+W0827 04:42:57.586475  3475 memory_location.cpp:73] Failed to get NUMA node, addr: 0x46a69900, len: 46336: Operation not permitted [1]
+W0827 04:42:57.586767  3477 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a2cd580, len: 23168: Operation not permitted [1]
+[2026-08-27 04:42:59] Dummy health check server started in background thread at 0.0.0.0:30000
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/d_healthy_d_node2.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/d_healthy_d_node2.log
new file mode 100644
index 0000000..97f23a1
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/d_healthy_d_node2.log
@@ -0,0 +1,498 @@
+/usr/local/lib/python3.12/dist-packages/torchao/quantization/quant_api.py:1731: SyntaxWarning: invalid escape sequence '\.'
+  """Configuration class for applying different quantization configs to modules or parameters based on their fully qualified names (FQNs).
+/sgl-workspace/sglang/python/sglang/launch_server.py:56: UserWarning: 'python -m sglang.launch_server' is still supported, but 'sglang serve' is the recommended entrypoint.
+  Example: sglang serve --model-path <model> [options]
+  warnings.warn(
+[2026-08-27 04:35:59] KV cache is forced as chunk cache for decode server
+[2026-08-27 04:36:03] Breakable CUDA graph is incompatible with MLA attention; disabling prefill CUDA graph.
+[2026-08-27 04:36:03] Attention backend not specified. Use triton backend by default.
+[2026-08-27 04:36:06] server_args=ServerArgs(model_path='/data/hf_models/Kimi-K3', tokenizer_path='/data/hf_models/Kimi-K3', tokenizer_mode='auto', tokenizer_backend='huggingface', tokenizer_worker_num=1, detokenizer_worker_num=1, skip_tokenizer_init=False, load_format='auto', model_loader_extra_config='{}', trust_remote_code=True, context_length=None, is_embedding=False, enable_multimodal=None, revision=None, model_impl='auto', model_config_parser='auto', json_model_override_args='{}', dtype='auto', quantization=None, quantization_param_path=None, kv_cache_dtype='auto', enable_fp32_lm_head=False, modelopt_quant=None, modelopt_checkpoint_restore_path=None, modelopt_checkpoint_save_path=None, modelopt_export_path=None, quantize_and_serve=False, rl_quant_profile=None, enable_tf32_matmul=False, mem_fraction_static=0.88, max_running_requests=None, max_queued_requests=None, max_total_tokens=None, chunked_prefill_size=8192, enable_dynamic_chunking=False, max_prefill_tokens=16384, prefill_max_requests=None, schedule_policy='fcfs', enable_priority_scheduling=False, disable_priority_preemption=False, default_priority_value=None, abort_on_priority_when_disabled=False, schedule_low_priority_values_first=False, priority_scheduling_preemption_threshold=10, retraction_policy='length', schedule_conservativeness=1.0, page_size=64, swa_full_tokens_ratio=0.8, disable_hybrid_swa_memory=False, radix_eviction_policy='lru', prefill_only_disable_kv_cache=False, disable_radix_cache=True, enable_page_major_kv_layout=False, enable_unified_memory=False, disable_chunked_prefix_cache=False, disable_overlap_schedule=False, num_continuous_decode_steps=1, scheduler_recv_interval=1, enable_mixed_chunk=False, nccl_port=None, dist_timeout=3600, dist_init_addr='174.1.60.5:20000', nnodes=4, node_rank=2, tp_size=32, dcp_size=1, pp_size=1, pp_max_micro_batch_size=None, pp_async_batch_depth=0, dp_size=1, load_balance_method='round_robin', attn_cp_size=1, moe_dp_size=1, dcp_comm_backend='ag_rs', dcp_replicate_q_proj=None, dwdp_size=1, enable_prefill_cp=False, cp_strategy=None, enable_dsa_cache_layer_split=False, enable_dsa_prefill_context_parallel=False, dsa_prefill_cp_mode='round-robin-split', enable_prefill_context_parallel=False, prefill_cp_mode='in-seq-split', enable_cp_decode_attn_tp=False, enable_dp_attention=False, enable_dp_attention_local_control_broadcast=False, enable_dp_lm_head=False, enable_attn_tp_input_scattered=False, disable_attn_tp_gather=False, enable_p2p_check=False, device='cuda', base_gpu_id=0, gpu_id_step=1, random_seed=212084647, watchdog_timeout=300, soft_watchdog_timeout=None, sleep_on_idle=False, use_ray=False, custom_sigquit_handler=None, numa_node=None, gc_threshold=None, host='0.0.0.0', port=30000, fastapi_root_path='', smg_grpc_mode=False, grpc_mode=False, grpc_port=None, sidecar=None, sidecar_args=None, skip_server_warmup=False, warmups=None, enable_http2=False, ssl_keyfile=None, ssl_certfile=None, ssl_ca_certs=None, ssl_keyfile_password=None, enable_ssl_refresh=False, api_key=None, admin_api_key=None, served_model_name='kimi-k3', weight_version='default', chat_template=None, hf_chat_template_name=None, completion_template=None, file_storage_path='sglang_storage', enable_cache_report=False, reasoning_parser=None, default_chat_template_kwargs=None, strip_thinking_cache=False, enable_strict_thinking=False, tool_call_parser=None, tool_server=None, sampling_defaults='model', asr_max_buffer_seconds=60, asr_max_concurrent_sessions=32, preferred_sampling_params=None, allow_auto_truncate=False, stream_interval=1, batch_notify_size=16, stream_response_default_include_usage=False, incremental_streaming_output=False, enable_streaming_session=False, enable_session_radix_cache=False, log_level='info', log_level_http=None, log_requests=False, log_requests_level=2, log_requests_format='text', log_requests_target=None, uvicorn_access_log_exclude_prefixes=[], crash_dump_folder=None, show_time_cost=False, enable_metrics=False, smg_http_sidecar_port=None, enable_mfu_metrics=False, enable_metrics_for_all_schedulers=False, load_snapshot_publish_interval=15, tokenizer_metrics_custom_labels_header='x-custom-labels', tokenizer_metrics_allowed_custom_labels=None, extra_metric_labels=None, bucket_time_to_first_token=None, bucket_inter_token_latency=None, bucket_e2e_request_latency=None, prompt_tokens_buckets=None, generation_tokens_buckets=None, gc_warning_threshold_secs=0.0, decode_log_interval=40, enable_request_time_stats_logging=False, kv_events_config=None, enable_forward_pass_metrics=False, forward_pass_metrics_worker_id='', forward_pass_metrics_ipc_name=None, enable_trace=False, trace_modules='request', otlp_traces_endpoint='localhost:4317', export_metrics_to_file=False, export_metrics_to_file_dir=None, stat_loggers=None, constrained_json_whitespace_pattern=None, constrained_json_disable_any_whitespace=False, attention_backend='triton', decode_attention_backend=None, prefill_attention_backend=None, sampling_backend='flashinfer', grammar_backend='xgrammar', radix_cache_backend=None, mm_attention_backend=None, fp8_gemm_runner_backend='auto', fp4_gemm_runner_backend='auto', bf16_gemm_backend='auto', dsa_prefill_backend=None, dsa_decode_backend=None, dsa_paged_mqa_logits_backend='auto', dsa_topk_backend='sgl-kernel', disable_flashinfer_autotune=False, flashinfer_autotune_skip_ops=None, mamba_backend='triton', cuda_graph_config=CudaGraphConfig(decode=PhaseConfig(backend='full', max_bs=16, bs=[1, 2, 4, 8, 12, 16], tc_compiler='eager', full_prefill_max_req=None), prefill=PhaseConfig(backend='disabled', max_bs=2048, bs=[4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], tc_compiler='eager', full_prefill_max_req=None)), cuda_graph_backend_decode=None, cuda_graph_backend_prefill=None, cuda_graph_max_bs_decode=16, cuda_graph_max_bs_prefill=None, cuda_graph_bs_decode=None, cuda_graph_bs_prefill=None, cuda_graph_tc_compiler=None, disable_prefill_cuda_graph=False, disable_decode_cuda_graph=False, disable_cuda_graph=False, disable_cuda_graph_padding=False, enable_profile_cuda_graph=False, enable_cudagraph_gc=False, debug_cuda_graph=False, enable_layerwise_nvtx_marker=False, enable_nccl_nvls=False, enable_symm_mem=False, triton_attention_reduce_in_fp32=False, triton_attention_num_kv_splits=8, triton_attention_split_tile_size=None, flashinfer_mla_disable_ragged=False, enable_fused_qk_norm_rope=False, enable_precise_embedding_interpolation=False, enable_fused_moe_sum_all_reduce=False, enable_deepseek_v4_fp4_indexer=False, disable_custom_all_reduce=False, enable_mscclpp=False, enable_torch_symm_mem=False, enable_scattered_sconv=False, pre_warm_nccl=False, enable_quant_communications=False, enable_flashinfer_allreduce_fusion=False, enforce_disable_flashinfer_allreduce_fusion=False, flashinfer_allreduce_fusion_backend=None, enable_aiter_allreduce_fusion=False, enable_torch_compile=False, enable_torch_compile_debug_mode=False, torch_compile_max_bs=32, torchao_config='', speculative_algorithm=None, speculative_draft_model_path=None, speculative_draft_model_revision=None, speculative_draft_load_format=None, speculative_num_steps=None, speculative_eagle_topk=None, speculative_num_draft_tokens=None, speculative_dflash_block_size=None, speculative_dspark_block_size=None, speculative_dspark_sps_table_path=None, speculative_dspark_confidence_sts_path=None, speculative_dspark_align_verify_tokens_to_graph_tier=False, speculative_accept_threshold_single=1.0, speculative_accept_threshold_acc=1.0, speculative_use_rejection_sampling=False, speculative_token_map=None, speculative_attention_mode='prefill', speculative_draft_attention_backend=None, speculative_draft_window_size=None, speculative_moe_runner_backend='marlin', speculative_moe_a2a_backend=None, speculative_draft_model_quantization=None, speculative_skip_dp_mlp_sync=False, enable_multi_layer_eagle=False, speculative_adaptive=False, speculative_adaptive_config=None, decoupled_spec_bind_endpoint=None, decoupled_spec_connect_endpoints=None, decoupled_spec_rank=None, decoupled_spec_role='null', spec_trace_dir=None, speculative_ngram_min_bfs_breadth=1, speculative_ngram_max_bfs_breadth=10, speculative_ngram_match_type='BFS', speculative_ngram_max_trie_depth=18, speculative_ngram_capacity=10000000, speculative_ngram_external_corpus_path=None, speculative_ngram_external_sam_budget=0, speculative_ngram_external_corpus_max_tokens=10000000, ep_size=32, moe_a2a_backend='none', moe_runner_backend='marlin', flashinfer_mxfp4_moe_precision='default', deepep_mode='auto', fuseep_mode=2, deepep_dispatcher_output_dtype='auto', ep_num_redundant_experts=0, ep_dispatch_algorithm=None, init_expert_location='trivial', enable_eplb=False, eplb_algorithm='auto', eplb_rebalance_num_iterations=1000, eplb_rebalance_layers_per_chunk=None, eplb_min_rebalancing_utilization_threshold=1.0, expert_distribution_recorder_mode=None, expert_distribution_recorder_buffer_size=1000, enable_expert_distribution_metrics=False, deepep_config=None, moe_dense_tp_size=None, elastic_ep_backend=None, enable_elastic_expert_backup=False, mooncake_ib_device=None, enable_waterfill=False, ep_join_mode=None, ep_join_rank_offset=0, elastic_ep_initial_size=None, max_ep_size=None, elastic_ep_scale_timeout=600, elastic_ep_rejoin=False, disable_flashinfer_cutlass_moe_fp4_allgather=False, disable_shared_experts_fusion=False, enforce_shared_experts_fusion=False, max_mamba_cache_size=None, mamba_ssm_dtype=None, mamba_max_states_per_path=-1, enable_mamba_cache_stochastic_rounding=False, mamba_cache_philox_rounds=0, mamba_full_memory_ratio=0.21, mamba_radix_cache_strategy='extra_buffer_lazy', uses_mamba_radix_cache=False, mamba_track_interval=256, enable_int8_mamba_checkpoint=False, int8_mamba_ckpt_size=None, linear_attn_backend='triton', linear_attn_decode_backend=None, linear_attn_prefill_backend=None, linear_attn_verify_backend=None, enable_linear_replayssm=False, linear_replayssm_cache_len=16, enable_linear_replayssm_spec=False, enable_hierarchical_cache=False, hicache_ratio=2.0, hicache_size=0, hicache_write_policy='write_through', hicache_io_backend='kernel', hicache_mem_layout='page_first', hicache_storage_backend=None, hicache_storage_prefetch_policy='timeout', hicache_storage_backend_extra_config=None, enable_hisparse=False, hisparse_config=None, enable_broadcast_mm_inputs_process=False, enable_prefix_mm_cache=False, mm_enable_dp_encoder=False, mm_process_config={}, mm_processor_worker_num=0, mm_io_worker_num=0, limit_mm_data_per_request=None, enable_mm_global_cache=False, disable_fast_image_processor=False, mm_feature_transport='cpu', keep_mm_feature_on_device=False, enable_lora=None, enable_lora_overlap_loading=None, max_lora_rank=None, lora_target_modules=None, lora_paths=None, max_loaded_loras=None, max_loras_per_batch=8, lora_eviction_policy='lru', lora_backend='csgmv', max_lora_chunk_size=16, experts_shared_outer_loras=None, lora_use_virtual_experts=False, lora_strict_loading=False, lora_drain_wait_threshold=0.0, enable_two_batch_overlap=False, enable_single_batch_overlap=False, tbo_token_distribution_threshold=0.48, cpu_offload_gb=0, offload_group_size=-1, offload_num_in_group=1, offload_prefetch_step=1, offload_mode='cpu', enable_lmcache=False, lmcache_config_file=None, enable_flexkv=False, flexkv_config_file=None, kt_weight_path=None, kt_method='AMXINT4', kt_cpuinfer=None, kt_threadpool_count=2, kt_num_gpu_experts=None, kt_max_deferred_experts_per_token=None, dllm_algorithm=None, dllm_algorithm_config=None, dllm_fdfo=True, disaggregation_mode='decode', disaggregation_transfer_backend='mooncake', disaggregation_bootstrap_port=28800, disaggregation_ib_device='mlx5_0,mlx5_1,mlx5_2,mlx5_3', disaggregation_decode_enable_radix_cache=False, disaggregation_decode_enable_offload_kvcache=False, num_reserved_decode_tokens=512, disaggregation_decode_extra_slots=0, disaggregation_decode_polling_interval=1, optimistic_prefill_attempts=0, encoder_only=False, language_only=False, encoder_transfer_backend='zmq_to_tokenizer', encoder_urls=[], encoder_bootstrap_port=8997, encoder_register_urls=[], enable_adaptive_dispatch_to_encoder=False, enable_pdmux=False, pdmux_config_path=None, sm_group_num=8, custom_weight_loader=[], weight_loader_disable_mmap=False, weight_loader_prefetch_checkpoints=False, weight_loader_prefetch_num_threads=4, weight_loader_drop_cache_after_load=False, remote_instance_weight_loader_seed_instance_ip=None, remote_instance_weight_loader_seed_instance_service_port=None, remote_instance_weight_loader_send_weights_group_ports=None, remote_instance_weight_loader_backend='nccl', remote_instance_weight_loader_start_seed_via_transfer_engine=False, engine_info_bootstrap_port=6789, modelexpress_config=None, download_dir=None, model_checksum=None, delete_ckpt_after_loading=False, decrypted_config_file=None, decrypted_draft_config_file=None, checkpoint_engine_wait_weights_before_ready=False, enable_prefill_delayer=False, prefill_delayer_max_delay_passes=30, prefill_delayer_token_usage_low_watermark=None, prefill_delayer_forward_passes_buckets=None, prefill_delayer_wait_seconds_buckets=None, prefill_delayer_queue_min_ratio=None, prefill_delayer_max_delay_ms=None, min_free_slots_delay=None, enable_deterministic_inference=False, rl_on_policy_target=None, kv_canary='none', kv_canary_real_data='none', kv_canary_sweep_interval=0, enable_dynamic_batch_tokenizer=False, dynamic_batch_tokenizer_batch_size=32, dynamic_batch_tokenizer_batch_timeout=0.002, enable_tokenizer_batch_encode=False, disable_tokenizer_batch_decode=False, debug_tensor_dump_output_folder=None, debug_tensor_dump_layers=None, debug_tensor_dump_input_file=None, enable_memory_saver=False, enable_weights_cpu_backup=False, enable_draft_weights_cpu_backup=False, enable_custom_logit_processor=False, enable_return_hidden_states=False, enable_return_routed_experts=False, enable_return_indexer_topk=False, disable_outlines_disk_cache=False, enable_mis=False, weight_cache_mode='off', weight_cache_socket=None, weight_cache_timeout=1800, forward_hooks=None, msprobe_dump_config=None)
+[2026-08-27 04:36:06] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:06] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:06] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:06] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:06] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:06] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:06] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:06] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:24 TP18 EP18] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:24 TP18 EP18] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:24 TP21 EP21] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:24 TP21 EP21] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:24 TP20 EP20] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:24 TP20 EP20] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:25 TP19 EP19] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:25 TP19 EP19] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:25 TP22 EP22] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:25 TP22 EP22] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:25 TP23 EP23] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:25 TP23 EP23] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:26 TP17 EP17] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:26 TP17 EP17] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:26 TP16 EP16] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:26 TP16 EP16] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:27 TP19 EP19] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:27 TP21 EP21] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:27 TP18 EP18] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:27 TP20 EP20] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:27 TP23 EP23] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:27 TP22 EP22] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:27 TP17 EP17] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:27 TP16 EP16] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:29.187384   239 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:29.187532   239 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:29.187558   239 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.7 port: 12001
+I0827 04:36:29.187624   239 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.7:16248
+I0827 04:36:29.187814   239 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:29.191661   235 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:29.191802   235 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:29.191823   235 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.7 port: 12001
+I0827 04:36:29.191901   235 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.7:16408
+I0827 04:36:29.192099   235 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:29.195215   242 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:29.195319   242 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:29.195333   242 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.7 port: 12001
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:29.195333   238 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:29.195374   242 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.7:15148
+I0827 04:36:29.195490   238 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:29.195513   238 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.7 port: 12001
+I0827 04:36:29.195519   242 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:29.195581   238 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.7:16909
+I0827 04:36:29.195815   238 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:29.198207   236 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:29.198346   236 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:29.198370   236 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.7 port: 12001
+I0827 04:36:29.198427   236 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.7:16738
+I0827 04:36:29.198611   236 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:29.200070   242 topology.cpp:127] Device mlx5_0 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:29.201086   237 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:29.201221   237 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:29.201239   237 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.7 port: 12001
+I0827 04:36:29.201300   237 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.7:16522
+I0827 04:36:29.201495   237 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:29.201511   240 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:29.201632   240 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:29.201648   240 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.7 port: 12001
+I0827 04:36:29.201701   240 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.7:16431
+I0827 04:36:29.201871   240 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:29.204236   241 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:29.204355   242 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:29.204380   241 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:29.204393   241 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.7 port: 12001
+I0827 04:36:29.204454   241 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.7:16508
+I0827 04:36:29.204633   241 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:29.204896   238 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:29.207675   237 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:29.208006   240 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:29.208549   242 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:29.212560   242 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:29.214218   242 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:29.214243   242 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:29.214290   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.214713   240 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:29.214757   237 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:29.219102   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:29.219870   239 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:29.220057   235 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:29.220343   242 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:07
+I0827 04:36:29.220369   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.222033   240 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:29.222069   237 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:29.223913   236 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:29.228577   235 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:29.228634   239 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:29.230391   237 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:29.230446   240 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:29.232455   236 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:29.233546   237 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:29.233578   237 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:29.233631   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.234635   240 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:29.234665   240 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:29.234716   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.235774   238 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:29.235827   241 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:29.237140   235 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:29.241117   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:29.241130   236 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:29.242069   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:29.242725   237 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:07
+I0827 04:36:29.242758   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.243661   240 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:07
+I0827 04:36:29.243691   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.245620   235 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:29.247138   238 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:29.247233   241 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:29.248172   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:29.248514   235 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:29.248550   235 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:29.248612   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.248719   236 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:29.249847   242 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:07
+I0827 04:36:29.249886   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.251354   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:29.251672   236 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:29.251719   236 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:29.251781   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.252889   240 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:07
+I0827 04:36:29.252921   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.255584   239 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:29.256263   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:29.257805   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:29.257980   235 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:07
+I0827 04:36:29.258016   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.258512   238 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:29.258567   241 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:29.259447   242 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:07
+I0827 04:36:29.259477   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.259642   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:29.261375   236 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:07
+I0827 04:36:29.261410   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.261518   238 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:29.261557   238 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:29.261615   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.261615   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:29.262753   239 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:29.263202   240 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:07
+I0827 04:36:29.263229   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.265475   239 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:29.265504   239 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:29.265558   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.269449   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:29.271157   236 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:07
+I0827 04:36:29.271186   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.271761   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:29.272920   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:29.273422   237 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:07
+I0827 04:36:29.273456   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.274566   239 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:07
+I0827 04:36:29.274600   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.281176   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:29.282341   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:29.282708   237 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:07
+I0827 04:36:29.282739   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.284137   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:29.284178   239 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:07
+I0827 04:36:29.284219   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.285904   235 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:07
+I0827 04:36:29.285953   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.288185   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:29.289810   242 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:07
+I0827 04:36:29.289940   242 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:29 TP23 EP23] Init torch distributed begin.
+I0827 04:36:29.291602   241 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:29.291882   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:29.291936   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:29.293576   240 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:07
+I0827 04:36:29.293699   240 transfer_engine_impl.cpp:388] installTransport, type=rdma
+I0827 04:36:29.293921   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+[2026-08-27 04:36:29 TP21 EP21] Init torch distributed begin.
+I0827 04:36:29.294121   238 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:07
+I0827 04:36:29.294169   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.294241   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:29.295475   239 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:07
+I0827 04:36:29.295507   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.295553   241 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:29.295591   241 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:29.295648   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.296002   235 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:07
+I0827 04:36:29.296041   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.296232   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:29.298113   236 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:07
+I0827 04:36:29.298154   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.303915   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:29.305596   235 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:07
+I0827 04:36:29.305743   235 transfer_engine_impl.cpp:388] installTransport, type=rdma
+I0827 04:36:29.306217   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+[2026-08-27 04:36:29 TP16 EP16] Init torch distributed begin.
+I0827 04:36:29.307994   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:29.308105   236 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:07
+I0827 04:36:29.308277   236 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:29 TP17 EP17] Init torch distributed begin.
+I0827 04:36:29.310081   237 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:07
+I0827 04:36:29.310261   237 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:29 TP18 EP18] Init torch distributed begin.
+I0827 04:36:29.320421   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:29.321812   239 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:07
+I0827 04:36:29.321919   239 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:29 TP20 EP20] Init torch distributed begin.
+I0827 04:36:29.323829   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:29.323894   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:29.325821   238 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:07
+I0827 04:36:29.325820   241 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:07
+I0827 04:36:29.325871   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.325872   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.360142   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:29.360253   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:29.362092   241 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:07
+I0827 04:36:29.362147   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.362201   238 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:07
+I0827 04:36:29.362243   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.392091   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:29.392130   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:29.394111   238 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:07
+I0827 04:36:29.394127   241 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:07
+I0827 04:36:29.394166   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.394286   238 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:29 TP19 EP19] Init torch distributed begin.
+I0827 04:36:29.420166   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:29.421864   241 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:07
+I0827 04:36:29.421993   241 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:29 TP22 EP22] Init torch distributed begin.
+[2026-08-27 04:36:48 TP23 EP23] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP22 EP22] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP21 EP21] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP20 EP20] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP19 EP19] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP18 EP18] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP17 EP17] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP16 EP16] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP22 EP22] Init torch distributed ends. elapsed=18.81 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP23 EP23] Init torch distributed ends. elapsed=18.94 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP21 EP21] Init torch distributed ends. elapsed=18.93 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP20 EP20] Init torch distributed ends. elapsed=18.91 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP19 EP19] Init torch distributed ends. elapsed=18.83 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP18 EP18] Init torch distributed ends. elapsed=18.92 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP17 EP17] Init torch distributed ends. elapsed=18.92 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP16 EP16] Init torch distributed ends. elapsed=18.92 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP21 EP21] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP16 EP16] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP22 EP22] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP20 EP20] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP23 EP23] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP18 EP18] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP19 EP19] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP22 EP22] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP20 EP20] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP21 EP21] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP16 EP16] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP23 EP23] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP19 EP19] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP18 EP18] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP17 EP17] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP17 EP17] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP20 EP20] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP22 EP22] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP18 EP18] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP23 EP23] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP19 EP19] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP21 EP21] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP20 EP20] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP16 EP16] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP18 EP18] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP22 EP22] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP19 EP19] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP21 EP21] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP23 EP23] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP16 EP16] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP17 EP17] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP17 EP17] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:50 TP23 EP23] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP22 EP22] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP21 EP21] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP19 EP19] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP18 EP18] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP20 EP20] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP16 EP16] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP17 EP17] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:38:53 TP16 EP16] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:53 TP16 EP16] Load weight end. elapsed=124.66 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:53 TP17 EP17] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:53 TP17 EP17] Load weight end. elapsed=124.69 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:53 TP23 EP23] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:53 TP23 EP23] Load weight end. elapsed=124.71 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:53 TP21 EP21] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:53 TP21 EP21] Load weight end. elapsed=124.74 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:53 TP18 EP18] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:53 TP18 EP18] Load weight end. elapsed=124.82 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:53 TP20 EP20] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:53 TP20 EP20] Load weight end. elapsed=124.82 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:53 TP22 EP22] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:53 TP22 EP22] Load weight end. elapsed=124.92 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:53 TP19 EP19] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:53 TP19 EP19] Load weight end. elapsed=125.03 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:39:01 TP17 EP17] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP17 EP17] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP23 EP23] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP23 EP23] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP16 EP16] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP16 EP16] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP18 EP18] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP18 EP18] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP22 EP22] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP22 EP22] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP21 EP21] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP21 EP21] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP20 EP20] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP20 EP20] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP19 EP19] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP19 EP19] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:02 TP23 EP23] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP18 EP18] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP21 EP21] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP22 EP22] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP20 EP20] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP17 EP17] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP19 EP19] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP16 EP16] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP21 EP21] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP22 EP22] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP23 EP23] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP17 EP17] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP16 EP16] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP20 EP20] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP18 EP18] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP19 EP19] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP21 EP21] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP23 EP23] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP22 EP22] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP17 EP17] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP23 EP23] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP21 EP21] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP20 EP20] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP16 EP16] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP22 EP22] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP17 EP17] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP20 EP20] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP16 EP16] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP18 EP18] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP19 EP19] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP18 EP18] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP19 EP19] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP23 EP23] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP23 EP23] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:39:02 TP17 EP17] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP17 EP17] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP16 EP16] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP16 EP16] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:39:02 TP21 EP21] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP21 EP21] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP22 EP22] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP22 EP22] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP20 EP20] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP20 EP20] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:39:02 TP18 EP18] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP18 EP18] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP19 EP19] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP19 EP19] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:42:56 TP19 EP19] Capture target decode CUDA graph end. elapsed=233.80 s, mem usage=0.62 GB, avail mem=8.51 GB.
+[2026-08-27 04:42:56 TP19 EP19] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP16 EP16] Capture target decode CUDA graph end. elapsed=233.84 s, mem usage=0.60 GB, avail mem=8.52 GB.
+[2026-08-27 04:42:56 TP16 EP16] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP22 EP22] Capture target decode CUDA graph end. elapsed=233.84 s, mem usage=0.56 GB, avail mem=8.59 GB.
+[2026-08-27 04:42:56 TP22 EP22] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP17 EP17] Capture target decode CUDA graph end. elapsed=233.88 s, mem usage=0.58 GB, avail mem=8.58 GB.
+[2026-08-27 04:42:56 TP17 EP17] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP21 EP21] Capture target decode CUDA graph end. elapsed=233.90 s, mem usage=0.58 GB, avail mem=8.58 GB.
+[2026-08-27 04:42:56 TP21 EP21] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP20 EP20] Capture target decode CUDA graph end. elapsed=233.91 s, mem usage=0.60 GB, avail mem=8.52 GB.
+[2026-08-27 04:42:56 TP20 EP20] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP23 EP23] Capture target decode CUDA graph end. elapsed=234.01 s, mem usage=0.62 GB, avail mem=8.51 GB.
+[2026-08-27 04:42:56 TP23 EP23] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP18 EP18] Capture target decode CUDA graph end. elapsed=234.00 s, mem usage=0.56 GB, avail mem=8.59 GB.
+[2026-08-27 04:42:56 TP18 EP18] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+W0827 04:42:57.619849  2423 memory_location.cpp:73] Failed to get NUMA node, addr: 0x9680d380, len: 23168: Operation not permitted [1]
+W0827 04:42:57.619855  2422 memory_location.cpp:73] Failed to get NUMA node, addr: 0x9696f4c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.619967  2424 memory_location.cpp:73] Failed to get NUMA node, addr: 0x966bf4c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.620213  2425 memory_location.cpp:73] Failed to get NUMA node, addr: 0x82b481c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.620281  2426 memory_location.cpp:73] Failed to get NUMA node, addr: 0x988cf580, len: 185344: Operation not permitted [1]
+W0827 04:42:57.620285  2427 memory_location.cpp:73] Failed to get NUMA node, addr: 0x988fc9c0, len: 185344: Operation not permitted [1]
+W0827 04:42:57.620325  2428 memory_location.cpp:73] Failed to get NUMA node, addr: 0x96988ec0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.620384  2429 memory_location.cpp:73] Failed to get NUMA node, addr: 0x96949540, len: 46336: Operation not permitted [1]
+W0827 04:42:57.620546  2430 memory_location.cpp:73] Failed to get NUMA node, addr: 0x96790780, len: 23168: Operation not permitted [1]
+W0827 04:42:57.620606  2431 memory_location.cpp:73] Failed to get NUMA node, addr: 0x967ccac0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.651996  2570 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6c8ac200, len: 23168: Operation not permitted [1]
+W0827 04:42:57.652108  2571 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6c6f5f80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.652212  2572 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6c8ea380, len: 23168: Operation not permitted [1]
+W0827 04:42:57.652417  2573 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6d2e8ec0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.652568  2574 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6ea4eb80, len: 185344: Operation not permitted [1]
+W0827 04:42:57.652587  2576 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6cb20d80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.652642  2577 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6c987340, len: 46336: Operation not permitted [1]
+W0827 04:42:57.652667  2575 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6ea7bfc0, len: 185344: Operation not permitted [1]
+W0827 04:42:57.652815  2578 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6c6eb200, len: 23168: Operation not permitted [1]
+W0827 04:42:57.652889  2579 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6cb0a2c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.713306  2681 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8a7feb00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.713331  2680 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ad55240, len: 23168: Operation not permitted [1]
+W0827 04:42:57.713303  2679 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ad28280, len: 23168: Operation not permitted [1]
+W0827 04:42:57.713424  2682 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8abc1200, len: 23168: Operation not permitted [1]
+W0827 04:42:57.713523  2683 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ca64980, len: 185344: Operation not permitted [1]
+W0827 04:42:57.713629  2685 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ca91dc0, len: 185344: Operation not permitted [1]
+W0827 04:42:57.713903  2687 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ac54c00, len: 46336: Operation not permitted [1]
+W0827 04:42:57.714316  2688 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ab93e40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.714335  2689 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8abe8700, len: 23168: Operation not permitted [1]
+W0827 04:42:57.714352  2686 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8a7f7880, len: 23168: Operation not permitted [1]
+W0827 04:42:57.717235  2716 memory_location.cpp:73] Failed to get NUMA node, addr: 0x739fc3c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.717247  2718 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ff359c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.717484  2725 memory_location.cpp:73] Failed to get NUMA node, addr: 0x75e8fa00, len: 185344: Operation not permitted [1]
+W0827 04:42:57.717736  2729 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ff46700, len: 46336: Operation not permitted [1]
+W0827 04:42:57.718014  2728 memory_location.cpp:73] Failed to get NUMA node, addr: 0x73b6a280, len: 23168: Operation not permitted [1]
+W0827 04:42:57.718112  2730 memory_location.cpp:73] Failed to get NUMA node, addr: 0x73dda340, len: 23168: Operation not permitted [1]
+W0827 04:42:57.718163  2731 memory_location.cpp:73] Failed to get NUMA node, addr: 0x73dbf000, len: 23168: Operation not permitted [1]
+W0827 04:42:57.718196  2721 memory_location.cpp:73] Failed to get NUMA node, addr: 0x73a8f640, len: 23168: Operation not permitted [1]
+W0827 04:42:57.718261  2723 memory_location.cpp:73] Failed to get NUMA node, addr: 0x73c5ba40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.724411  2727 memory_location.cpp:73] Failed to get NUMA node, addr: 0x75ebce40, len: 185344: Operation not permitted [1]
+W0827 04:42:57.755818  2858 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5e996800, len: 23168: Operation not permitted [1]
+W0827 04:42:57.755931  2859 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ec52140, len: 23168: Operation not permitted [1]
+W0827 04:42:57.756026  2861 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5e767ec0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.756074  2862 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60bbb440, len: 185344: Operation not permitted [1]
+W0827 04:42:57.756379  2865 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4aecca80, len: 46336: Operation not permitted [1]
+W0827 04:42:57.756517  2863 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60be8880, len: 185344: Operation not permitted [1]
+W0827 04:42:57.756536  2864 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5eb0d880, len: 23168: Operation not permitted [1]
+W0827 04:42:57.756650  2866 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ea7fe80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.757373  2860 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5eb4d800, len: 23168: Operation not permitted [1]
+W0827 04:42:57.757380  2867 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ea97dc0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.818434  3070 memory_location.cpp:73] Failed to get NUMA node, addr: 0x965568c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.818459  3075 memory_location.cpp:73] Failed to get NUMA node, addr: 0x96700fc0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.818470  3072 memory_location.cpp:73] Failed to get NUMA node, addr: 0x96541bc0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.820174  3079 memory_location.cpp:73] Failed to get NUMA node, addr: 0x984fb440, len: 185344: Operation not permitted [1]
+W0827 04:42:57.822142  3084 memory_location.cpp:73] Failed to get NUMA node, addr: 0x962df200, len: 23168: Operation not permitted [1]
+W0827 04:42:57.822369  3086 memory_location.cpp:73] Failed to get NUMA node, addr: 0x831c0780, len: 46336: Operation not permitted [1]
+W0827 04:42:57.822384  3110 memory_location.cpp:73] Failed to get NUMA node, addr: 0x965cd3c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.822431  3082 memory_location.cpp:73] Failed to get NUMA node, addr: 0x98528880, len: 185344: Operation not permitted [1]
+W0827 04:42:57.822527  3127 memory_location.cpp:73] Failed to get NUMA node, addr: 0x9642bbc0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.829262  3077 memory_location.cpp:73] Failed to get NUMA node, addr: 0x9640b580, len: 23168: Operation not permitted [1]
+W0827 04:42:57.832078  3193 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ce567c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.832136  3194 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8cbe8d00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.832149  3195 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8cc94640, len: 23168: Operation not permitted [1]
+W0827 04:42:57.832294  3196 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8d79f140, len: 23168: Operation not permitted [1]
+W0827 04:42:57.832419  3199 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ce4bec0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.832562  3201 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8d7e4b40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.832683  3202 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8cbbd940, len: 23168: Operation not permitted [1]
+W0827 04:42:57.833209  3198 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ef88080, len: 185344: Operation not permitted [1]
+W0827 04:42:57.833267  3200 memory_location.cpp:73] Failed to get NUMA node, addr: 0x79c278c0, len: 46336: Operation not permitted [1]
+W0827 04:42:57.837312  3197 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ef5ac40, len: 185344: Operation not permitted [1]
+W0827 04:42:58.193480  3458 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59a23f40, len: 23168: Operation not permitted [1]
+W0827 04:42:58.193507  3459 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59aa70c0, len: 23168: Operation not permitted [1]
+W0827 04:42:58.193545  3460 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5999c100, len: 23168: Operation not permitted [1]
+W0827 04:42:58.193585  3461 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59bc3040, len: 23168: Operation not permitted [1]
+W0827 04:42:58.193631  3462 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5bd1e5c0, len: 185344: Operation not permitted [1]
+W0827 04:42:58.193667  3463 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5bd4ba00, len: 185344: Operation not permitted [1]
+W0827 04:42:58.193720  3464 memory_location.cpp:73] Failed to get NUMA node, addr: 0x599fa280, len: 23168: Operation not permitted [1]
+W0827 04:42:58.193746  3466 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59dd6100, len: 23168: Operation not permitted [1]
+W0827 04:42:58.193764  3467 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59cbd400, len: 23168: Operation not permitted [1]
+W0827 04:42:58.193776  3465 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59db0340, len: 46336: Operation not permitted [1]
+[2026-08-27 04:43:00] Dummy health check server started in background thread at 0.0.0.0:30000
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/d_healthy_d_node3.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/d_healthy_d_node3.log
new file mode 100644
index 0000000..9791774
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/d_healthy_d_node3.log
@@ -0,0 +1,498 @@
+/usr/local/lib/python3.12/dist-packages/torchao/quantization/quant_api.py:1731: SyntaxWarning: invalid escape sequence '\.'
+  """Configuration class for applying different quantization configs to modules or parameters based on their fully qualified names (FQNs).
+/sgl-workspace/sglang/python/sglang/launch_server.py:56: UserWarning: 'python -m sglang.launch_server' is still supported, but 'sglang serve' is the recommended entrypoint.
+  Example: sglang serve --model-path <model> [options]
+  warnings.warn(
+[2026-08-27 04:36:00] KV cache is forced as chunk cache for decode server
+[2026-08-27 04:36:03] Breakable CUDA graph is incompatible with MLA attention; disabling prefill CUDA graph.
+[2026-08-27 04:36:03] Attention backend not specified. Use triton backend by default.
+[2026-08-27 04:36:05] server_args=ServerArgs(model_path='/data/hf_models/Kimi-K3', tokenizer_path='/data/hf_models/Kimi-K3', tokenizer_mode='auto', tokenizer_backend='huggingface', tokenizer_worker_num=1, detokenizer_worker_num=1, skip_tokenizer_init=False, load_format='auto', model_loader_extra_config='{}', trust_remote_code=True, context_length=None, is_embedding=False, enable_multimodal=None, revision=None, model_impl='auto', model_config_parser='auto', json_model_override_args='{}', dtype='auto', quantization=None, quantization_param_path=None, kv_cache_dtype='auto', enable_fp32_lm_head=False, modelopt_quant=None, modelopt_checkpoint_restore_path=None, modelopt_checkpoint_save_path=None, modelopt_export_path=None, quantize_and_serve=False, rl_quant_profile=None, enable_tf32_matmul=False, mem_fraction_static=0.88, max_running_requests=None, max_queued_requests=None, max_total_tokens=None, chunked_prefill_size=8192, enable_dynamic_chunking=False, max_prefill_tokens=16384, prefill_max_requests=None, schedule_policy='fcfs', enable_priority_scheduling=False, disable_priority_preemption=False, default_priority_value=None, abort_on_priority_when_disabled=False, schedule_low_priority_values_first=False, priority_scheduling_preemption_threshold=10, retraction_policy='length', schedule_conservativeness=1.0, page_size=64, swa_full_tokens_ratio=0.8, disable_hybrid_swa_memory=False, radix_eviction_policy='lru', prefill_only_disable_kv_cache=False, disable_radix_cache=True, enable_page_major_kv_layout=False, enable_unified_memory=False, disable_chunked_prefix_cache=False, disable_overlap_schedule=False, num_continuous_decode_steps=1, scheduler_recv_interval=1, enable_mixed_chunk=False, nccl_port=None, dist_timeout=3600, dist_init_addr='174.1.60.5:20000', nnodes=4, node_rank=3, tp_size=32, dcp_size=1, pp_size=1, pp_max_micro_batch_size=None, pp_async_batch_depth=0, dp_size=1, load_balance_method='round_robin', attn_cp_size=1, moe_dp_size=1, dcp_comm_backend='ag_rs', dcp_replicate_q_proj=None, dwdp_size=1, enable_prefill_cp=False, cp_strategy=None, enable_dsa_cache_layer_split=False, enable_dsa_prefill_context_parallel=False, dsa_prefill_cp_mode='round-robin-split', enable_prefill_context_parallel=False, prefill_cp_mode='in-seq-split', enable_cp_decode_attn_tp=False, enable_dp_attention=False, enable_dp_attention_local_control_broadcast=False, enable_dp_lm_head=False, enable_attn_tp_input_scattered=False, disable_attn_tp_gather=False, enable_p2p_check=False, device='cuda', base_gpu_id=0, gpu_id_step=1, random_seed=297832107, watchdog_timeout=300, soft_watchdog_timeout=None, sleep_on_idle=False, use_ray=False, custom_sigquit_handler=None, numa_node=None, gc_threshold=None, host='0.0.0.0', port=30000, fastapi_root_path='', smg_grpc_mode=False, grpc_mode=False, grpc_port=None, sidecar=None, sidecar_args=None, skip_server_warmup=False, warmups=None, enable_http2=False, ssl_keyfile=None, ssl_certfile=None, ssl_ca_certs=None, ssl_keyfile_password=None, enable_ssl_refresh=False, api_key=None, admin_api_key=None, served_model_name='kimi-k3', weight_version='default', chat_template=None, hf_chat_template_name=None, completion_template=None, file_storage_path='sglang_storage', enable_cache_report=False, reasoning_parser=None, default_chat_template_kwargs=None, strip_thinking_cache=False, enable_strict_thinking=False, tool_call_parser=None, tool_server=None, sampling_defaults='model', asr_max_buffer_seconds=60, asr_max_concurrent_sessions=32, preferred_sampling_params=None, allow_auto_truncate=False, stream_interval=1, batch_notify_size=16, stream_response_default_include_usage=False, incremental_streaming_output=False, enable_streaming_session=False, enable_session_radix_cache=False, log_level='info', log_level_http=None, log_requests=False, log_requests_level=2, log_requests_format='text', log_requests_target=None, uvicorn_access_log_exclude_prefixes=[], crash_dump_folder=None, show_time_cost=False, enable_metrics=False, smg_http_sidecar_port=None, enable_mfu_metrics=False, enable_metrics_for_all_schedulers=False, load_snapshot_publish_interval=15, tokenizer_metrics_custom_labels_header='x-custom-labels', tokenizer_metrics_allowed_custom_labels=None, extra_metric_labels=None, bucket_time_to_first_token=None, bucket_inter_token_latency=None, bucket_e2e_request_latency=None, prompt_tokens_buckets=None, generation_tokens_buckets=None, gc_warning_threshold_secs=0.0, decode_log_interval=40, enable_request_time_stats_logging=False, kv_events_config=None, enable_forward_pass_metrics=False, forward_pass_metrics_worker_id='', forward_pass_metrics_ipc_name=None, enable_trace=False, trace_modules='request', otlp_traces_endpoint='localhost:4317', export_metrics_to_file=False, export_metrics_to_file_dir=None, stat_loggers=None, constrained_json_whitespace_pattern=None, constrained_json_disable_any_whitespace=False, attention_backend='triton', decode_attention_backend=None, prefill_attention_backend=None, sampling_backend='flashinfer', grammar_backend='xgrammar', radix_cache_backend=None, mm_attention_backend=None, fp8_gemm_runner_backend='auto', fp4_gemm_runner_backend='auto', bf16_gemm_backend='auto', dsa_prefill_backend=None, dsa_decode_backend=None, dsa_paged_mqa_logits_backend='auto', dsa_topk_backend='sgl-kernel', disable_flashinfer_autotune=False, flashinfer_autotune_skip_ops=None, mamba_backend='triton', cuda_graph_config=CudaGraphConfig(decode=PhaseConfig(backend='full', max_bs=16, bs=[1, 2, 4, 8, 12, 16], tc_compiler='eager', full_prefill_max_req=None), prefill=PhaseConfig(backend='disabled', max_bs=2048, bs=[4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], tc_compiler='eager', full_prefill_max_req=None)), cuda_graph_backend_decode=None, cuda_graph_backend_prefill=None, cuda_graph_max_bs_decode=16, cuda_graph_max_bs_prefill=None, cuda_graph_bs_decode=None, cuda_graph_bs_prefill=None, cuda_graph_tc_compiler=None, disable_prefill_cuda_graph=False, disable_decode_cuda_graph=False, disable_cuda_graph=False, disable_cuda_graph_padding=False, enable_profile_cuda_graph=False, enable_cudagraph_gc=False, debug_cuda_graph=False, enable_layerwise_nvtx_marker=False, enable_nccl_nvls=False, enable_symm_mem=False, triton_attention_reduce_in_fp32=False, triton_attention_num_kv_splits=8, triton_attention_split_tile_size=None, flashinfer_mla_disable_ragged=False, enable_fused_qk_norm_rope=False, enable_precise_embedding_interpolation=False, enable_fused_moe_sum_all_reduce=False, enable_deepseek_v4_fp4_indexer=False, disable_custom_all_reduce=False, enable_mscclpp=False, enable_torch_symm_mem=False, enable_scattered_sconv=False, pre_warm_nccl=False, enable_quant_communications=False, enable_flashinfer_allreduce_fusion=False, enforce_disable_flashinfer_allreduce_fusion=False, flashinfer_allreduce_fusion_backend=None, enable_aiter_allreduce_fusion=False, enable_torch_compile=False, enable_torch_compile_debug_mode=False, torch_compile_max_bs=32, torchao_config='', speculative_algorithm=None, speculative_draft_model_path=None, speculative_draft_model_revision=None, speculative_draft_load_format=None, speculative_num_steps=None, speculative_eagle_topk=None, speculative_num_draft_tokens=None, speculative_dflash_block_size=None, speculative_dspark_block_size=None, speculative_dspark_sps_table_path=None, speculative_dspark_confidence_sts_path=None, speculative_dspark_align_verify_tokens_to_graph_tier=False, speculative_accept_threshold_single=1.0, speculative_accept_threshold_acc=1.0, speculative_use_rejection_sampling=False, speculative_token_map=None, speculative_attention_mode='prefill', speculative_draft_attention_backend=None, speculative_draft_window_size=None, speculative_moe_runner_backend='marlin', speculative_moe_a2a_backend=None, speculative_draft_model_quantization=None, speculative_skip_dp_mlp_sync=False, enable_multi_layer_eagle=False, speculative_adaptive=False, speculative_adaptive_config=None, decoupled_spec_bind_endpoint=None, decoupled_spec_connect_endpoints=None, decoupled_spec_rank=None, decoupled_spec_role='null', spec_trace_dir=None, speculative_ngram_min_bfs_breadth=1, speculative_ngram_max_bfs_breadth=10, speculative_ngram_match_type='BFS', speculative_ngram_max_trie_depth=18, speculative_ngram_capacity=10000000, speculative_ngram_external_corpus_path=None, speculative_ngram_external_sam_budget=0, speculative_ngram_external_corpus_max_tokens=10000000, ep_size=32, moe_a2a_backend='none', moe_runner_backend='marlin', flashinfer_mxfp4_moe_precision='default', deepep_mode='auto', fuseep_mode=2, deepep_dispatcher_output_dtype='auto', ep_num_redundant_experts=0, ep_dispatch_algorithm=None, init_expert_location='trivial', enable_eplb=False, eplb_algorithm='auto', eplb_rebalance_num_iterations=1000, eplb_rebalance_layers_per_chunk=None, eplb_min_rebalancing_utilization_threshold=1.0, expert_distribution_recorder_mode=None, expert_distribution_recorder_buffer_size=1000, enable_expert_distribution_metrics=False, deepep_config=None, moe_dense_tp_size=None, elastic_ep_backend=None, enable_elastic_expert_backup=False, mooncake_ib_device=None, enable_waterfill=False, ep_join_mode=None, ep_join_rank_offset=0, elastic_ep_initial_size=None, max_ep_size=None, elastic_ep_scale_timeout=600, elastic_ep_rejoin=False, disable_flashinfer_cutlass_moe_fp4_allgather=False, disable_shared_experts_fusion=False, enforce_shared_experts_fusion=False, max_mamba_cache_size=None, mamba_ssm_dtype=None, mamba_max_states_per_path=-1, enable_mamba_cache_stochastic_rounding=False, mamba_cache_philox_rounds=0, mamba_full_memory_ratio=0.21, mamba_radix_cache_strategy='extra_buffer_lazy', uses_mamba_radix_cache=False, mamba_track_interval=256, enable_int8_mamba_checkpoint=False, int8_mamba_ckpt_size=None, linear_attn_backend='triton', linear_attn_decode_backend=None, linear_attn_prefill_backend=None, linear_attn_verify_backend=None, enable_linear_replayssm=False, linear_replayssm_cache_len=16, enable_linear_replayssm_spec=False, enable_hierarchical_cache=False, hicache_ratio=2.0, hicache_size=0, hicache_write_policy='write_through', hicache_io_backend='kernel', hicache_mem_layout='page_first', hicache_storage_backend=None, hicache_storage_prefetch_policy='timeout', hicache_storage_backend_extra_config=None, enable_hisparse=False, hisparse_config=None, enable_broadcast_mm_inputs_process=False, enable_prefix_mm_cache=False, mm_enable_dp_encoder=False, mm_process_config={}, mm_processor_worker_num=0, mm_io_worker_num=0, limit_mm_data_per_request=None, enable_mm_global_cache=False, disable_fast_image_processor=False, mm_feature_transport='cpu', keep_mm_feature_on_device=False, enable_lora=None, enable_lora_overlap_loading=None, max_lora_rank=None, lora_target_modules=None, lora_paths=None, max_loaded_loras=None, max_loras_per_batch=8, lora_eviction_policy='lru', lora_backend='csgmv', max_lora_chunk_size=16, experts_shared_outer_loras=None, lora_use_virtual_experts=False, lora_strict_loading=False, lora_drain_wait_threshold=0.0, enable_two_batch_overlap=False, enable_single_batch_overlap=False, tbo_token_distribution_threshold=0.48, cpu_offload_gb=0, offload_group_size=-1, offload_num_in_group=1, offload_prefetch_step=1, offload_mode='cpu', enable_lmcache=False, lmcache_config_file=None, enable_flexkv=False, flexkv_config_file=None, kt_weight_path=None, kt_method='AMXINT4', kt_cpuinfer=None, kt_threadpool_count=2, kt_num_gpu_experts=None, kt_max_deferred_experts_per_token=None, dllm_algorithm=None, dllm_algorithm_config=None, dllm_fdfo=True, disaggregation_mode='decode', disaggregation_transfer_backend='mooncake', disaggregation_bootstrap_port=28800, disaggregation_ib_device='mlx5_0,mlx5_1,mlx5_2,mlx5_3', disaggregation_decode_enable_radix_cache=False, disaggregation_decode_enable_offload_kvcache=False, num_reserved_decode_tokens=512, disaggregation_decode_extra_slots=0, disaggregation_decode_polling_interval=1, optimistic_prefill_attempts=0, encoder_only=False, language_only=False, encoder_transfer_backend='zmq_to_tokenizer', encoder_urls=[], encoder_bootstrap_port=8997, encoder_register_urls=[], enable_adaptive_dispatch_to_encoder=False, enable_pdmux=False, pdmux_config_path=None, sm_group_num=8, custom_weight_loader=[], weight_loader_disable_mmap=False, weight_loader_prefetch_checkpoints=False, weight_loader_prefetch_num_threads=4, weight_loader_drop_cache_after_load=False, remote_instance_weight_loader_seed_instance_ip=None, remote_instance_weight_loader_seed_instance_service_port=None, remote_instance_weight_loader_send_weights_group_ports=None, remote_instance_weight_loader_backend='nccl', remote_instance_weight_loader_start_seed_via_transfer_engine=False, engine_info_bootstrap_port=6789, modelexpress_config=None, download_dir=None, model_checksum=None, delete_ckpt_after_loading=False, decrypted_config_file=None, decrypted_draft_config_file=None, checkpoint_engine_wait_weights_before_ready=False, enable_prefill_delayer=False, prefill_delayer_max_delay_passes=30, prefill_delayer_token_usage_low_watermark=None, prefill_delayer_forward_passes_buckets=None, prefill_delayer_wait_seconds_buckets=None, prefill_delayer_queue_min_ratio=None, prefill_delayer_max_delay_ms=None, min_free_slots_delay=None, enable_deterministic_inference=False, rl_on_policy_target=None, kv_canary='none', kv_canary_real_data='none', kv_canary_sweep_interval=0, enable_dynamic_batch_tokenizer=False, dynamic_batch_tokenizer_batch_size=32, dynamic_batch_tokenizer_batch_timeout=0.002, enable_tokenizer_batch_encode=False, disable_tokenizer_batch_decode=False, debug_tensor_dump_output_folder=None, debug_tensor_dump_layers=None, debug_tensor_dump_input_file=None, enable_memory_saver=False, enable_weights_cpu_backup=False, enable_draft_weights_cpu_backup=False, enable_custom_logit_processor=False, enable_return_hidden_states=False, enable_return_routed_experts=False, enable_return_indexer_topk=False, disable_outlines_disk_cache=False, enable_mis=False, weight_cache_mode='off', weight_cache_socket=None, weight_cache_timeout=1800, forward_hooks=None, msprobe_dump_config=None)
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:16 TP30 EP30] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:16 TP30 EP30] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:19 TP30 EP30] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:19.321749   241 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:19.321810   241 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:19.321820   241 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.8 port: 12001
+I0827 04:36:19.321856   241 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.8:16569
+I0827 04:36:19.321940   241 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:19.343153   241 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:19.347820   241 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:19.373265   241 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:19.380316   241 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:19.383023   241 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:19.383050   241 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:19.383101   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:19.390141   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:19.391583   241 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:08
+I0827 04:36:19.391615   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:19.417495   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:19.418977   241 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:08
+I0827 04:36:19.419003   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:19.445492   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:19.446861   241 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:08
+I0827 04:36:19.446884   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:19.473483   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:19.474817   241 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:08
+I0827 04:36:19.474922   241 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:19 TP30 EP30] Init torch distributed begin.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:20 TP27 EP27] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:20 TP27 EP27] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:21 TP25 EP25] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:21 TP25 EP25] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:21 TP26 EP26] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:21 TP26 EP26] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:21 TP28 EP28] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:21 TP28 EP28] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:21 TP24 EP24] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:21 TP24 EP24] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:21 TP29 EP29] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:21 TP29 EP29] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:21 TP31 EP31] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:21 TP31 EP31] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:22 TP27 EP27] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:22.734625   238 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:22.734687   238 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:22.734699   238 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.8 port: 12001
+I0827 04:36:22.734735   238 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.8:16566
+I0827 04:36:22.734848   238 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:22.760079   238 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:22.768187   238 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:22.777189   238 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:22.786154   238 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:22.789806   238 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:22.789830   238 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:22.789880   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:22.797986   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:22.799417   238 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:08
+I0827 04:36:22.799446   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:22.825555   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:22.826912   238 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:08
+I0827 04:36:22.826946   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:22.853536   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:22.854885   238 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:08
+I0827 04:36:22.854909   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:22.881553   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:22.882910   238 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:08
+I0827 04:36:22.883030   238 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:22 TP27 EP27] Init torch distributed begin.
+[2026-08-27 04:36:22 TP25 EP25] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:22 TP28 EP28] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:22 TP26 EP26] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:23 TP29 EP29] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:23 TP24 EP24] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:23.476856   236 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:23.476951   236 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:23.476967   236 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.8 port: 12001
+I0827 04:36:23.477013   236 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.8:15425
+I0827 04:36:23.477150   236 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:23.477173   239 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:23.477272   239 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:23.477288   239 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.8 port: 12001
+I0827 04:36:23.477344   239 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.8:15710
+I0827 04:36:23.477483   239 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:23.483197   236 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:23.483803   239 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:23.488940   236 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:23.489802   239 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:23.494866   236 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:23.495663   239 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:23.500973   236 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:23.501806   239 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:23.503927   236 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:23.503952   236 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:23.503998   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.504750   239 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:23.504770   239 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:23.504812   239 rdma_context.cpp:205] Using SIEVE endpoint store
+[2026-08-27 04:36:23 TP31 EP31] Multimodal data loading enabled with 16 worker threads (auto).
+I0827 04:36:23.510515   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:23.511322   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:23.511830   236 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:08
+I0827 04:36:23.511855   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.512616   239 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:08
+I0827 04:36:23.512640   239 rdma_context.cpp:205] Using SIEVE endpoint store
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:23.522899   237 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:23.523005   237 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:23.523020   237 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.8 port: 12001
+I0827 04:36:23.523065   237 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.8:16748
+I0827 04:36:23.523202   237 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:23.529469   237 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:23.535485   237 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:23.537460   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:23.537528   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:23.538903   236 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:08
+I0827 04:36:23.538913   239 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:08
+I0827 04:36:23.538928   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.538945   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.541347   237 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:23.546439   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:23.547672   237 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:23.547796   239 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:08
+I0827 04:36:23.547820   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.550041   237 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:23.550061   237 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:23.550102   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.555119   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:23.556490   239 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:08
+I0827 04:36:23.556598   239 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:23 TP28 EP28] Init torch distributed begin.
+I0827 04:36:23.557152   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:23.558562   237 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:08
+I0827 04:36:23.558589   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.565423   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:23.566766   236 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:08
+I0827 04:36:23.566788   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.589463   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:23.590811   237 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:08
+I0827 04:36:23.590835   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.593444   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:23.594800   236 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:08
+I0827 04:36:23.594898   236 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:23 TP25 EP25] Init torch distributed begin.
+I0827 04:36:23.598161   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:23.599563   237 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:08
+I0827 04:36:23.599589   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.625564   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:23.626922   237 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:08
+I0827 04:36:23.627038   237 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:23 TP26 EP26] Init torch distributed begin.
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:23.899124   240 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:23.899245   240 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:23.899261   240 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.8 port: 12001
+I0827 04:36:23.899313   240 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.8:15127
+I0827 04:36:23.899492   240 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:23.907753   235 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:23.907863   235 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:23.907878   235 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.8 port: 12001
+I0827 04:36:23.907927   235 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.8:15330
+I0827 04:36:23.908092   235 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:23.914011   235 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:23.919618   235 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:23.925108   235 topology.cpp:127] Device mlx5_2 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:23.925450   242 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:23.925558   242 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:23.925576   242 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.8 port: 12001
+I0827 04:36:23.925631   242 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.8:16235
+I0827 04:36:23.925693   240 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:23.925761   242 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:23.930419   242 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:23.933247   240 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:23.934639   242 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:23.939190   242 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:23.940605   240 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:23.943866   242 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:23.945531   242 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:23.945554   242 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:23.945595   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.947860   240 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:23.949263   235 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:23.951108   240 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:23.951148   240 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:23.951216   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.952417   235 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:23.952445   235 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:23.952502   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.958555   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:23.959836   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:23.960163   240 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:08
+I0827 04:36:23.960197   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.961452   235 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:08
+I0827 04:36:23.961483   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.969228   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:23.969322   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:23.970860   235 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:08
+I0827 04:36:23.970890   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.971050   242 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:08
+I0827 04:36:23.971084   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.979120   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:23.980706   242 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:08
+I0827 04:36:23.980736   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.989722   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:23.991313   240 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:08
+I0827 04:36:23.991346   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:24.001633   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:24.003211   235 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:08
+I0827 04:36:24.003237   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:24.005753   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:24.007426   242 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:08
+I0827 04:36:24.007457   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:24.021715   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:24.023293   240 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:08
+I0827 04:36:24.023322   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:24.034039   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:24.034071   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:24.035696   235 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:08
+I0827 04:36:24.035816   235 transfer_engine_impl.cpp:388] installTransport, type=rdma
+I0827 04:36:24.035794   242 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:08
+I0827 04:36:24.035987   242 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:24 TP24 EP24] Init torch distributed begin.
+[2026-08-27 04:36:24 TP31 EP31] Init torch distributed begin.
+I0827 04:36:24.049829   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:24.051388   240 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:08
+I0827 04:36:24.051532   240 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:24 TP29 EP29] Init torch distributed begin.
+[2026-08-27 04:36:48 TP31 EP31] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP30 EP30] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP28 EP28] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP29 EP29] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP27 EP27] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP26 EP26] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP25 EP25] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP24 EP24] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP31 EP31] Init torch distributed ends. elapsed=24.20 s, mem usage=0.40 GB
+[2026-08-27 04:36:48 TP30 EP30] Init torch distributed ends. elapsed=28.76 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP29 EP29] Init torch distributed ends. elapsed=24.18 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP28 EP28] Init torch distributed ends. elapsed=24.68 s, mem usage=0.40 GB
+[2026-08-27 04:36:48 TP27 EP27] Init torch distributed ends. elapsed=25.35 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP26 EP26] Init torch distributed ends. elapsed=24.61 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP25 EP25] Init torch distributed ends. elapsed=24.64 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP24 EP24] Init torch distributed ends. elapsed=24.20 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP28 EP28] Load weight begin. avail mem=82.19 GB
+[2026-08-27 04:36:48 TP27 EP27] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP30 EP30] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP24 EP24] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP29 EP29] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP26 EP26] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP25 EP25] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP28 EP28] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP30 EP30] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP27 EP27] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP29 EP29] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP24 EP24] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP26 EP26] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP25 EP25] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP31 EP31] Load weight begin. avail mem=82.19 GB
+[2026-08-27 04:36:48 TP31 EP31] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP27 EP27] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP28 EP28] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP24 EP24] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP27 EP27] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP30 EP30] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP29 EP29] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP26 EP26] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP25 EP25] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP28 EP28] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP24 EP24] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP29 EP29] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP31 EP31] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP30 EP30] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP26 EP26] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP25 EP25] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP31 EP31] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:50 TP29 EP29] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP28 EP28] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP30 EP30] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP26 EP26] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP25 EP25] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP31 EP31] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP27 EP27] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP24 EP24] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:38:54 TP26 EP26] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:54 TP27 EP27] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:54 TP24 EP24] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:54 TP26 EP26] Load weight end. elapsed=125.61 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:54 TP29 EP29] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:54 TP27 EP27] Load weight end. elapsed=125.61 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:54 TP24 EP24] Load weight end. elapsed=125.61 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:54 TP29 EP29] Load weight end. elapsed=125.61 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:54 TP25 EP25] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:54 TP25 EP25] Load weight end. elapsed=125.64 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:54 TP28 EP28] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:54 TP28 EP28] Load weight end. elapsed=125.69 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:54 TP30 EP30] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:54 TP30 EP30] Load weight end. elapsed=125.71 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:54 TP31 EP31] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:54 TP31 EP31] Load weight end. elapsed=125.77 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:39:01 TP26 EP26] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP26 EP26] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP30 EP30] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP30 EP30] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP28 EP28] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP28 EP28] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP27 EP27] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP27 EP27] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP25 EP25] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP25 EP25] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP31 EP31] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP31 EP31] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP24 EP24] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP24 EP24] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP29 EP29] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP29 EP29] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:02 TP29 EP29] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP28 EP28] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP26 EP26] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP27 EP27] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP30 EP30] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP25 EP25] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP31 EP31] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP24 EP24] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP29 EP29] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP30 EP30] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP28 EP28] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP26 EP26] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP27 EP27] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP25 EP25] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP24 EP24] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP31 EP31] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP29 EP29] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP29 EP29] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP28 EP28] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP28 EP28] Memory pool end. avail mem=9.18 GB
+[2026-08-27 04:39:02 TP27 EP27] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP24 EP24] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP26 EP26] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP27 EP27] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP31 EP31] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP24 EP24] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP26 EP26] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP31 EP31] Memory pool end. avail mem=9.18 GB
+[2026-08-27 04:39:02 TP30 EP30] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP25 EP25] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP30 EP30] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP25 EP25] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP29 EP29] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP29 EP29] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP24 EP24] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP24 EP24] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:39:02 TP28 EP28] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP28 EP28] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.17 GB
+[2026-08-27 04:39:02 TP25 EP25] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP31 EP31] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP25 EP25] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP31 EP31] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.17 GB
+[2026-08-27 04:39:02 TP26 EP26] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP26 EP26] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP27 EP27] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP27 EP27] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:39:02 TP30 EP30] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP30 EP30] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:42:56 TP30 EP30] Capture target decode CUDA graph end. elapsed=233.69 s, mem usage=0.56 GB, avail mem=8.59 GB.
+[2026-08-27 04:42:56 TP30 EP30] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP27 EP27] Capture target decode CUDA graph end. elapsed=233.73 s, mem usage=0.62 GB, avail mem=8.51 GB.
+[2026-08-27 04:42:56 TP27 EP27] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP31 EP31] Capture target decode CUDA graph end. elapsed=233.74 s, mem usage=0.62 GB, avail mem=8.55 GB.
+[2026-08-27 04:42:56 TP31 EP31] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP26 EP26] Capture target decode CUDA graph end. elapsed=233.76 s, mem usage=0.56 GB, avail mem=8.59 GB.
+[2026-08-27 04:42:56 TP26 EP26] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP29 EP29] Capture target decode CUDA graph end. elapsed=233.82 s, mem usage=0.58 GB, avail mem=8.58 GB.
+[2026-08-27 04:42:56 TP29 EP29] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP28 EP28] Capture target decode CUDA graph end. elapsed=233.79 s, mem usage=0.60 GB, avail mem=8.56 GB.
+[2026-08-27 04:42:56 TP28 EP28] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP24 EP24] Capture target decode CUDA graph end. elapsed=233.88 s, mem usage=0.60 GB, avail mem=8.52 GB.
+[2026-08-27 04:42:56 TP24 EP24] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP25 EP25] Capture target decode CUDA graph end. elapsed=233.89 s, mem usage=0.58 GB, avail mem=8.58 GB.
+[2026-08-27 04:42:56 TP25 EP25] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+W0827 04:42:57.132246  2431 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5913c7c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.132292  2432 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58be65c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.132460  2435 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5af7a600, len: 185344: Operation not permitted [1]
+W0827 04:42:57.132488  2433 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58ed5900, len: 23168: Operation not permitted [1]
+W0827 04:42:57.132501  2436 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5afa7a40, len: 185344: Operation not permitted [1]
+W0827 04:42:57.132512  2437 memory_location.cpp:73] Failed to get NUMA node, addr: 0x590583c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.132552  2434 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58c15080, len: 23168: Operation not permitted [1]
+W0827 04:42:57.132607  2439 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59046600, len: 23168: Operation not permitted [1]
+W0827 04:42:57.132640  2438 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5afd4e80, len: 46336: Operation not permitted [1]
+W0827 04:42:57.132716  2440 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58e5fc80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.276726  2580 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a339340, len: 23168: Operation not permitted [1]
+W0827 04:42:57.276731  2579 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59705880, len: 23168: Operation not permitted [1]
+W0827 04:42:57.276762  2581 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a32b080, len: 23168: Operation not permitted [1]
+W0827 04:42:57.276839  2582 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a331fc0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.276887  2583 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5bb81840, len: 185344: Operation not permitted [1]
+W0827 04:42:57.276939  2584 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5bbaec80, len: 185344: Operation not permitted [1]
+W0827 04:42:57.276983  2585 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59b7b6c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.277042  2587 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59953f00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.277055  2586 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59517380, len: 46336: Operation not permitted [1]
+W0827 04:42:57.277086  2588 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59b87640, len: 23168: Operation not permitted [1]
+W0827 04:42:57.448414  2727 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a4bd3c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.448436  2728 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ac9eb00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.448488  2729 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a2fe240, len: 23168: Operation not permitted [1]
+W0827 04:42:57.448798  2735 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a6135c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.448812  2734 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59ed25c0, len: 46336: Operation not permitted [1]
+W0827 04:42:57.448840  2733 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a3c3080, len: 23168: Operation not permitted [1]
+W0827 04:42:57.448849  2736 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a181e80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.451195  2730 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5acaf240, len: 23168: Operation not permitted [1]
+W0827 04:42:57.451217  2731 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5c61b840, len: 185344: Operation not permitted [1]
+W0827 04:42:57.455173  2732 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5c648c80, len: 185344: Operation not permitted [1]
+W0827 04:42:57.460536  2816 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a391f40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.460651  2817 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ae89700, len: 23168: Operation not permitted [1]
+W0827 04:42:57.460843  2819 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ae0fa00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.460901  2826 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a4e2c80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.460979  2830 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ae7a300, len: 23168: Operation not permitted [1]
+W0827 04:42:57.461052  2821 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5c6d6880, len: 185344: Operation not permitted [1]
+W0827 04:42:57.461104  2820 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a22ed40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.462260  2823 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5c703cc0, len: 185344: Operation not permitted [1]
+W0827 04:42:57.463069  2827 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4678c140, len: 46336: Operation not permitted [1]
+W0827 04:42:57.463085  2829 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a645e80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.466511  2955 memory_location.cpp:73] Failed to get NUMA node, addr: 0x597592c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.466544  2958 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59908b40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.466676  2960 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a2f9c80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.466849  2953 memory_location.cpp:73] Failed to get NUMA node, addr: 0x599f9a80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.467304  2964 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5b9e4d40, len: 185344: Operation not permitted [1]
+W0827 04:42:57.470024  2966 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5993fa80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.471112  2970 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59898900, len: 23168: Operation not permitted [1]
+W0827 04:42:57.471313  2962 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5b9b7900, len: 185344: Operation not permitted [1]
+W0827 04:42:57.475107  2968 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5991bac0, len: 46336: Operation not permitted [1]
+W0827 04:42:57.475121  2972 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59b5ea40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.493731  3089 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59564740, len: 23168: Operation not permitted [1]
+W0827 04:42:57.493901  3097 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5b355380, len: 185344: Operation not permitted [1]
+W0827 04:42:57.494069  3104 memory_location.cpp:73] Failed to get NUMA node, addr: 0x45dbac00, len: 46336: Operation not permitted [1]
+W0827 04:42:57.494072  3105 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5902f500, len: 23168: Operation not permitted [1]
+W0827 04:42:57.494150  3094 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59498600, len: 23168: Operation not permitted [1]
+W0827 04:42:57.494194  3099 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5b3827c0, len: 185344: Operation not permitted [1]
+W0827 04:42:57.499953  3107 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59486740, len: 23168: Operation not permitted [1]
+W0827 04:42:57.502707  3102 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59016c80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.504236  3091 memory_location.cpp:73] Failed to get NUMA node, addr: 0x594d2200, len: 23168: Operation not permitted [1]
+W0827 04:42:57.504359  3095 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59354e40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.514688  3200 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5885e940, len: 23168: Operation not permitted [1]
+W0827 04:42:57.514716  3202 memory_location.cpp:73] Failed to get NUMA node, addr: 0x587d55c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.514720  3201 memory_location.cpp:73] Failed to get NUMA node, addr: 0x586bd5c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.514801  3204 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5aa56c00, len: 185344: Operation not permitted [1]
+W0827 04:42:57.514810  3205 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5aa84040, len: 185344: Operation not permitted [1]
+W0827 04:42:57.514727  3203 memory_location.cpp:73] Failed to get NUMA node, addr: 0x44d09c40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.514963  3206 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58692200, len: 23168: Operation not permitted [1]
+W0827 04:42:57.515017  3209 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58865100, len: 23168: Operation not permitted [1]
+W0827 04:42:57.515022  3207 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58ae8bc0, len: 46336: Operation not permitted [1]
+W0827 04:42:57.515030  3208 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58683a40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.548924  3433 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5895be80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.548955  3434 memory_location.cpp:73] Failed to get NUMA node, addr: 0x593aa600, len: 23168: Operation not permitted [1]
+W0827 04:42:57.549010  3435 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59399e80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.549088  3436 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58a4fa40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.549110  3437 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ab36800, len: 185344: Operation not permitted [1]
+W0827 04:42:57.549167  3438 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ab63c40, len: 185344: Operation not permitted [1]
+W0827 04:42:57.551182  3439 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5936aac0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.551239  3441 memory_location.cpp:73] Failed to get NUMA node, addr: 0x455ad780, len: 23168: Operation not permitted [1]
+W0827 04:42:57.551265  3442 memory_location.cpp:73] Failed to get NUMA node, addr: 0x594169c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.551280  3440 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5853c980, len: 46336: Operation not permitted [1]
+[2026-08-27 04:42:59] Dummy health check server started in background thread at 0.0.0.0:30000
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/d_healthy_mc_master.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/d_healthy_mc_master.log
new file mode 100644
index 0000000..cf99d27
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/d_healthy_mc_master.log
@@ -0,0 +1,94 @@
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+I0827 04:28:16.835302    99 master.cpp:1366] Master service started on port 50051, max_threads=16, enable_metric_reporting=1, metrics_port=9003, default_kv_lease_ttl=10000, default_kv_soft_pin_ttl=1800000, allow_evict_soft_pinned_objects=1, eviction_ratio=0.05, eviction_high_watermark_ratio=0.9, enable_ha=0, enable_offload=0, enable_kv_events=0, kv_events_bind_endpoint=, kv_events_backend_id=, offload_on_evict=0, offload_force_evict=0, offloading_queue_limit=50000, offload_cap_ratio=0.5, ha_backend_type=etcd, ha_backend_connstring=, etcd_endpoints=, client_ttl=10, rpc_thread_num=16, rpc_port=50051, rpc_address=0.0.0.0, rpc_interface=, rpc_conn_timeout_seconds=0, rpc_enable_tcp_no_delay=1, rpc protocol=tcp, cluster_id=mooncake_cluster, root_fs_dir=, global_file_segment_size=9223372036854775807, memory_allocator=offset, enable_http_metadata_server=0, http_metadata_server_port=8080, http_metadata_server_host=0.0.0.0, enable_metadata_cleanup_on_timeout=0, put_start_discard_timeout_sec=30, put_start_release_timeout_sec=600, max_total_finished_tasks=10000, max_total_pending_tasks=10000, max_total_processing_tasks=10000, pending_task_timeout_sec=300, processing_task_timeout_sec=300, enable_snapshot=0, enable_snapshot_restore=0, snapshot_interval_seconds=600, snapshot_backup_dir=, snapshot_object_store_type=, snapshot_catalog_store_type=, snapshot_retention_count=2, max_retry_attempts=10, enable_cxl=0, cxl_path=/dev/dax0.0, cxl_size=8589934592
+I0827 04:28:16.866606   108 master_service.cpp:1520] Task cleanup thread started
+I0827 04:28:16.867491    99 master_admin_service.cpp:311] Master admin server started on port 9003
+I0827 04:28:16.869802   111 master_admin_service.cpp:302] Master Admin Metrics: role=standby, state=starting, service_ready=false, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:28:26.870203   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:28:36.870568   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:28:46.870873   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:28:56.871232   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:29:06.871615   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:29:16.871969   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:29:26.872316   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:29:36.872687   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:29:46.873068   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:29:56.873467   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:30:06.873914   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:30:16.874214   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:30:26.874567   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:30:36.874953   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:30:46.875332   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:30:56.875741   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:31:06.876119   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:31:16.876497   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:31:26.876866   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:31:36.877233   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:31:46.877605   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:31:56.877948   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:32:06.878437   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:32:16.879210   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:32:26.879711   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:32:36.880080   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:32:46.880565   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:32:56.881004   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:33:06.881398   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:33:16.881786   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:33:26.882170   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:33:36.882550   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:33:46.882939   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:33:56.883315   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:34:06.883746   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:34:16.884127   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:34:26.884568   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:34:36.884953   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:34:46.885334   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:34:56.885706   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:35:06.886085   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:35:16.886464   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:35:26.886835   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:35:36.887218   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:35:46.887596   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:35:56.887964   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:36:06.888356   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:36:16.888742   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:36:26.889113   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:36:36.889495   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:36:46.889882   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:36:56.890261   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:37:06.890656   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:37:16.891040   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:37:26.891430   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:37:36.891836   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:37:46.892215   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:37:56.892597   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:38:06.892982   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:38:16.893360   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:38:26.893747   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:38:36.894131   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:38:46.894515   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:38:56.894920   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:39:06.895303   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:39:16.895701   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:39:26.896085   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:39:36.896471   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:39:46.896852   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:39:56.897240   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:40:06.897629   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:40:16.898005   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:40:26.898384   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:40:36.898761   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:40:46.899144   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:40:56.899530   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:41:06.899937   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:41:16.900316   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:41:26.900695   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:41:36.901089   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:41:46.901479   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:41:56.901856   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:42:06.902242   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:42:16.902666   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:42:26.903064   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:42:36.903537   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:42:46.903987   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:42:56.904397   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:43:06.904863   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/d_healthy_p_node0.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/d_healthy_p_node0.log
new file mode 100644
index 0000000..85cfdec
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/d_healthy_p_node0.log
@@ -0,0 +1,941 @@
+/usr/local/lib/python3.12/dist-packages/torchao/quantization/quant_api.py:1731: SyntaxWarning: invalid escape sequence '\.'
+  """Configuration class for applying different quantization configs to modules or parameters based on their fully qualified names (FQNs).
+/sgl-workspace/sglang/python/sglang/launch_server.py:56: UserWarning: 'python -m sglang.launch_server' is still supported, but 'sglang serve' is the recommended entrypoint.
+  Example: sglang serve --model-path <model> [options]
+  warnings.warn(
+[2026-08-27 04:28:54] Breakable CUDA graph is incompatible with MLA attention; disabling prefill CUDA graph.
+[2026-08-27 04:28:55] Attention backend not specified. Use triton backend by default.
+[2026-08-27 04:28:55] Pipeline parallelism is incompatible with overlap schedule.
+[2026-08-27 04:28:58] server_args=ServerArgs(model_path='/data/hf_models/Kimi-K3', tokenizer_path='/data/hf_models/Kimi-K3', tokenizer_mode='auto', tokenizer_backend='huggingface', tokenizer_worker_num=1, detokenizer_worker_num=1, skip_tokenizer_init=False, load_format='auto', model_loader_extra_config='{}', trust_remote_code=True, context_length=None, is_embedding=False, enable_multimodal=None, revision=None, model_impl='auto', model_config_parser='auto', json_model_override_args='{}', dtype='auto', quantization=None, quantization_param_path=None, kv_cache_dtype='auto', enable_fp32_lm_head=False, modelopt_quant=None, modelopt_checkpoint_restore_path=None, modelopt_checkpoint_save_path=None, modelopt_export_path=None, quantize_and_serve=False, rl_quant_profile=None, enable_tf32_matmul=False, mem_fraction_static=0.88, max_running_requests=None, max_queued_requests=None, max_total_tokens=None, chunked_prefill_size=8192, enable_dynamic_chunking=False, max_prefill_tokens=16384, prefill_max_requests=None, schedule_policy='fcfs', enable_priority_scheduling=False, disable_priority_preemption=False, default_priority_value=None, abort_on_priority_when_disabled=False, schedule_low_priority_values_first=False, priority_scheduling_preemption_threshold=10, retraction_policy='length', schedule_conservativeness=1.0, page_size=64, swa_full_tokens_ratio=0.8, disable_hybrid_swa_memory=False, radix_eviction_policy='lru', prefill_only_disable_kv_cache=False, disable_radix_cache=True, enable_page_major_kv_layout=False, enable_unified_memory=False, disable_chunked_prefix_cache=False, disable_overlap_schedule=True, num_continuous_decode_steps=1, scheduler_recv_interval=1, enable_mixed_chunk=False, nccl_port=None, dist_timeout=3600, dist_init_addr='174.1.60.1:20000', nnodes=4, node_rank=0, tp_size=4, dcp_size=1, pp_size=8, pp_max_micro_batch_size=None, pp_async_batch_depth=0, dp_size=1, load_balance_method='follow_bootstrap_room', attn_cp_size=1, moe_dp_size=1, dcp_comm_backend='ag_rs', dcp_replicate_q_proj=None, dwdp_size=1, enable_prefill_cp=False, cp_strategy=None, enable_dsa_cache_layer_split=False, enable_dsa_prefill_context_parallel=False, dsa_prefill_cp_mode='round-robin-split', enable_prefill_context_parallel=False, prefill_cp_mode='in-seq-split', enable_cp_decode_attn_tp=False, enable_dp_attention=False, enable_dp_attention_local_control_broadcast=False, enable_dp_lm_head=False, enable_attn_tp_input_scattered=False, disable_attn_tp_gather=False, enable_p2p_check=False, device='cuda', base_gpu_id=0, gpu_id_step=1, random_seed=984607766, watchdog_timeout=300, soft_watchdog_timeout=None, sleep_on_idle=False, use_ray=False, custom_sigquit_handler=None, numa_node=None, gc_threshold=None, host='0.0.0.0', port=30000, fastapi_root_path='', smg_grpc_mode=False, grpc_mode=False, grpc_port=None, sidecar=None, sidecar_args=None, skip_server_warmup=False, warmups=None, enable_http2=False, ssl_keyfile=None, ssl_certfile=None, ssl_ca_certs=None, ssl_keyfile_password=None, enable_ssl_refresh=False, api_key=None, admin_api_key=None, served_model_name='kimi-k3', weight_version='default', chat_template=None, hf_chat_template_name=None, completion_template=None, file_storage_path='sglang_storage', enable_cache_report=False, reasoning_parser=None, default_chat_template_kwargs=None, strip_thinking_cache=False, enable_strict_thinking=False, tool_call_parser=None, tool_server=None, sampling_defaults='model', asr_max_buffer_seconds=60, asr_max_concurrent_sessions=32, preferred_sampling_params=None, allow_auto_truncate=False, stream_interval=1, batch_notify_size=16, stream_response_default_include_usage=False, incremental_streaming_output=False, enable_streaming_session=False, enable_session_radix_cache=False, log_level='info', log_level_http=None, log_requests=False, log_requests_level=2, log_requests_format='text', log_requests_target=None, uvicorn_access_log_exclude_prefixes=[], crash_dump_folder=None, show_time_cost=False, enable_metrics=False, smg_http_sidecar_port=None, enable_mfu_metrics=False, enable_metrics_for_all_schedulers=False, load_snapshot_publish_interval=15, tokenizer_metrics_custom_labels_header='x-custom-labels', tokenizer_metrics_allowed_custom_labels=None, extra_metric_labels=None, bucket_time_to_first_token=None, bucket_inter_token_latency=None, bucket_e2e_request_latency=None, prompt_tokens_buckets=None, generation_tokens_buckets=None, gc_warning_threshold_secs=0.0, decode_log_interval=40, enable_request_time_stats_logging=False, kv_events_config=None, enable_forward_pass_metrics=False, forward_pass_metrics_worker_id='', forward_pass_metrics_ipc_name=None, enable_trace=False, trace_modules='request', otlp_traces_endpoint='localhost:4317', export_metrics_to_file=False, export_metrics_to_file_dir=None, stat_loggers=None, constrained_json_whitespace_pattern=None, constrained_json_disable_any_whitespace=False, attention_backend='triton', decode_attention_backend=None, prefill_attention_backend=None, sampling_backend='flashinfer', grammar_backend='xgrammar', radix_cache_backend=None, mm_attention_backend=None, fp8_gemm_runner_backend='auto', fp4_gemm_runner_backend='auto', bf16_gemm_backend='auto', dsa_prefill_backend=None, dsa_decode_backend=None, dsa_paged_mqa_logits_backend='auto', dsa_topk_backend='sgl-kernel', disable_flashinfer_autotune=False, flashinfer_autotune_skip_ops=None, mamba_backend='triton', cuda_graph_config=CudaGraphConfig(decode=PhaseConfig(backend='disabled', max_bs=16, bs=[1, 2, 4, 8, 12, 16], tc_compiler='eager', full_prefill_max_req=None), prefill=PhaseConfig(backend='disabled', max_bs=2048, bs=[4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], tc_compiler='eager', full_prefill_max_req=None)), cuda_graph_backend_decode=None, cuda_graph_backend_prefill=None, cuda_graph_max_bs_decode=16, cuda_graph_max_bs_prefill=None, cuda_graph_bs_decode=None, cuda_graph_bs_prefill=None, cuda_graph_tc_compiler=None, disable_prefill_cuda_graph=False, disable_decode_cuda_graph=False, disable_cuda_graph=False, disable_cuda_graph_padding=False, enable_profile_cuda_graph=False, enable_cudagraph_gc=False, debug_cuda_graph=False, enable_layerwise_nvtx_marker=False, enable_nccl_nvls=False, enable_symm_mem=False, triton_attention_reduce_in_fp32=False, triton_attention_num_kv_splits=8, triton_attention_split_tile_size=None, flashinfer_mla_disable_ragged=False, enable_fused_qk_norm_rope=False, enable_precise_embedding_interpolation=False, enable_fused_moe_sum_all_reduce=False, enable_deepseek_v4_fp4_indexer=False, disable_custom_all_reduce=False, enable_mscclpp=False, enable_torch_symm_mem=False, enable_scattered_sconv=False, pre_warm_nccl=False, enable_quant_communications=False, enable_flashinfer_allreduce_fusion=False, enforce_disable_flashinfer_allreduce_fusion=False, flashinfer_allreduce_fusion_backend=None, enable_aiter_allreduce_fusion=False, enable_torch_compile=False, enable_torch_compile_debug_mode=False, torch_compile_max_bs=32, torchao_config='', speculative_algorithm=None, speculative_draft_model_path=None, speculative_draft_model_revision=None, speculative_draft_load_format=None, speculative_num_steps=None, speculative_eagle_topk=None, speculative_num_draft_tokens=None, speculative_dflash_block_size=None, speculative_dspark_block_size=None, speculative_dspark_sps_table_path=None, speculative_dspark_confidence_sts_path=None, speculative_dspark_align_verify_tokens_to_graph_tier=False, speculative_accept_threshold_single=1.0, speculative_accept_threshold_acc=1.0, speculative_use_rejection_sampling=False, speculative_token_map=None, speculative_attention_mode='prefill', speculative_draft_attention_backend=None, speculative_draft_window_size=None, speculative_moe_runner_backend='flashinfer_mxfp4', speculative_moe_a2a_backend=None, speculative_draft_model_quantization=None, speculative_skip_dp_mlp_sync=False, enable_multi_layer_eagle=False, speculative_adaptive=False, speculative_adaptive_config=None, decoupled_spec_bind_endpoint=None, decoupled_spec_connect_endpoints=None, decoupled_spec_rank=None, decoupled_spec_role='null', spec_trace_dir=None, speculative_ngram_min_bfs_breadth=1, speculative_ngram_max_bfs_breadth=10, speculative_ngram_match_type='BFS', speculative_ngram_max_trie_depth=18, speculative_ngram_capacity=10000000, speculative_ngram_external_corpus_path=None, speculative_ngram_external_sam_budget=0, speculative_ngram_external_corpus_max_tokens=10000000, ep_size=4, moe_a2a_backend='none', moe_runner_backend='flashinfer_mxfp4', flashinfer_mxfp4_moe_precision='default', deepep_mode='auto', fuseep_mode=2, deepep_dispatcher_output_dtype='auto', ep_num_redundant_experts=0, ep_dispatch_algorithm=None, init_expert_location='trivial', enable_eplb=False, eplb_algorithm='auto', eplb_rebalance_num_iterations=1000, eplb_rebalance_layers_per_chunk=None, eplb_min_rebalancing_utilization_threshold=1.0, expert_distribution_recorder_mode=None, expert_distribution_recorder_buffer_size=1000, enable_expert_distribution_metrics=False, deepep_config=None, moe_dense_tp_size=None, elastic_ep_backend=None, enable_elastic_expert_backup=False, mooncake_ib_device=None, enable_waterfill=False, ep_join_mode=None, ep_join_rank_offset=0, elastic_ep_initial_size=None, max_ep_size=None, elastic_ep_scale_timeout=600, elastic_ep_rejoin=False, disable_flashinfer_cutlass_moe_fp4_allgather=False, disable_shared_experts_fusion=False, enforce_shared_experts_fusion=False, max_mamba_cache_size=None, mamba_ssm_dtype=None, mamba_max_states_per_path=-1, enable_mamba_cache_stochastic_rounding=False, mamba_cache_philox_rounds=0, mamba_full_memory_ratio=0.36, mamba_radix_cache_strategy='extra_buffer_lazy', uses_mamba_radix_cache=False, mamba_track_interval=256, enable_int8_mamba_checkpoint=False, int8_mamba_ckpt_size=None, linear_attn_backend='triton', linear_attn_decode_backend=None, linear_attn_prefill_backend=None, linear_attn_verify_backend=None, enable_linear_replayssm=False, linear_replayssm_cache_len=16, enable_linear_replayssm_spec=False, enable_hierarchical_cache=False, hicache_ratio=2.0, hicache_size=0, hicache_write_policy='write_through', hicache_io_backend='kernel', hicache_mem_layout='page_first', hicache_storage_backend=None, hicache_storage_prefetch_policy='timeout', hicache_storage_backend_extra_config=None, enable_hisparse=False, hisparse_config=None, enable_broadcast_mm_inputs_process=False, enable_prefix_mm_cache=False, mm_enable_dp_encoder=False, mm_process_config={}, mm_processor_worker_num=0, mm_io_worker_num=0, limit_mm_data_per_request=None, enable_mm_global_cache=False, disable_fast_image_processor=False, mm_feature_transport='cpu', keep_mm_feature_on_device=False, enable_lora=None, enable_lora_overlap_loading=None, max_lora_rank=None, lora_target_modules=None, lora_paths=None, max_loaded_loras=None, max_loras_per_batch=8, lora_eviction_policy='lru', lora_backend='csgmv', max_lora_chunk_size=16, experts_shared_outer_loras=None, lora_use_virtual_experts=False, lora_strict_loading=False, lora_drain_wait_threshold=0.0, enable_two_batch_overlap=False, enable_single_batch_overlap=False, tbo_token_distribution_threshold=0.48, cpu_offload_gb=0, offload_group_size=-1, offload_num_in_group=1, offload_prefetch_step=1, offload_mode='cpu', enable_lmcache=False, lmcache_config_file=None, enable_flexkv=False, flexkv_config_file=None, kt_weight_path=None, kt_method='AMXINT4', kt_cpuinfer=None, kt_threadpool_count=2, kt_num_gpu_experts=None, kt_max_deferred_experts_per_token=None, dllm_algorithm=None, dllm_algorithm_config=None, dllm_fdfo=True, disaggregation_mode='prefill', disaggregation_transfer_backend='mooncake', disaggregation_bootstrap_port=28800, disaggregation_ib_device='mlx5_0,mlx5_1,mlx5_2,mlx5_3', disaggregation_decode_enable_radix_cache=False, disaggregation_decode_enable_offload_kvcache=False, num_reserved_decode_tokens=512, disaggregation_decode_extra_slots=None, disaggregation_decode_polling_interval=1, optimistic_prefill_attempts=0, encoder_only=False, language_only=False, encoder_transfer_backend='zmq_to_tokenizer', encoder_urls=[], encoder_bootstrap_port=8997, encoder_register_urls=[], enable_adaptive_dispatch_to_encoder=False, enable_pdmux=False, pdmux_config_path=None, sm_group_num=8, custom_weight_loader=[], weight_loader_disable_mmap=False, weight_loader_prefetch_checkpoints=False, weight_loader_prefetch_num_threads=4, weight_loader_drop_cache_after_load=False, remote_instance_weight_loader_seed_instance_ip=None, remote_instance_weight_loader_seed_instance_service_port=None, remote_instance_weight_loader_send_weights_group_ports=None, remote_instance_weight_loader_backend='nccl', remote_instance_weight_loader_start_seed_via_transfer_engine=False, engine_info_bootstrap_port=6789, modelexpress_config=None, download_dir=None, model_checksum=None, delete_ckpt_after_loading=False, decrypted_config_file=None, decrypted_draft_config_file=None, checkpoint_engine_wait_weights_before_ready=False, enable_prefill_delayer=False, prefill_delayer_max_delay_passes=30, prefill_delayer_token_usage_low_watermark=None, prefill_delayer_forward_passes_buckets=None, prefill_delayer_wait_seconds_buckets=None, prefill_delayer_queue_min_ratio=None, prefill_delayer_max_delay_ms=None, min_free_slots_delay=None, enable_deterministic_inference=False, rl_on_policy_target=None, kv_canary='none', kv_canary_real_data='none', kv_canary_sweep_interval=0, enable_dynamic_batch_tokenizer=False, dynamic_batch_tokenizer_batch_size=32, dynamic_batch_tokenizer_batch_timeout=0.002, enable_tokenizer_batch_encode=False, disable_tokenizer_batch_decode=False, debug_tensor_dump_output_folder=None, debug_tensor_dump_layers=None, debug_tensor_dump_input_file=None, enable_memory_saver=False, enable_weights_cpu_backup=False, enable_draft_weights_cpu_backup=False, enable_custom_logit_processor=False, enable_return_hidden_states=False, enable_return_routed_experts=False, enable_return_indexer_topk=False, disable_outlines_disk_cache=False, enable_mis=False, weight_cache_mode='off', weight_cache_socket=None, weight_cache_timeout=1800, forward_hooks=None, msprobe_dump_config=None)
+[2026-08-27 04:28:58] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:58] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:58] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:58] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:58] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:58] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:58] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:58] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:01] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:01] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:01] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:03] Multimodal processor concurrency enabled with 2 isolated worker threads (auto).
+[2026-08-27 04:29:03] CommonKVBootstrapServer started successfully on 0.0.0.0:28800
+[2026-08-27 04:29:03] No HuggingFace chat template found
+[2026-08-27 04:29:03] No chat template found, defaulting to 'string' content format
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:15 PP0 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:15 PP0 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:16] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:16] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:16 PP1 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:16 PP1 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+/sgl-workspace/sglang/python/sglang/srt/utils/hf_transformers/tokenizer.py:423: UserWarning: Using a slow tokenizer. This might cause a significant slowdown. Consider using a fast tokenizer instead.
+  warnings.warn(
+[2026-08-27 04:29:16] Applying special tokens cache patch for Kimi tokenizer: <class 'transformers_modules.Kimi_hyphen_K3.caf40b2773f9029e.tokenization_kimi.TikTokenTokenizer'>
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:16 PP1 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:16 PP1 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:16 PP0 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:16 PP0 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:16 PP1 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:16 PP1 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:17 PP0 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:17 PP0 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:17 PP1 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:17 PP1 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:17 PP0 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:17 PP0 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:17 PP0 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:17.587481   237 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:17.587555   237 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:17.587566   237 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.1 port: 12001
+I0827 04:29:17.587602   237 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.1:16830
+I0827 04:29:17.587723   237 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:17.613691   237 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:17.620473   237 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:17.627439   237 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:17.653980   237 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:17.656663   237 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:17.656692   237 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:17.656746   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:17.663833   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:17.665359   237 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:01
+I0827 04:29:17.665387   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:17.690173   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:17.691654   237 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:01
+I0827 04:29:17.691689   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:17.718215   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:17.719650   237 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:01
+I0827 04:29:17.719686   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:17.746297   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:17.747779   237 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:01
+I0827 04:29:17.747891   237 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:17 PP0 TP1 EP1] Init torch distributed begin.
+[2026-08-27 04:29:17 PP1 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:18.219898   241 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:18.219964   241 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:18.219973   241 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.1 port: 12001
+I0827 04:29:18.220007   241 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.1:15425
+I0827 04:29:18.220109   241 transfer_engine_impl.cpp:258] Auto-discovering topology...
+[2026-08-27 04:29:18 PP1 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+I0827 04:29:18.244532   241 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:18.252967   241 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:18.261787   241 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:18.289944   241 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:18.293610   241 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:18.293633   241 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:18.293681   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.301677   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:18.303126   241 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:01
+I0827 04:29:18.303155   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.330181   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:18.331590   241 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:01
+I0827 04:29:18.331614   241 rdma_context.cpp:205] Using SIEVE endpoint store
+[2026-08-27 04:29:18 PP0 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+I0827 04:29:18.358213   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:18.359581   241 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:01
+I0827 04:29:18.359604   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.386204   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:18.387562   241 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:01
+I0827 04:29:18.387681   241 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:18 PP1 TP1 EP1] Init torch distributed begin.
+[2026-08-27 04:29:18 PP1 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:18.588883   242 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:18.588970   242 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:18.588984   242 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.1 port: 12001
+I0827 04:29:18.589022   242 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.1:15650
+I0827 04:29:18.589143   242 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:18.594419   242 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:18.599439   242 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:18.604553   242 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:18.609825   242 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:18.612036   242 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:18.612056   242 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:18.612097   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.618021   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:18.619308   242 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:01
+I0827 04:29:18.619333   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.646080   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:18.647465   242 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:01
+I0827 04:29:18.647495   242 rdma_context.cpp:205] Using SIEVE endpoint store
+[2026-08-27 04:29:18 PP0 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+I0827 04:29:18.674199   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:18.675598   242 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:01
+I0827 04:29:18.675621   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.682866   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:18.684276   242 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:01
+I0827 04:29:18.684391   242 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:18 PP1 TP2 EP2] Init torch distributed begin.
+[2026-08-27 04:29:18 PP1 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:18.695660   238 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:18.695755   238 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:18.695767   238 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.1 port: 12001
+I0827 04:29:18.695806   238 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.1:15123
+I0827 04:29:18.695909   238 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:18.701093   238 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:18.706161   238 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:18.711323   238 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:18.737910   238 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:18.740556   238 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:18.740602   238 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:18.740651   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.747717   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:18.749193   238 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:01
+I0827 04:29:18.749220   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.774145   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:18.775534   238 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:01
+I0827 04:29:18.775558   238 rdma_context.cpp:205] Using SIEVE endpoint store
+[2026-08-27 04:29:18 PP0 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+I0827 04:29:18.802196   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:18.803623   238 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:01
+I0827 04:29:18.803649   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.810976   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:18.812341   238 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:01
+I0827 04:29:18.812443   238 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:18 PP0 TP2 EP2] Init torch distributed begin.
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:19.084213   243 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:19.084332   243 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:19.084348   243 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.1 port: 12001
+I0827 04:29:19.084400   243 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.1:15411
+I0827 04:29:19.084602   243 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:19.114387   243 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:19.121773   243 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:19.146405   243 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:19.153759   243 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:19.156515   243 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:19.156548   243 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:19.156603   243 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.163982   243 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:19.165694   243 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:01
+I0827 04:29:19.165726   243 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.173307   243 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:19.174904   243 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:01
+I0827 04:29:19.174932   243 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.206454   243 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:19.208142   243 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:01
+I0827 04:29:19.208168   243 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.215909   243 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:19.217583   243 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:01
+I0827 04:29:19.217706   243 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:19 PP1 TP3 EP3] Init torch distributed begin.
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:19.239588   236 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:19.239699   236 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:19.239715   236 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.1 port: 12001
+I0827 04:29:19.239763   236 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.1:15605
+I0827 04:29:19.239933   236 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:19.240546   240 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:19.240644   240 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:19.240658   240 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.1 port: 12001
+I0827 04:29:19.240702   240 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.1:15487
+I0827 04:29:19.240862   240 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:19.242301   239 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:19.242408   239 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:19.242422   239 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.1 port: 12001
+I0827 04:29:19.242470   239 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.1:15552
+I0827 04:29:19.242636   239 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:19.246201   240 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:19.248451   239 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:19.251204   240 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:19.254226   239 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:19.256254   240 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:19.259852   239 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:19.261372   240 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:19.263411   240 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:19.263432   240 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:19.263473   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.265838   239 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:19.268059   239 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:19.268082   239 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:19.268126   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.269106   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:19.270298   236 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:19.270401   240 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:01
+I0827 04:29:19.270423   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.274480   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:19.275991   239 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:01
+I0827 04:29:19.276019   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.276877   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:19.278193   240 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:01
+I0827 04:29:19.278220   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.297820   236 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:19.302485   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:19.302623   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:19.303958   239 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:01
+I0827 04:29:19.303983   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.304121   240 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:01
+I0827 04:29:19.304147   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.311621   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:19.313341   240 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:01
+I0827 04:29:19.313478   240 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:19 PP1 TP0 EP0] Init torch distributed begin.
+I0827 04:29:19.322126   236 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:19.330281   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:19.331759   239 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:01
+I0827 04:29:19.331784   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.339274   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:19.340766   239 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:01
+I0827 04:29:19.340884   239 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:19 PP0 TP3 EP3] Init torch distributed begin.
+I0827 04:29:19.345958   236 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:19.348632   236 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:19.348659   236 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:19.348712   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.355962   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:19.417273   236 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:01
+I0827 04:29:19.417309   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.424894   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:19.426437   236 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:01
+I0827 04:29:19.426491   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.434231   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:19.435868   236 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:01
+I0827 04:29:19.435896   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.443578   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:19.445148   236 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:01
+I0827 04:29:19.445264   236 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:19 PP0 TP0 EP0] Init torch distributed begin.
+[2026-08-27 04:29:26 PP0 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:26 PP1 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:27 PP1 TP2 EP2] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP1 TP0 EP0] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP1 TP3 EP3] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP1 TP1 EP1] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP0 TP3 EP3] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP0 TP0 EP0] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP0 TP1 EP1] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP0 TP2 EP2] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP0 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:27 PP0 TP3 EP3] sglang is using nccl==2.28.9
+[2026-08-27 04:29:27 PP0 TP1 EP1] sglang is using nccl==2.28.9
+[2026-08-27 04:29:27 PP0 TP2 EP2] sglang is using nccl==2.28.9
+[2026-08-27 04:29:27 PP0 TP0 EP0] Init torch distributed ends. elapsed=8.22 s, mem usage=0.46 GB
+[2026-08-27 04:29:27 PP1 TP3 EP3] Init torch distributed ends. elapsed=8.45 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP1 TP1 EP1] Init torch distributed ends. elapsed=9.28 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP1 TP2 EP2] Init torch distributed ends. elapsed=8.98 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP1 TP0 EP0] Init torch distributed ends. elapsed=8.35 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP0 TP2 EP2] Init torch distributed ends. elapsed=8.85 s, mem usage=0.46 GB
+[2026-08-27 04:29:27 PP0 TP3 EP3] Init torch distributed ends. elapsed=8.32 s, mem usage=0.46 GB
+[2026-08-27 04:29:27 PP0 TP1 EP1] Init torch distributed ends. elapsed=9.92 s, mem usage=0.46 GB
+[2026-08-27 04:29:27 PP1 TP0 EP0] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP1 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP0 TP0 EP0] Load weight begin. avail mem=82.12 GB
+[2026-08-27 04:29:27 PP1 TP3 EP3] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP1 TP2 EP2] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP0 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP1 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP1 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP1 TP1 EP1] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP0 TP2 EP2] Load weight begin. avail mem=82.12 GB
+[2026-08-27 04:29:27 PP1 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP0 TP3 EP3] Load weight begin. avail mem=82.12 GB
+[2026-08-27 04:29:27 PP0 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP0 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP1 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:27 PP1 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:27 PP0 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:27 PP1 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:27 PP1 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:27 PP1 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:27 PP0 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:27 PP1 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:27 PP1 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:27 PP1 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:27 PP0 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:27 PP0 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:27 PP0 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:27 PP0 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:27 PP0 TP1 EP1] Load weight begin. avail mem=82.12 GB
+[2026-08-27 04:29:27 PP0 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP0 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP0 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+
Multi-thread loading shards:   0% Completed | 0/96 [00:00<?, ?it/s]
Multi-thread loading shards:   1% Completed | 1/96 [00:00<01:01,  1.55it/s]
Multi-thread loading shards:   2% Completed | 2/96 [00:09<08:49,  5.63s/it]
Multi-thread loading shards:   3% Completed | 3/96 [00:12<06:35,  4.25s/it]
Multi-thread loading shards:   4% Completed | 4/96 [00:14<05:26,  3.55s/it]
Multi-thread loading shards:   5% Completed | 5/96 [00:17<04:43,  3.11s/it]
Multi-thread loading shards:   6% Completed | 6/96 [00:19<04:17,  2.86s/it]
Multi-thread loading shards:   7% Completed | 7/96 [00:21<04:00,  2.71s/it]
Multi-thread loading shards:   8% Completed | 8/96 [00:24<03:49,  2.61s/it]
Multi-thread loading shards:   9% Completed | 9/96 [00:26<03:44,  2.58s/it]
Multi-thread loading shards:  10% Completed | 10/96 [00:29<03:40,  2.56s/it]
Multi-thread loading shards:  11% Completed | 11/96 [00:31<03:36,  2.54s/it]
Multi-thread loading shards:  12% Completed | 12/96 [00:32<02:43,  1.95s/it]
Multi-thread loading shards:  14% Completed | 13/96 [00:32<02:05,  1.52s/it]
Multi-thread loading shards:  15% Completed | 14/96 [00:33<01:33,  1.14s/it]
Multi-thread loading shards:  16% Completed | 15/96 [00:33<01:21,  1.01s/it]
Multi-thread loading shards:  17% Completed | 16/96 [00:34<01:19,  1.00it/s]
Multi-thread loading shards:  18% Completed | 17/96 [00:35<01:02,  1.26it/s]
Multi-thread loading shards:  19% Completed | 18/96 [00:35<00:51,  1.53it/s]
Multi-thread loading shards:  20% Completed | 19/96 [00:35<00:42,  1.80it/s]
Multi-thread loading shards:  21% Completed | 20/96 [00:36<00:37,  2.01it/s]
Multi-thread loading shards:  22% Completed | 21/96 [00:36<00:33,  2.23it/s]
Multi-thread loading shards:  23% Completed | 22/96 [00:36<00:30,  2.43it/s]
Multi-thread loading shards:  24% Completed | 23/96 [00:37<00:27,  2.68it/s]
Multi-thread loading shards:  25% Completed | 24/96 [00:37<00:25,  2.80it/s]
Multi-thread loading shards:  26% Completed | 25/96 [00:37<00:24,  2.89it/s]
Multi-thread loading shards:  27% Completed | 26/96 [00:38<00:23,  2.99it/s]
Multi-thread loading shards:  28% Completed | 27/96 [00:38<00:22,  3.01it/s]
Multi-thread loading shards:  29% Completed | 28/96 [00:38<00:22,  2.99it/s]
Multi-thread loading shards:  30% Completed | 29/96 [00:39<00:22,  2.98it/s]
Multi-thread loading shards:  31% Completed | 30/96 [00:39<00:22,  2.98it/s]
Multi-thread loading shards:  32% Completed | 31/96 [00:39<00:21,  3.01it/s]
Multi-thread loading shards:  33% Completed | 32/96 [00:40<00:21,  3.00it/s]
Multi-thread loading shards:  34% Completed | 33/96 [00:41<00:35,  1.79it/s]
Multi-thread loading shards:  35% Completed | 34/96 [00:41<00:30,  2.02it/s]
Multi-thread loading shards:  36% Completed | 35/96 [00:41<00:27,  2.23it/s]
Multi-thread loading shards:  38% Completed | 36/96 [00:42<00:24,  2.41it/s]
Multi-thread loading shards:  39% Completed | 37/96 [00:42<00:22,  2.57it/s]
Multi-thread loading shards:  40% Completed | 38/96 [00:42<00:21,  2.64it/s]
Multi-thread loading shards:  41% Completed | 39/96 [00:43<00:20,  2.73it/s]
Multi-thread loading shards:  42% Completed | 40/96 [00:43<00:20,  2.80it/s]
Multi-thread loading shards:  43% Completed | 41/96 [00:43<00:19,  2.79it/s]
Multi-thread loading shards:  44% Completed | 42/96 [00:44<00:19,  2.80it/s]
Multi-thread loading shards:  45% Completed | 43/96 [00:44<00:22,  2.39it/s]
Multi-thread loading shards:  46% Completed | 44/96 [00:45<00:20,  2.51it/s]
Multi-thread loading shards:  47% Completed | 45/96 [00:45<00:19,  2.64it/s]
Multi-thread loading shards:  48% Completed | 46/96 [00:45<00:18,  2.67it/s]
Multi-thread loading shards:  49% Completed | 47/96 [00:46<00:18,  2.61it/s]
Multi-thread loading shards:  50% Completed | 48/96 [00:46<00:17,  2.72it/s]
Multi-thread loading shards:  51% Completed | 49/96 [00:47<00:16,  2.84it/s]
Multi-thread loading shards:  52% Completed | 50/96 [00:47<00:16,  2.81it/s]
Multi-thread loading shards:  53% Completed | 51/96 [00:47<00:15,  2.93it/s]
Multi-thread loading shards:  54% Completed | 52/96 [00:48<00:21,  2.05it/s]
Multi-thread loading shards:  55% Completed | 53/96 [00:49<00:33,  1.30it/s]
Multi-thread loading shards:  56% Completed | 54/96 [00:50<00:27,  1.55it/s]
Multi-thread loading shards:  57% Completed | 55/96 [00:50<00:22,  1.82it/s]
Multi-thread loading shards:  58% Completed | 56/96 [00:50<00:19,  2.06it/s]
Multi-thread loading shards:  59% Completed | 57/96 [00:51<00:17,  2.25it/s]
Multi-thread loading shards:  60% Completed | 58/96 [00:51<00:15,  2.40it/s]
Multi-thread loading shards:  61% Completed | 59/96 [00:51<00:14,  2.56it/s]
Multi-thread loading shards:  62% Completed | 60/96 [00:52<00:13,  2.67it/s]
Multi-thread loading shards:  64% Completed | 61/96 [00:52<00:12,  2.77it/s]
Multi-thread loading shards:  65% Completed | 62/96 [00:52<00:11,  2.84it/s]
Multi-thread loading shards:  66% Completed | 63/96 [00:53<00:11,  2.89it/s]
Multi-thread loading shards:  67% Completed | 64/96 [00:53<00:11,  2.87it/s]
Multi-thread loading shards:  68% Completed | 65/96 [00:54<00:10,  2.89it/s]
Multi-thread loading shards:  69% Completed | 66/96 [00:54<00:10,  2.91it/s]
Multi-thread loading shards:  70% Completed | 67/96 [00:54<00:10,  2.90it/s]
Multi-thread loading shards:  71% Completed | 68/96 [00:55<00:09,  2.90it/s]
Multi-thread loading shards:  72% Completed | 69/96 [00:55<00:09,  2.89it/s]
Multi-thread loading shards:  73% Completed | 70/96 [00:56<00:14,  1.85it/s]
Multi-thread loading shards:  74% Completed | 71/96 [00:56<00:11,  2.10it/s]
Multi-thread loading shards:  75% Completed | 72/96 [00:57<00:10,  2.23it/s]
Multi-thread loading shards:  76% Completed | 73/96 [00:57<00:09,  2.39it/s]
Multi-thread loading shards:  77% Completed | 74/96 [00:57<00:08,  2.48it/s]
Multi-thread loading shards:  78% Completed | 75/96 [00:58<00:09,  2.15it/s]
Multi-thread loading shards:  79% Completed | 76/96 [00:59<00:10,  1.89it/s]
Multi-thread loading shards:  80% Completed | 77/96 [00:59<00:09,  2.05it/s]
Multi-thread loading shards:  81% Completed | 78/96 [00:59<00:08,  2.24it/s]
Multi-thread loading shards:  82% Completed | 79/96 [01:00<00:07,  2.39it/s]
Multi-thread loading shards:  83% Completed | 80/96 [01:00<00:06,  2.51it/s]
Multi-thread loading shards:  84% Completed | 81/96 [01:00<00:05,  2.65it/s]
Multi-thread loading shards:  85% Completed | 82/96 [01:01<00:04,  2.84it/s]
Multi-thread loading shards:  86% Completed | 83/96 [01:02<00:06,  1.99it/s]
Multi-thread loading shards:  88% Completed | 84/96 [01:02<00:06,  1.99it/s]
Multi-thread loading shards:  90% Completed | 86/96 [01:02<00:03,  3.27it/s]
Multi-thread loading shards:  92% Completed | 88/96 [01:02<00:01,  4.72it/s]
Multi-thread loading shards:  94% Completed | 90/96 [01:02<00:00,  6.49it/s]
Multi-thread loading shards:  96% Completed | 92/96 [01:03<00:00,  8.37it/s]
Multi-thread loading shards:  98% Completed | 94/96 [01:03<00:00,  8.91it/s]
Multi-thread loading shards: 100% Completed | 96/96 [01:03<00:00,  9.35it/s]
Multi-thread loading shards: 100% Completed | 96/96 [01:03<00:00,  1.51it/s]
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:33 PP0 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:33 PP0 TP2 EP2] Load weight end. elapsed=65.63 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=39.03 GB, mem usage=43.09 GB.
+[2026-08-27 04:30:33 PP0 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:33 PP0 TP0 EP0] Load weight end. elapsed=65.81 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=39.03 GB, mem usage=43.09 GB.
+[2026-08-27 04:30:35 PP0 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:35 PP0 TP3 EP3] Load weight end. elapsed=67.37 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=39.03 GB, mem usage=43.09 GB.
+[2026-08-27 04:30:35 PP0 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:35 PP0 TP1 EP1] Load weight end. elapsed=67.46 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=39.03 GB, mem usage=43.09 GB.
+[2026-08-27 04:30:35 PP0 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:35 PP0 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:35 PP0 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:35 PP0 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:35 PP0 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:35 PP0 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:35 PP0 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:35 PP0 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:36 PP1 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:36 PP1 TP3 EP3] Load weight end. elapsed=68.32 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=36.12 GB, mem usage=45.92 GB.
+[2026-08-27 04:30:36 PP1 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP1 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP1 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP1 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP1 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP1 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP1 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP1 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:37 PP1 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:37 PP1 TP1 EP1] Load weight end. elapsed=69.55 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=36.12 GB, mem usage=45.92 GB.
+[2026-08-27 04:30:37 PP1 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:37 PP1 TP0 EP0] Load weight end. elapsed=69.66 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=36.12 GB, mem usage=45.92 GB.
+[2026-08-27 04:30:39 PP1 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:39 PP1 TP2 EP2] Load weight end. elapsed=71.33 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=36.12 GB, mem usage=45.92 GB.
+[2026-08-27 04:30:39 PP1 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:39 PP1 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:39 PP1 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:39 PP1 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:39 PP1 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:39 PP1 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:39 PP1 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:39 PP1 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:53 PP0 TP0 EP0] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:53 PP1 TP3 EP3] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:53 PP1 TP2 EP2] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:53 PP1 TP1 EP1] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:53 PP1 TP0 EP0] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:53 PP0 TP2 EP2] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:53 PP0 TP3 EP3] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:53 PP0 TP1 EP1] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:53 PP0 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:53 PP1 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:53 PP1 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:53 PP1 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:53 PP0 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:53 PP1 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:53 PP0 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:53 PP0 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:53 PP0 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 8.16 GB
+[2026-08-27 04:30:53 PP0 TP0 EP0] Memory pool end. avail mem=30.14 GB
+[2026-08-27 04:30:53 PP0 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 8.16 GB
+[2026-08-27 04:30:53 PP0 TP2 EP2] Memory pool end. avail mem=30.14 GB
+[2026-08-27 04:30:53 PP1 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:53 PP1 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:53 PP1 TP0 EP0] Memory pool end. avail mem=23.24 GB
+[2026-08-27 04:30:53 PP1 TP2 EP2] Memory pool end. avail mem=23.24 GB
+[2026-08-27 04:30:53 PP0 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 8.16 GB
+[2026-08-27 04:30:53 PP1 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:53 PP0 TP3 EP3] Memory pool end. avail mem=30.14 GB
+[2026-08-27 04:30:53 PP1 TP1 EP1] Memory pool end. avail mem=23.24 GB
+[2026-08-27 04:30:53 PP1 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:53 PP0 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 8.16 GB
+[2026-08-27 04:30:53 PP1 TP3 EP3] Memory pool end. avail mem=23.24 GB
+[2026-08-27 04:30:53 PP0 TP1 EP1] Memory pool end. avail mem=30.14 GB
+[2026-08-27 04:30:54 PP0 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:30:54 PP0 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:30:54 PP0 TP0 EP0] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp0_pp0_dp0.json
+[2026-08-27 04:30:54 PP1 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:30:54 PP1 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:30:54 PP0 TP2 EP2] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp2_pp0_dp0.json
+[2026-08-27 04:30:54 PP1 TP2 EP2] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp2_pp1_dp0.json
+[2026-08-27 04:30:54 PP1 TP0 EP0] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp0_pp1_dp0.json
+[2026-08-27 04:30:54 PP1 TP1 EP1] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp1_pp1_dp0.json
+[2026-08-27 04:30:54 PP1 TP3 EP3] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp3_pp1_dp0.json
+[2026-08-27 04:30:54 PP0 TP1 EP1] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp1_pp0_dp0.json
+[2026-08-27 04:30:54 PP0 TP3 EP3] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp3_pp0_dp0.json
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.46profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.32profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.51profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  6.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.09profile/s]
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.07profile/s]
+[TensorRT-LLM][INFO] Set logger level to INFO
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.48profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.11profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.29profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.37profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.16profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.82profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.18profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.58profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:01<00:02,  1.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.66profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.03profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.49profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:01<00:01,  2.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.25profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:01,  2.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:02<00:00,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:02<00:00,  3.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.20profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.01profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  9.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  9.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  9.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 12.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 12.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 11.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 11.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.78profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.16profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  3.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.75profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.50profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.92profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.68profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  3.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.97profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.87profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 10.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.51profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.74profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 11.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.30profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.70profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.72profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.63profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.53profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.29profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.58profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  8.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 10.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.71profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.62profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.43profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.18profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.29profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.37profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.53profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.37profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.35profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.00profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.71profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.60profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.78profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.79profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  8.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.86profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  8.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.16profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.78profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  3.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.15profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.94profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.93profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.25profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.23profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.97profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.79profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.96profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.64profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.12profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.75profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.12profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.25profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.91profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.05profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.66profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.90profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  9.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.73profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.55profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.49profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.13profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.74profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.17profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  3.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.72profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.57profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.59profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.64profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.83profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 13.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.49profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 12.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  9.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  7.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.32profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.93profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.29profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.72profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.75profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.46profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.51profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  3.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.90profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.16profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.72profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.63profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.83profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.70profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.59profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.55profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  6.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.93profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.33profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.75profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.30profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  6.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  7.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.79profile/s]
+[2026-08-27 04:31:21 PP0 TP1 EP1] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.53profile/s][2026-08-27 04:31:21 PP0 TP0 EP0] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.97profile/s]
+[2026-08-27 04:31:21 PP0 TP3 EP3] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.43profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.06profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s][2026-08-27 04:31:21 PP0 TP2 EP2] FlashInfer autotune completed.
+[2026-08-27 04:31:21 PP0 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:21 PP0 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:21 PP0 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:21 PP0 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:21 PP0 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:21 PP0 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:21 PP0 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:21 PP0 TP0 EP0] max_total_num_tokens=3801408, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=54, context_len=1048576, available_gpu_mem=27.35 GB
+[2026-08-27 04:31:21 PP0 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.86profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.58profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:01<00:07,  1.18s/profile]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  2.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.27profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.55profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  3.96profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.86profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  6.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  7.01profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.23profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.05profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  6.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  7.06profile/s]
+[2026-08-27 04:31:26 PP1 TP1 EP1] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.40profile/s]
+[2026-08-27 04:31:26 PP1 TP2 EP2] FlashInfer autotune completed.
+[2026-08-27 04:31:26 PP1 TP0 EP0] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.81profile/s]
+[2026-08-27 04:31:26 PP1 TP3 EP3] FlashInfer autotune completed.
+[2026-08-27 04:31:26 PP1 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:26 PP1 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:26 PP1 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:26 PP1 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:26 PP1 TP0 EP0] max_total_num_tokens=3801408, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=54, context_len=1048576, available_gpu_mem=20.45 GB
+[2026-08-27 04:31:26 PP1 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:26 PP1 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:26 PP1 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:26 PP1 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+W0827 04:31:30.805795  2607 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fe48940, len: 6912: Operation not permitted [1]
+W0827 04:31:30.805797  2608 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4f684400, len: 6912: Operation not permitted [1]
+W0827 04:31:30.805945  2609 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4f962700, len: 6912: Operation not permitted [1]
+W0827 04:31:30.805979  2610 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4f9e8a00, len: 6912: Operation not permitted [1]
+W0827 04:31:30.806130  2612 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fe5f640, len: 55296: Operation not permitted [1]
+W0827 04:31:30.806133  2611 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4f83f000, len: 55296: Operation not permitted [1]
+W0827 04:31:30.806264  2613 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fe328c0, len: 6912: Operation not permitted [1]
+W0827 04:31:30.806334  2614 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fa75cc0, len: 13824: Operation not permitted [1]
+W0827 04:31:30.806449  2615 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fa87f80, len: 6912: Operation not permitted [1]
+W0827 04:31:30.806493  2616 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fa29a80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.203712  2635 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50c65e80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.203859  2636 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50c6bb40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.203975  2637 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50d1ffc0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.204025  2638 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50a0c300, len: 6912: Operation not permitted [1]
+W0827 04:31:31.204075  2640 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50de1940, len: 55296: Operation not permitted [1]
+W0827 04:31:31.204082  2639 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50dc93c0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.204438  2641 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50c87940, len: 6912: Operation not permitted [1]
+W0827 04:31:31.204561  2642 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50cffb00, len: 13824: Operation not permitted [1]
+W0827 04:31:31.204641  2644 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50dc7380, len: 6912: Operation not permitted [1]
+W0827 04:31:31.204722  2643 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50c243c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.323861  2661 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50b13c40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.323863  2662 memory_location.cpp:73] Failed to get NUMA node, addr: 0x511191c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.323912  2663 memory_location.cpp:73] Failed to get NUMA node, addr: 0x511b37c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.324069  2664 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50b2da80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.324110  2666 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5115c980, len: 55296: Operation not permitted [1]
+W0827 04:31:31.324155  2665 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5116ac40, len: 55296: Operation not permitted [1]
+W0827 04:31:31.324479  2667 memory_location.cpp:73] Failed to get NUMA node, addr: 0x511a7e80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.324587  2668 memory_location.cpp:73] Failed to get NUMA node, addr: 0x511b8cc0, len: 13824: Operation not permitted [1]
+W0827 04:31:31.324621  2669 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50fc6800, len: 6912: Operation not permitted [1]
+W0827 04:31:31.324682  2670 memory_location.cpp:73] Failed to get NUMA node, addr: 0x511bc780, len: 6912: Operation not permitted [1]
+W0827 04:31:31.543082  2689 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4902e440, len: 6912: Operation not permitted [1]
+W0827 04:31:31.543203  2690 memory_location.cpp:73] Failed to get NUMA node, addr: 0x49484200, len: 6912: Operation not permitted [1]
+W0827 04:31:31.543314  2691 memory_location.cpp:73] Failed to get NUMA node, addr: 0x49119600, len: 6912: Operation not permitted [1]
+W0827 04:31:31.543375  2692 memory_location.cpp:73] Failed to get NUMA node, addr: 0x493dc780, len: 6912: Operation not permitted [1]
+W0827 04:31:31.543467  2693 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51ac5fc0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.543601  2694 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51a9e780, len: 55296: Operation not permitted [1]
+W0827 04:31:31.543751  2695 memory_location.cpp:73] Failed to get NUMA node, addr: 0x49149a40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.543912  2696 memory_location.cpp:73] Failed to get NUMA node, addr: 0x518c0840, len: 13824: Operation not permitted [1]
+W0827 04:31:31.543947  2697 memory_location.cpp:73] Failed to get NUMA node, addr: 0x49303540, len: 6912: Operation not permitted [1]
+W0827 04:31:31.543990  2698 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51951a80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681525  2717 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50376e40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681525  2718 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5058ee40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681525  2716 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5008b980, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681548  2715 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fe0e840, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681727  2719 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5040eb80, len: 55296: Operation not permitted [1]
+W0827 04:31:31.681779  2724 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50209ac0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681811  2720 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5041c3c0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.681828  2722 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5048cc40, len: 13824: Operation not permitted [1]
+W0827 04:31:31.681844  2723 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fde7a00, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681851  2721 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50386580, len: 6912: Operation not permitted [1]
+W0827 04:31:31.730942  2745 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fce4240, len: 6912: Operation not permitted [1]
+W0827 04:31:31.730918  2744 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fed7d40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.730950  2743 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fdcaf80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.731007  2748 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5074c400, len: 55296: Operation not permitted [1]
+W0827 04:31:31.731073  2747 memory_location.cpp:73] Failed to get NUMA node, addr: 0x507bd9c0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.731086  2746 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4f931480, len: 6912: Operation not permitted [1]
+W0827 04:31:31.731294  2749 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd958c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.731361  2751 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd31d80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.731361  2750 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4ffabe00, len: 13824: Operation not permitted [1]
+W0827 04:31:31.731566  2752 memory_location.cpp:73] Failed to get NUMA node, addr: 0x508527c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.808167  2778 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd3ce40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.808190  2777 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd20740, len: 6912: Operation not permitted [1]
+W0827 04:31:31.808192  2776 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4ff86c80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.808637  2782 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fe88780, len: 6912: Operation not permitted [1]
+W0827 04:31:31.808724  2780 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fe48c80, len: 55296: Operation not permitted [1]
+W0827 04:31:31.808760  2781 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fe564c0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.809056  2785 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4ffa7e40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.809049  2779 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4feaa740, len: 6912: Operation not permitted [1]
+W0827 04:31:31.809075  2784 memory_location.cpp:73] Failed to get NUMA node, addr: 0x502fabc0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.809239  2783 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4ffa1f80, len: 13824: Operation not permitted [1]
+W0827 04:31:31.851475  2807 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4ff6fcc0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.851596  2808 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fc0a280, len: 6912: Operation not permitted [1]
+W0827 04:31:31.851615  2809 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd7d700, len: 6912: Operation not permitted [1]
+W0827 04:31:31.851732  2810 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd761c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.851951  2811 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fde6cc0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.851970  2812 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4feeb000, len: 55296: Operation not permitted [1]
+W0827 04:31:31.851962  2813 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fbe2680, len: 6912: Operation not permitted [1]
+W0827 04:31:31.851987  2816 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fbf07c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.852048  2815 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4ffa9940, len: 6912: Operation not permitted [1]
+W0827 04:31:31.852046  2814 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd235c0, len: 13824: Operation not permitted [1]
+[2026-08-27 04:31:32] INFO:     Started server process [1]
+[2026-08-27 04:31:32] INFO:     Waiting for application startup.
+[2026-08-27 04:31:32] Calling super().encode with {'add_special_tokens': False}
+[2026-08-27 04:31:32] Calling super().encode with {'add_special_tokens': False}
+[2026-08-27 04:31:32] INFO:     Application startup complete.
+[2026-08-27 04:31:32] INFO:     Uvicorn running on http://0.0.0.0:30000 (Press CTRL+C to quit)
+[2026-08-27 04:31:33] INFO:     127.0.0.1:60348 - "GET /model_info HTTP/1.1" 200 OK
+[2026-08-27 04:31:33] Start of pd disaggregation warmup ...
+[2026-08-27 04:31:37] INFO:     174.1.60.1:33088 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:38] INFO:     174.1.60.1:33092 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:39] INFO:     174.1.60.1:33106 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:40] INFO:     174.1.60.1:33110 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:41] INFO:     174.1.60.1:42028 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:42] INFO:     174.1.60.1:42030 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:43] INFO:     174.1.60.1:42046 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:44] INFO:     174.1.60.1:42050 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:45] INFO:     174.1.60.1:42060 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:46] INFO:     174.1.60.1:42072 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:47] INFO:     174.1.60.1:42086 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:48] INFO:     174.1.60.1:42092 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:49] INFO:     174.1.60.1:42098 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:50] INFO:     174.1.60.1:42100 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:51] INFO:     174.1.60.1:46852 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:52] INFO:     174.1.60.1:46856 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:53] INFO:     174.1.60.1:46872 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:54] INFO:     174.1.60.1:46876 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:55] INFO:     174.1.60.1:46886 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:56] INFO:     174.1.60.1:46890 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:57] INFO:     174.1.60.1:46898 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:58] INFO:     174.1.60.1:46912 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:59] INFO:     174.1.60.1:46928 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:00] INFO:     174.1.60.1:46936 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:01] INFO:     174.1.60.1:57672 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:02] INFO:     174.1.60.1:57676 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:03] INFO:     174.1.60.1:57686 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:04] INFO:     174.1.60.1:57688 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:05] INFO:     174.1.60.1:57696 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:06] INFO:     174.1.60.1:57708 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:12] INFO:     174.1.60.1:39714 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:13] INFO:     174.1.60.1:39728 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:14] INFO:     174.1.60.1:39736 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:15] INFO:     174.1.60.1:39746 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:16] INFO:     174.1.60.1:39760 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:17] INFO:     174.1.60.1:39772 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:18] INFO:     174.1.60.1:39782 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:19] INFO:     174.1.60.1:39786 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:20] INFO:     174.1.60.1:39800 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:21] INFO:     174.1.60.1:41036 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:22] INFO:     174.1.60.1:41052 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:23] INFO:     174.1.60.1:41058 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:24] INFO:     174.1.60.1:41068 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:25] INFO:     174.1.60.1:41074 - "GET /health HTTP/1.1" 503 Service Unavailable
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank1]:[W827 04:32:26.019157362 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank0]:[W827 04:32:26.019963648 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank3]:[W827 04:32:26.020028119 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank2]:[W827 04:32:26.020057305 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank6]:[W827 04:32:26.045976535 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank7]:[W827 04:32:26.046176399 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank5]:[W827 04:32:26.048033943 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank4]:[W827 04:32:26.048331224 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[2026-08-27 04:32:26] INFO:     174.1.60.1:41078 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:27] INFO:     174.1.60.1:41090 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:28] INFO:     174.1.60.1:41102 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:29] INFO:     174.1.60.1:41104 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:30] INFO:     174.1.60.1:41120 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:31] INFO:     174.1.60.1:57002 - "GET /health HTTP/1.1" 503 Service Unavailable
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank6]:[W827 04:32:31.709756552 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank7]:[W827 04:32:31.709839299 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank5]:[W827 04:32:31.709991082 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank4]:[W827 04:32:31.710123265 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[2026-08-27 04:32:32] INFO:     174.1.60.1:57016 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:33] INFO:     174.1.60.1:57030 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:34] INFO:     174.1.60.1:57046 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:35] INFO:     174.1.60.1:57056 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:36] INFO:     174.1.60.1:57066 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:37] INFO:     174.1.60.1:57070 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:38] INFO:     174.1.60.1:57072 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:39] INFO:     174.1.60.1:57082 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:40] INFO:     174.1.60.1:57098 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:41] INFO:     174.1.60.1:56606 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:50] INFO:     174.1.60.1:56620 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:51] INFO:     174.1.60.1:45658 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:52] INFO:     174.1.60.1:45672 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:53] INFO:     174.1.60.1:45674 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:54] INFO:     174.1.60.1:45690 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:55] INFO:     174.1.60.1:45704 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:56] INFO:     174.1.60.1:45708 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:57] INFO:     174.1.60.1:45714 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:58] INFO:     174.1.60.1:45722 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:00] INFO:     174.1.60.1:45736 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:01] INFO:     174.1.60.1:45746 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:02] INFO:     174.1.60.1:33310 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:03] INFO:     174.1.60.1:33320 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:04] INFO:     174.1.60.1:33324 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:05] INFO:     174.1.60.1:33328 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:06] INFO:     174.1.60.1:33336 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:07] INFO:     174.1.60.1:33350 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:08] INFO:     174.1.60.1:33354 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:09] INFO:     174.1.60.1:33356 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:10] INFO:     174.1.60.1:33368 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:11] INFO:     174.1.60.1:39318 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:12] INFO:     174.1.60.1:39328 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:13] INFO:     174.1.60.1:39336 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:14] INFO:     174.1.60.1:39344 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:15] INFO:     174.1.60.1:39346 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:16] INFO:     174.1.60.1:39352 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:17] INFO:     174.1.60.1:39364 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:18] INFO:     174.1.60.1:39378 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:19] INFO:     174.1.60.1:39390 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:20] INFO:     174.1.60.1:39402 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:25] INFO:     174.1.60.1:37792 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:26] INFO:     174.1.60.1:37804 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:27] INFO:     174.1.60.1:37812 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:28] INFO:     174.1.60.1:37826 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:29] INFO:     174.1.60.1:37830 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:30] INFO:     174.1.60.1:37844 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:31] INFO:     174.1.60.1:44208 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:32] INFO:     174.1.60.1:44220 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:33] INFO:     174.1.60.1:44226 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:34] INFO:     174.1.60.1:44228 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:35] INFO:     174.1.60.1:44238 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:37] INFO:     174.1.60.1:44248 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:38] INFO:     174.1.60.1:44258 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:39] INFO:     174.1.60.1:44272 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:40] INFO:     174.1.60.1:44276 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:41] INFO:     174.1.60.1:44194 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:42] INFO:     174.1.60.1:44198 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:43] INFO:     174.1.60.1:44212 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:44] INFO:     174.1.60.1:44216 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:45] INFO:     174.1.60.1:44222 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:46] INFO:     174.1.60.1:44236 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:47] INFO:     174.1.60.1:44246 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:48] INFO:     174.1.60.1:44252 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:49] INFO:     174.1.60.1:44264 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:50] INFO:     174.1.60.1:44278 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:51] INFO:     174.1.60.1:35728 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:52] INFO:     174.1.60.1:35736 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:53] INFO:     174.1.60.1:35752 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:54] INFO:     174.1.60.1:35754 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:55] INFO:     174.1.60.1:35758 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:04] INFO:     174.1.60.1:44514 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:05] INFO:     174.1.60.1:44518 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:06] INFO:     174.1.60.1:44520 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:07] INFO:     174.1.60.1:44526 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:08] INFO:     174.1.60.1:44530 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:09] INFO:     174.1.60.1:44546 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:10] INFO:     174.1.60.1:44550 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:11] INFO:     174.1.60.1:45098 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:12] INFO:     127.0.0.1:59722 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:12] INFO:     174.1.60.1:45106 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:13] INFO:     174.1.60.1:45122 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:14] INFO:     174.1.60.1:45124 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:15] INFO:     174.1.60.1:45136 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:16] INFO:     174.1.60.1:45146 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:17] INFO:     174.1.60.1:45152 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:19] INFO:     174.1.60.1:45158 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:20] INFO:     174.1.60.1:45170 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:21] INFO:     174.1.60.1:45182 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:22] INFO:     174.1.60.1:39946 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:23] INFO:     174.1.60.1:39958 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:24] INFO:     174.1.60.1:39962 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:25] INFO:     174.1.60.1:39978 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:26] INFO:     174.1.60.1:39990 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:27] INFO:     174.1.60.1:39994 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:28] INFO:     174.1.60.1:39996 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:29] INFO:     174.1.60.1:40002 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:30] INFO:     174.1.60.1:40012 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:31] INFO:     174.1.60.1:46978 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:32] INFO:     174.1.60.1:46988 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:33] INFO:     174.1.60.1:46996 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:34] INFO:     174.1.60.1:47004 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:39] INFO:     174.1.60.1:47020 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:40] INFO:     174.1.60.1:47022 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:41] INFO:     174.1.60.1:59158 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:42] INFO:     174.1.60.1:59170 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:43] INFO:     174.1.60.1:59172 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:44] INFO:     174.1.60.1:59180 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:45] INFO:     174.1.60.1:59184 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:46] INFO:     174.1.60.1:59186 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:47] INFO:     174.1.60.1:59200 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:48] INFO:     174.1.60.1:59202 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:49] INFO:     174.1.60.1:59214 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:50] INFO:     174.1.60.1:59220 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:52] INFO:     174.1.60.1:40566 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:53] INFO:     174.1.60.1:40576 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:54] INFO:     174.1.60.1:40588 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:55] INFO:     174.1.60.1:40596 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:56] INFO:     174.1.60.1:40598 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:57] INFO:     174.1.60.1:40604 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:58] INFO:     174.1.60.1:40608 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:59] INFO:     174.1.60.1:40610 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:00] INFO:     174.1.60.1:40622 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:01] INFO:     174.1.60.1:41102 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:02] INFO:     174.1.60.1:41108 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:03] INFO:     174.1.60.1:41122 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:04] INFO:     174.1.60.1:41124 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:05] INFO:     174.1.60.1:41134 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:06] INFO:     174.1.60.1:41140 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:07] INFO:     174.1.60.1:41142 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:08] INFO:     174.1.60.1:41158 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:09] INFO:     174.1.60.1:41160 - "GET /health HTTP/1.1" 503 Service Unavailable
+[rank3]:[W827 04:35:10.359210278 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank2]:[W827 04:35:10.359272236 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank0]:[W827 04:35:10.360047094 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank1]:[W827 04:35:10.370610409 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[2026-08-27 04:35:10 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.28
+[2026-08-27 04:35:10 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.29
+[2026-08-27 04:35:10] INFO:     127.0.0.1:60364 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 04:35:10] Disaggregation warmup requests completed for all 1 DP ranks
+[2026-08-27 04:35:10] End of disaggregation warmup
+[2026-08-27 04:35:10] The server is fired up and ready to roll!
+[2026-08-27 04:35:29 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.39
+[2026-08-27 04:35:29 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.39
+[2026-08-27 04:35:31 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25.83
+[2026-08-27 04:35:31 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25.84
+[2026-08-27 04:35:32] INFO:     174.1.60.1:49902 - "GET /health HTTP/1.1" 200 OK
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/d_healthy_p_node1.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/d_healthy_p_node1.log
new file mode 100644
index 0000000..5675a96
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/d_healthy_p_node1.log
@@ -0,0 +1,745 @@
+/usr/local/lib/python3.12/dist-packages/torchao/quantization/quant_api.py:1731: SyntaxWarning: invalid escape sequence '\.'
+  """Configuration class for applying different quantization configs to modules or parameters based on their fully qualified names (FQNs).
+/sgl-workspace/sglang/python/sglang/launch_server.py:56: UserWarning: 'python -m sglang.launch_server' is still supported, but 'sglang serve' is the recommended entrypoint.
+  Example: sglang serve --model-path <model> [options]
+  warnings.warn(
+[2026-08-27 04:28:43] Breakable CUDA graph is incompatible with MLA attention; disabling prefill CUDA graph.
+[2026-08-27 04:28:43] Attention backend not specified. Use triton backend by default.
+[2026-08-27 04:28:43] Pipeline parallelism is incompatible with overlap schedule.
+[2026-08-27 04:28:46] server_args=ServerArgs(model_path='/data/hf_models/Kimi-K3', tokenizer_path='/data/hf_models/Kimi-K3', tokenizer_mode='auto', tokenizer_backend='huggingface', tokenizer_worker_num=1, detokenizer_worker_num=1, skip_tokenizer_init=False, load_format='auto', model_loader_extra_config='{}', trust_remote_code=True, context_length=None, is_embedding=False, enable_multimodal=None, revision=None, model_impl='auto', model_config_parser='auto', json_model_override_args='{}', dtype='auto', quantization=None, quantization_param_path=None, kv_cache_dtype='auto', enable_fp32_lm_head=False, modelopt_quant=None, modelopt_checkpoint_restore_path=None, modelopt_checkpoint_save_path=None, modelopt_export_path=None, quantize_and_serve=False, rl_quant_profile=None, enable_tf32_matmul=False, mem_fraction_static=0.88, max_running_requests=None, max_queued_requests=None, max_total_tokens=None, chunked_prefill_size=8192, enable_dynamic_chunking=False, max_prefill_tokens=16384, prefill_max_requests=None, schedule_policy='fcfs', enable_priority_scheduling=False, disable_priority_preemption=False, default_priority_value=None, abort_on_priority_when_disabled=False, schedule_low_priority_values_first=False, priority_scheduling_preemption_threshold=10, retraction_policy='length', schedule_conservativeness=1.0, page_size=64, swa_full_tokens_ratio=0.8, disable_hybrid_swa_memory=False, radix_eviction_policy='lru', prefill_only_disable_kv_cache=False, disable_radix_cache=True, enable_page_major_kv_layout=False, enable_unified_memory=False, disable_chunked_prefix_cache=False, disable_overlap_schedule=True, num_continuous_decode_steps=1, scheduler_recv_interval=1, enable_mixed_chunk=False, nccl_port=None, dist_timeout=3600, dist_init_addr='174.1.60.1:20000', nnodes=4, node_rank=1, tp_size=4, dcp_size=1, pp_size=8, pp_max_micro_batch_size=None, pp_async_batch_depth=0, dp_size=1, load_balance_method='follow_bootstrap_room', attn_cp_size=1, moe_dp_size=1, dcp_comm_backend='ag_rs', dcp_replicate_q_proj=None, dwdp_size=1, enable_prefill_cp=False, cp_strategy=None, enable_dsa_cache_layer_split=False, enable_dsa_prefill_context_parallel=False, dsa_prefill_cp_mode='round-robin-split', enable_prefill_context_parallel=False, prefill_cp_mode='in-seq-split', enable_cp_decode_attn_tp=False, enable_dp_attention=False, enable_dp_attention_local_control_broadcast=False, enable_dp_lm_head=False, enable_attn_tp_input_scattered=False, disable_attn_tp_gather=False, enable_p2p_check=False, device='cuda', base_gpu_id=0, gpu_id_step=1, random_seed=577755302, watchdog_timeout=300, soft_watchdog_timeout=None, sleep_on_idle=False, use_ray=False, custom_sigquit_handler=None, numa_node=None, gc_threshold=None, host='0.0.0.0', port=30000, fastapi_root_path='', smg_grpc_mode=False, grpc_mode=False, grpc_port=None, sidecar=None, sidecar_args=None, skip_server_warmup=False, warmups=None, enable_http2=False, ssl_keyfile=None, ssl_certfile=None, ssl_ca_certs=None, ssl_keyfile_password=None, enable_ssl_refresh=False, api_key=None, admin_api_key=None, served_model_name='kimi-k3', weight_version='default', chat_template=None, hf_chat_template_name=None, completion_template=None, file_storage_path='sglang_storage', enable_cache_report=False, reasoning_parser=None, default_chat_template_kwargs=None, strip_thinking_cache=False, enable_strict_thinking=False, tool_call_parser=None, tool_server=None, sampling_defaults='model', asr_max_buffer_seconds=60, asr_max_concurrent_sessions=32, preferred_sampling_params=None, allow_auto_truncate=False, stream_interval=1, batch_notify_size=16, stream_response_default_include_usage=False, incremental_streaming_output=False, enable_streaming_session=False, enable_session_radix_cache=False, log_level='info', log_level_http=None, log_requests=False, log_requests_level=2, log_requests_format='text', log_requests_target=None, uvicorn_access_log_exclude_prefixes=[], crash_dump_folder=None, show_time_cost=False, enable_metrics=False, smg_http_sidecar_port=None, enable_mfu_metrics=False, enable_metrics_for_all_schedulers=False, load_snapshot_publish_interval=15, tokenizer_metrics_custom_labels_header='x-custom-labels', tokenizer_metrics_allowed_custom_labels=None, extra_metric_labels=None, bucket_time_to_first_token=None, bucket_inter_token_latency=None, bucket_e2e_request_latency=None, prompt_tokens_buckets=None, generation_tokens_buckets=None, gc_warning_threshold_secs=0.0, decode_log_interval=40, enable_request_time_stats_logging=False, kv_events_config=None, enable_forward_pass_metrics=False, forward_pass_metrics_worker_id='', forward_pass_metrics_ipc_name=None, enable_trace=False, trace_modules='request', otlp_traces_endpoint='localhost:4317', export_metrics_to_file=False, export_metrics_to_file_dir=None, stat_loggers=None, constrained_json_whitespace_pattern=None, constrained_json_disable_any_whitespace=False, attention_backend='triton', decode_attention_backend=None, prefill_attention_backend=None, sampling_backend='flashinfer', grammar_backend='xgrammar', radix_cache_backend=None, mm_attention_backend=None, fp8_gemm_runner_backend='auto', fp4_gemm_runner_backend='auto', bf16_gemm_backend='auto', dsa_prefill_backend=None, dsa_decode_backend=None, dsa_paged_mqa_logits_backend='auto', dsa_topk_backend='sgl-kernel', disable_flashinfer_autotune=False, flashinfer_autotune_skip_ops=None, mamba_backend='triton', cuda_graph_config=CudaGraphConfig(decode=PhaseConfig(backend='disabled', max_bs=16, bs=[1, 2, 4, 8, 12, 16], tc_compiler='eager', full_prefill_max_req=None), prefill=PhaseConfig(backend='disabled', max_bs=2048, bs=[4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], tc_compiler='eager', full_prefill_max_req=None)), cuda_graph_backend_decode=None, cuda_graph_backend_prefill=None, cuda_graph_max_bs_decode=16, cuda_graph_max_bs_prefill=None, cuda_graph_bs_decode=None, cuda_graph_bs_prefill=None, cuda_graph_tc_compiler=None, disable_prefill_cuda_graph=False, disable_decode_cuda_graph=False, disable_cuda_graph=False, disable_cuda_graph_padding=False, enable_profile_cuda_graph=False, enable_cudagraph_gc=False, debug_cuda_graph=False, enable_layerwise_nvtx_marker=False, enable_nccl_nvls=False, enable_symm_mem=False, triton_attention_reduce_in_fp32=False, triton_attention_num_kv_splits=8, triton_attention_split_tile_size=None, flashinfer_mla_disable_ragged=False, enable_fused_qk_norm_rope=False, enable_precise_embedding_interpolation=False, enable_fused_moe_sum_all_reduce=False, enable_deepseek_v4_fp4_indexer=False, disable_custom_all_reduce=False, enable_mscclpp=False, enable_torch_symm_mem=False, enable_scattered_sconv=False, pre_warm_nccl=False, enable_quant_communications=False, enable_flashinfer_allreduce_fusion=False, enforce_disable_flashinfer_allreduce_fusion=False, flashinfer_allreduce_fusion_backend=None, enable_aiter_allreduce_fusion=False, enable_torch_compile=False, enable_torch_compile_debug_mode=False, torch_compile_max_bs=32, torchao_config='', speculative_algorithm=None, speculative_draft_model_path=None, speculative_draft_model_revision=None, speculative_draft_load_format=None, speculative_num_steps=None, speculative_eagle_topk=None, speculative_num_draft_tokens=None, speculative_dflash_block_size=None, speculative_dspark_block_size=None, speculative_dspark_sps_table_path=None, speculative_dspark_confidence_sts_path=None, speculative_dspark_align_verify_tokens_to_graph_tier=False, speculative_accept_threshold_single=1.0, speculative_accept_threshold_acc=1.0, speculative_use_rejection_sampling=False, speculative_token_map=None, speculative_attention_mode='prefill', speculative_draft_attention_backend=None, speculative_draft_window_size=None, speculative_moe_runner_backend='flashinfer_mxfp4', speculative_moe_a2a_backend=None, speculative_draft_model_quantization=None, speculative_skip_dp_mlp_sync=False, enable_multi_layer_eagle=False, speculative_adaptive=False, speculative_adaptive_config=None, decoupled_spec_bind_endpoint=None, decoupled_spec_connect_endpoints=None, decoupled_spec_rank=None, decoupled_spec_role='null', spec_trace_dir=None, speculative_ngram_min_bfs_breadth=1, speculative_ngram_max_bfs_breadth=10, speculative_ngram_match_type='BFS', speculative_ngram_max_trie_depth=18, speculative_ngram_capacity=10000000, speculative_ngram_external_corpus_path=None, speculative_ngram_external_sam_budget=0, speculative_ngram_external_corpus_max_tokens=10000000, ep_size=4, moe_a2a_backend='none', moe_runner_backend='flashinfer_mxfp4', flashinfer_mxfp4_moe_precision='default', deepep_mode='auto', fuseep_mode=2, deepep_dispatcher_output_dtype='auto', ep_num_redundant_experts=0, ep_dispatch_algorithm=None, init_expert_location='trivial', enable_eplb=False, eplb_algorithm='auto', eplb_rebalance_num_iterations=1000, eplb_rebalance_layers_per_chunk=None, eplb_min_rebalancing_utilization_threshold=1.0, expert_distribution_recorder_mode=None, expert_distribution_recorder_buffer_size=1000, enable_expert_distribution_metrics=False, deepep_config=None, moe_dense_tp_size=None, elastic_ep_backend=None, enable_elastic_expert_backup=False, mooncake_ib_device=None, enable_waterfill=False, ep_join_mode=None, ep_join_rank_offset=0, elastic_ep_initial_size=None, max_ep_size=None, elastic_ep_scale_timeout=600, elastic_ep_rejoin=False, disable_flashinfer_cutlass_moe_fp4_allgather=False, disable_shared_experts_fusion=False, enforce_shared_experts_fusion=False, max_mamba_cache_size=None, mamba_ssm_dtype=None, mamba_max_states_per_path=-1, enable_mamba_cache_stochastic_rounding=False, mamba_cache_philox_rounds=0, mamba_full_memory_ratio=0.36, mamba_radix_cache_strategy='extra_buffer_lazy', uses_mamba_radix_cache=False, mamba_track_interval=256, enable_int8_mamba_checkpoint=False, int8_mamba_ckpt_size=None, linear_attn_backend='triton', linear_attn_decode_backend=None, linear_attn_prefill_backend=None, linear_attn_verify_backend=None, enable_linear_replayssm=False, linear_replayssm_cache_len=16, enable_linear_replayssm_spec=False, enable_hierarchical_cache=False, hicache_ratio=2.0, hicache_size=0, hicache_write_policy='write_through', hicache_io_backend='kernel', hicache_mem_layout='page_first', hicache_storage_backend=None, hicache_storage_prefetch_policy='timeout', hicache_storage_backend_extra_config=None, enable_hisparse=False, hisparse_config=None, enable_broadcast_mm_inputs_process=False, enable_prefix_mm_cache=False, mm_enable_dp_encoder=False, mm_process_config={}, mm_processor_worker_num=0, mm_io_worker_num=0, limit_mm_data_per_request=None, enable_mm_global_cache=False, disable_fast_image_processor=False, mm_feature_transport='cpu', keep_mm_feature_on_device=False, enable_lora=None, enable_lora_overlap_loading=None, max_lora_rank=None, lora_target_modules=None, lora_paths=None, max_loaded_loras=None, max_loras_per_batch=8, lora_eviction_policy='lru', lora_backend='csgmv', max_lora_chunk_size=16, experts_shared_outer_loras=None, lora_use_virtual_experts=False, lora_strict_loading=False, lora_drain_wait_threshold=0.0, enable_two_batch_overlap=False, enable_single_batch_overlap=False, tbo_token_distribution_threshold=0.48, cpu_offload_gb=0, offload_group_size=-1, offload_num_in_group=1, offload_prefetch_step=1, offload_mode='cpu', enable_lmcache=False, lmcache_config_file=None, enable_flexkv=False, flexkv_config_file=None, kt_weight_path=None, kt_method='AMXINT4', kt_cpuinfer=None, kt_threadpool_count=2, kt_num_gpu_experts=None, kt_max_deferred_experts_per_token=None, dllm_algorithm=None, dllm_algorithm_config=None, dllm_fdfo=True, disaggregation_mode='prefill', disaggregation_transfer_backend='mooncake', disaggregation_bootstrap_port=28800, disaggregation_ib_device='mlx5_0,mlx5_1,mlx5_2,mlx5_3', disaggregation_decode_enable_radix_cache=False, disaggregation_decode_enable_offload_kvcache=False, num_reserved_decode_tokens=512, disaggregation_decode_extra_slots=None, disaggregation_decode_polling_interval=1, optimistic_prefill_attempts=0, encoder_only=False, language_only=False, encoder_transfer_backend='zmq_to_tokenizer', encoder_urls=[], encoder_bootstrap_port=8997, encoder_register_urls=[], enable_adaptive_dispatch_to_encoder=False, enable_pdmux=False, pdmux_config_path=None, sm_group_num=8, custom_weight_loader=[], weight_loader_disable_mmap=False, weight_loader_prefetch_checkpoints=False, weight_loader_prefetch_num_threads=4, weight_loader_drop_cache_after_load=False, remote_instance_weight_loader_seed_instance_ip=None, remote_instance_weight_loader_seed_instance_service_port=None, remote_instance_weight_loader_send_weights_group_ports=None, remote_instance_weight_loader_backend='nccl', remote_instance_weight_loader_start_seed_via_transfer_engine=False, engine_info_bootstrap_port=6789, modelexpress_config=None, download_dir=None, model_checksum=None, delete_ckpt_after_loading=False, decrypted_config_file=None, decrypted_draft_config_file=None, checkpoint_engine_wait_weights_before_ready=False, enable_prefill_delayer=False, prefill_delayer_max_delay_passes=30, prefill_delayer_token_usage_low_watermark=None, prefill_delayer_forward_passes_buckets=None, prefill_delayer_wait_seconds_buckets=None, prefill_delayer_queue_min_ratio=None, prefill_delayer_max_delay_ms=None, min_free_slots_delay=None, enable_deterministic_inference=False, rl_on_policy_target=None, kv_canary='none', kv_canary_real_data='none', kv_canary_sweep_interval=0, enable_dynamic_batch_tokenizer=False, dynamic_batch_tokenizer_batch_size=32, dynamic_batch_tokenizer_batch_timeout=0.002, enable_tokenizer_batch_encode=False, disable_tokenizer_batch_decode=False, debug_tensor_dump_output_folder=None, debug_tensor_dump_layers=None, debug_tensor_dump_input_file=None, enable_memory_saver=False, enable_weights_cpu_backup=False, enable_draft_weights_cpu_backup=False, enable_custom_logit_processor=False, enable_return_hidden_states=False, enable_return_routed_experts=False, enable_return_indexer_topk=False, disable_outlines_disk_cache=False, enable_mis=False, weight_cache_mode='off', weight_cache_socket=None, weight_cache_timeout=1800, forward_hooks=None, msprobe_dump_config=None)
+[2026-08-27 04:28:46] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:46] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:46] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:46] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:46] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:46] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:46] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:46] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:02 PP3 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:02 PP3 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:03 PP3 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:03 PP3 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:03 PP3 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:03 PP3 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:03 PP2 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:03 PP2 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:03 PP2 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:03 PP2 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:04 PP2 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:04 PP2 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:04 PP2 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:04 PP2 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:04 PP3 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:04 PP3 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:05 PP3 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:05 PP2 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:05 PP3 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:05 PP3 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:05 PP2 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:06.044592   241 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:06.044694   241 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:06.044710   241 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.2 port: 12001
+I0827 04:29:06.044759   241 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.2:16343
+I0827 04:29:06.044883   241 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:06.053642   242 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:06.053738   242 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:06.053754   242 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.2 port: 12001
+I0827 04:29:06.053797   242 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.2:15170
+I0827 04:29:06.053925   242 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:06.058039   237 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:06.058135   237 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:06.058149   237 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.2 port: 12001
+I0827 04:29:06.058192   237 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.2:15486
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:06.058215   236 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:06.058310   236 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:06.058316   237 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:06.058326   236 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.2 port: 12001
+I0827 04:29:06.058367   236 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.2:16140
+I0827 04:29:06.058492   236 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:06.059734   239 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:06.059828   239 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:06.059844   239 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.2 port: 12001
+I0827 04:29:06.059890   239 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.2:15755
+I0827 04:29:06.060027   239 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:06.064433   237 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:06.064513   236 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:06.065588   239 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:06.069945   237 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:06.070852   236 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:06.071113   239 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:06.071978   241 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:06.079348   236 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:06.079967   242 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:06.080070   241 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:06.087044   241 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:06.087074   242 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:06.094024   242 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:06.095021   241 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:06.095822   237 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:06.097810   241 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:06.097839   241 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:06.097889   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.099939   239 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:06.100984   242 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:06.103683   242 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:06.103708   242 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:06.103756   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.103785   236 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:06.103833   237 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:06.106014   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:06.106982   236 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:06.107007   236 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:06.107053   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.107446   241 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:02
+I0827 04:29:06.107475   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.107936   239 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:06.107971   237 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:06.107997   237 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:06.108048   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.110760   239 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:06.110790   239 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:06.110847   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.111824   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:06.113214   242 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:02
+I0827 04:29:06.113242   242 rdma_context.cpp:205] Using SIEVE endpoint store
+[2026-08-27 04:29:06 PP2 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+I0827 04:29:06.114821   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:06.115196   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+[2026-08-27 04:29:06 PP2 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+I0827 04:29:06.116189   241 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:02
+I0827 04:29:06.116214   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.116621   237 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:02
+I0827 04:29:06.116650   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.118325   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:06.119874   239 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:02
+I0827 04:29:06.119906   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.121601   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:06.122952   242 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:02
+I0827 04:29:06.122977   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.127393   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:06.128841   239 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:02
+I0827 04:29:06.128867   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.132040   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:06.133428   236 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:02
+I0827 04:29:06.133458   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.136312   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:06.137817   239 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:02
+I0827 04:29:06.137846   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.144285   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:06.144373   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:06.145742   237 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:02
+I0827 04:29:06.145776   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.145833   241 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:02
+I0827 04:29:06.145866   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.152084   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:06.153820   242 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:02
+I0827 04:29:06.153860   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.154352   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:06.155728   241 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:02
+I0827 04:29:06.155835   241 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:06 PP3 TP2 EP2] Init torch distributed begin.
+I0827 04:29:06.163965   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:06.164443   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:06.164605   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:06.165319   236 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:02
+I0827 04:29:06.165345   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.165951   239 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:02
+I0827 04:29:06.166070   239 transfer_engine_impl.cpp:388] installTransport, type=rdma
+I0827 04:29:06.166186   242 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:02
+I0827 04:29:06.166311   242 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:06 PP3 TP0 EP0] Init torch distributed begin.
+[2026-08-27 04:29:06 PP3 TP3 EP3] Init torch distributed begin.
+I0827 04:29:06.172065   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:06.173699   237 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:02
+I0827 04:29:06.173743   237 rdma_context.cpp:205] Using SIEVE endpoint store
+[2026-08-27 04:29:06 PP3 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+I0827 04:29:06.192251   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:06.193763   236 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:02
+I0827 04:29:06.193797   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.200259   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:06.201810   237 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:02
+I0827 04:29:06.201937   237 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:06 PP2 TP2 EP2] Init torch distributed begin.
+I0827 04:29:06.220070   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:06.221505   236 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:02
+I0827 04:29:06.221653   236 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:06 PP2 TP1 EP1] Init torch distributed begin.
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:06.613179   235 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:06.613286   235 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:06.613301   235 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.2 port: 12001
+I0827 04:29:06.613344   235 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.2:16382
+I0827 04:29:06.613492   235 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:06.618471   235 topology.cpp:127] Device mlx5_0 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:06.618719   238 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:06.618824   238 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:06.618839   238 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.2 port: 12001
+I0827 04:29:06.618882   238 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.2:16170
+I0827 04:29:06.619020   238 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:06.624203   238 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:06.629066   238 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:06.633975   238 topology.cpp:127] Device mlx5_2 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:06.636366   240 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:06.636454   240 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:06.636466   240 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.2 port: 12001
+I0827 04:29:06.636509   240 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.2:15840
+I0827 04:29:06.636675   240 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:06.638876   238 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:06.640780   238 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:06.640802   238 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:06.640842   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.641396   240 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:06.643196   235 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:06.645701   240 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:06.649896   235 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:06.650310   240 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:06.654819   240 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:06.656375   240 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:06.656396   240 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:06.656432   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.657156   235 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:06.659863   235 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:06.659895   235 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:06.659950   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.661671   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:06.662956   240 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:02
+I0827 04:29:06.662979   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.664074   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:06.665762   238 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:02
+I0827 04:29:06.665796   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.667351   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:06.668808   235 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:02
+I0827 04:29:06.668838   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.676294   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:06.677721   235 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:02
+I0827 04:29:06.677748   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.692129   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:06.693621   240 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:02
+I0827 04:29:06.693648   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.696089   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:06.697444   238 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:02
+I0827 04:29:06.697470   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.707720   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:06.709362   235 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:02
+I0827 04:29:06.709394   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.719987   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:06.721463   240 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:02
+I0827 04:29:06.721489   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.728108   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:06.728855   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:06.729467   238 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:02
+I0827 04:29:06.729491   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.730392   240 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:02
+I0827 04:29:06.730533   240 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:06 PP3 TP1 EP1] Init torch distributed begin.
+I0827 04:29:06.736888   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:06.738317   238 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:02
+I0827 04:29:06.738422   238 transfer_engine_impl.cpp:388] installTransport, type=rdma
+I0827 04:29:06.738466   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+[2026-08-27 04:29:06 PP2 TP3 EP3] Init torch distributed begin.
+I0827 04:29:06.739869   235 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:02
+I0827 04:29:06.739982   235 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:06 PP2 TP0 EP0] Init torch distributed begin.
+[2026-08-27 04:29:26 PP3 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:26 PP2 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:27 PP2 TP3 EP3] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP2 TP1 EP1] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP2 TP2 EP2] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP2 TP0 EP0] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP3 TP3 EP3] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP3 TP0 EP0] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP3 TP2 EP2] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP3 TP1 EP1] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP3 TP3 EP3] Init torch distributed ends. elapsed=21.61 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP3 TP2 EP2] Init torch distributed ends. elapsed=21.62 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP3 TP1 EP1] Init torch distributed ends. elapsed=21.04 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP3 TP0 EP0] Init torch distributed ends. elapsed=21.61 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP2 TP2 EP2] Init torch distributed ends. elapsed=21.57 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP2 TP1 EP1] Init torch distributed ends. elapsed=21.55 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP2 TP0 EP0] Init torch distributed ends. elapsed=21.03 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP2 TP3 EP3] Init torch distributed ends. elapsed=21.04 s, mem usage=0.54 GB
+[2026-08-27 04:29:28 PP3 TP3 EP3] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP3 TP0 EP0] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP2 TP0 EP0] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP3 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP2 TP1 EP1] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP3 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP2 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP3 TP2 EP2] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP2 TP3 EP3] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP2 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP2 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP3 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP2 TP2 EP2] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP3 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP3 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP2 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP3 TP1 EP1] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP2 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP3 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP2 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP3 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP3 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP2 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP3 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP2 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP3 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP2 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP2 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP2 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP2 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP3 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP3 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:30:31 PP2 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:31 PP2 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:31 PP2 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:31 PP2 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:31 PP2 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:31 PP2 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:31 PP2 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:31 PP2 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP2 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:32 PP2 TP0 EP0] Load weight end. elapsed=64.92 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=36.14 GB, mem usage=45.91 GB.
+[2026-08-27 04:30:33 PP3 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:33 PP3 TP1 EP1] Load weight end. elapsed=65.31 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:34 PP2 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:34 PP2 TP1 EP1] Load weight end. elapsed=66.07 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=36.14 GB, mem usage=45.91 GB.
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:35 PP3 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:35 PP3 TP0 EP0] Load weight end. elapsed=67.41 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:36 PP2 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:36 PP2 TP3 EP3] Load weight end. elapsed=68.04 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=36.14 GB, mem usage=45.91 GB.
+[2026-08-27 04:30:36 PP2 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:36 PP2 TP2 EP2] Load weight end. elapsed=68.23 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=36.14 GB, mem usage=45.91 GB.
+[2026-08-27 04:30:36 PP2 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:36 PP2 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:36 PP2 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:36 PP2 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:36 PP2 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:36 PP2 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:36 PP2 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:36 PP2 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:36 PP3 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:36 PP3 TP3 EP3] Load weight end. elapsed=68.86 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:37 PP3 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:37 PP3 TP2 EP2] Load weight end. elapsed=69.68 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:38 PP3 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:38 PP3 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:38 PP3 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:38 PP3 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:38 PP3 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:38 PP3 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:38 PP3 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:38 PP3 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:54 PP3 TP3 EP3] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP3 TP2 EP2] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP3 TP1 EP1] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP2 TP3 EP3] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP3 TP0 EP0] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP2 TP2 EP2] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP2 TP1 EP1] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP2 TP0 EP0] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP2 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:54 PP2 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:54 PP3 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP3 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP2 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:54 PP3 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP3 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP2 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:54 PP3 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP3 TP3 EP3] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP2 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP2 TP2 EP2] Memory pool end. avail mem=23.25 GB
+[2026-08-27 04:30:54 PP2 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP2 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP2 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP3 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP3 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP2 TP3 EP3] Memory pool end. avail mem=23.25 GB
+[2026-08-27 04:30:54 PP2 TP1 EP1] Memory pool end. avail mem=23.25 GB
+[2026-08-27 04:30:54 PP3 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP3 TP1 EP1] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP3 TP2 EP2] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP2 TP0 EP0] Memory pool end. avail mem=23.25 GB
+[2026-08-27 04:30:54 PP3 TP0 EP0] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP3 TP3 EP3] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp3_pp3_dp0.json
+[2026-08-27 04:30:54 PP2 TP1 EP1] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp1_pp2_dp0.json
+[2026-08-27 04:30:54 PP2 TP2 EP2] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp2_pp2_dp0.json
+[2026-08-27 04:30:54 PP3 TP1 EP1] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp1_pp3_dp0.json
+[2026-08-27 04:30:54 PP2 TP3 EP3] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp3_pp2_dp0.json
+[2026-08-27 04:30:54 PP3 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:30:54 PP3 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO[TensorRT-LLM][INFO] Set logger level to INFO
+
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[2026-08-27 04:30:54 PP3 TP2 EP2] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp2_pp3_dp0.json
+[2026-08-27 04:30:54 PP2 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:30:54 PP2 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:30:54 PP3 TP0 EP0] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp0_pp3_dp0.json
+[2026-08-27 04:30:54 PP2 TP0 EP0] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp0_pp2_dp0.json
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.65profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.60profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.79profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.51profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.86profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.27profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.62profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.98profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.86profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 12.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 13.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.93profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.92profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.09profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.63profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  6.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  7.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.33profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.58profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  8.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.56profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.23profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.63profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.36profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.09profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.97profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  8.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.30profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 13.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 13.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  8.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  8.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 11.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 11.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.23profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.48profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.23profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.53profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.27profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.59profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.62profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.97profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.60profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.46profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.12profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.68profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.77profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.62profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  6.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.87profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.30profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.40profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.83profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.94profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.59profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.66profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.29profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.23profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:01,  3.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.68profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.75profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  7.01profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.22profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.65profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.62profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.46profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.22profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  7.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:01<00:06,  1.03s/profile]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:01,  3.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.50profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.93profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.48profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.11profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.12profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.30profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.68profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  7.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 11.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  9.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 12.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 11.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 11.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 11.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.11profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.31profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.65profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.31profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.35profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.49profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:01<00:01,  2.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.65profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.68profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  7.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.09profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 11.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.88profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.98profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.50profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.62profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.51profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.91profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:01,  3.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.31profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.86profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.27profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.59profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.17profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.76profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.73profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  9.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.37profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.15profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.98profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.65profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.97profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.23profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.46profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  9.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.03profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.57profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.40profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.40profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.57profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.13profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.56profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  6.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.73profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.23profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  3.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.86profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.77profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.92profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.37profile/s]
+[2026-08-27 04:31:26 PP2 TP3 EP3] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.39profile/s]
+[2026-08-27 04:31:26 PP2 TP2 EP2] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.75profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.46profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.35profile/s]
+[2026-08-27 04:31:26 PP2 TP1 EP1] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.12profile/s]
+[2026-08-27 04:31:26 PP2 TP0 EP0] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.66profile/s][2026-08-27 04:31:26 PP2 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:26 PP2 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:26 PP2 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:26 PP2 TP0 EP0] max_total_num_tokens=3801408, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=54, context_len=1048576, available_gpu_mem=20.47 GB
+[2026-08-27 04:31:26 PP2 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:26 PP2 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:26 PP2 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:26 PP2 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:26 PP2 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.93profile/s]
+[2026-08-27 04:31:27 PP3 TP1 EP1] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.16profile/s]
+[2026-08-27 04:31:27 PP3 TP0 EP0] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.43profile/s]
+[2026-08-27 04:31:27 PP3 TP2 EP2] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.05profile/s]
+[2026-08-27 04:31:27 PP3 TP3 EP3] FlashInfer autotune completed.
+[2026-08-27 04:31:27 PP3 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:27 PP3 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:27 PP3 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:27 PP3 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:27 PP3 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:27 PP3 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:27 PP3 TP0 EP0] max_total_num_tokens=3801408, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=54, context_len=1048576, available_gpu_mem=16.54 GB
+[2026-08-27 04:31:27 PP3 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:27 PP3 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+W0827 04:31:30.590376  2521 memory_location.cpp:73] Failed to get NUMA node, addr: 0x513c5ec0, len: 6912: Operation not permitted [1]
+W0827 04:31:30.590432  2522 memory_location.cpp:73] Failed to get NUMA node, addr: 0x514a0900, len: 6912: Operation not permitted [1]
+W0827 04:31:30.590495  2525 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51226dc0, len: 55296: Operation not permitted [1]
+W0827 04:31:30.590492  2523 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5128db40, len: 6912: Operation not permitted [1]
+W0827 04:31:30.590533  2524 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5156ff80, len: 6912: Operation not permitted [1]
+W0827 04:31:30.590600  2526 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51320ac0, len: 55296: Operation not permitted [1]
+W0827 04:31:30.590636  2527 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51768540, len: 6912: Operation not permitted [1]
+W0827 04:31:30.590826  2528 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51483dc0, len: 13824: Operation not permitted [1]
+W0827 04:31:30.590914  2529 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51119380, len: 6912: Operation not permitted [1]
+W0827 04:31:30.590979  2530 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50fe1e00, len: 6912: Operation not permitted [1]
+W0827 04:31:30.846120  2547 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50ceebc0, len: 6912: Operation not permitted [1]
+W0827 04:31:30.846271  2548 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50c57640, len: 6912: Operation not permitted [1]
+W0827 04:31:30.846328  2549 memory_location.cpp:73] Failed to get NUMA node, addr: 0x502ed680, len: 6912: Operation not permitted [1]
+W0827 04:31:30.846354  2550 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50d1c800, len: 6912: Operation not permitted [1]
+W0827 04:31:30.846448  2551 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50510a40, len: 55296: Operation not permitted [1]
+W0827 04:31:30.846473  2552 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5051e280, len: 55296: Operation not permitted [1]
+W0827 04:31:30.846571  2553 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50d01340, len: 6912: Operation not permitted [1]
+W0827 04:31:30.846689  2554 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50ce3440, len: 13824: Operation not permitted [1]
+W0827 04:31:30.846858  2555 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5030f440, len: 6912: Operation not permitted [1]
+W0827 04:31:30.846926  2556 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50cae980, len: 6912: Operation not permitted [1]
+W0827 04:31:31.198441  2575 memory_location.cpp:73] Failed to get NUMA node, addr: 0x514f46c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.198501  2576 memory_location.cpp:73] Failed to get NUMA node, addr: 0x509c6400, len: 6912: Operation not permitted [1]
+W0827 04:31:31.198501  2577 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51513300, len: 6912: Operation not permitted [1]
+W0827 04:31:31.198679  2578 memory_location.cpp:73] Failed to get NUMA node, addr: 0x506eef40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.198787  2579 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50dc9180, len: 55296: Operation not permitted [1]
+W0827 04:31:31.198915  2580 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50904700, len: 55296: Operation not permitted [1]
+W0827 04:31:31.198982  2581 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50aa9440, len: 6912: Operation not permitted [1]
+W0827 04:31:31.199081  2582 memory_location.cpp:73] Failed to get NUMA node, addr: 0x515012c0, len: 13824: Operation not permitted [1]
+W0827 04:31:31.199148  2583 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5161ac80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.199185  2584 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50dd69c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.763885  2603 memory_location.cpp:73] Failed to get NUMA node, addr: 0x519f1500, len: 6912: Operation not permitted [1]
+W0827 04:31:31.763998  2605 memory_location.cpp:73] Failed to get NUMA node, addr: 0x517df940, len: 6912: Operation not permitted [1]
+W0827 04:31:31.763923  2604 memory_location.cpp:73] Failed to get NUMA node, addr: 0x514ab440, len: 6912: Operation not permitted [1]
+W0827 04:31:31.764139  2606 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51556180, len: 6912: Operation not permitted [1]
+W0827 04:31:31.764169  2608 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50ed8140, len: 55296: Operation not permitted [1]
+W0827 04:31:31.764184  2607 memory_location.cpp:73] Failed to get NUMA node, addr: 0x515df040, len: 55296: Operation not permitted [1]
+W0827 04:31:31.764356  2609 memory_location.cpp:73] Failed to get NUMA node, addr: 0x517e3580, len: 6912: Operation not permitted [1]
+W0827 04:31:31.764564  2610 memory_location.cpp:73] Failed to get NUMA node, addr: 0x519cdb80, len: 13824: Operation not permitted [1]
+W0827 04:31:31.764624  2611 memory_location.cpp:73] Failed to get NUMA node, addr: 0x519dbf80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.764714  2612 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50f08d80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.868036  2629 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50a7a680, len: 6912: Operation not permitted [1]
+W0827 04:31:31.868046  2630 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50b6e400, len: 6912: Operation not permitted [1]
+W0827 04:31:31.868139  2631 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50e5f580, len: 6912: Operation not permitted [1]
+W0827 04:31:31.868261  2632 memory_location.cpp:73] Failed to get NUMA node, addr: 0x503c7100, len: 6912: Operation not permitted [1]
+W0827 04:31:31.868330  2633 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50c94ac0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.868432  2635 memory_location.cpp:73] Failed to get NUMA node, addr: 0x509b2880, len: 6912: Operation not permitted [1]
+W0827 04:31:31.868502  2634 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50b14c80, len: 55296: Operation not permitted [1]
+W0827 04:31:31.868564  2636 memory_location.cpp:73] Failed to get NUMA node, addr: 0x503ee840, len: 13824: Operation not permitted [1]
+W0827 04:31:31.868708  2637 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51057d80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.868835  2638 memory_location.cpp:73] Failed to get NUMA node, addr: 0x509cd940, len: 6912: Operation not permitted [1]
+W0827 04:31:32.003983  2656 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50715b00, len: 6912: Operation not permitted [1]
+W0827 04:31:32.004096  2657 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51346e80, len: 6912: Operation not permitted [1]
+W0827 04:31:32.004005  2655 memory_location.cpp:73] Failed to get NUMA node, addr: 0x507176c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.004179  2658 memory_location.cpp:73] Failed to get NUMA node, addr: 0x507e3a40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.004302  2659 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5078aa80, len: 55296: Operation not permitted [1]
+W0827 04:31:32.004357  2660 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5135c980, len: 55296: Operation not permitted [1]
+W0827 04:31:32.004410  2661 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5099f180, len: 6912: Operation not permitted [1]
+W0827 04:31:32.004585  2662 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50b287c0, len: 13824: Operation not permitted [1]
+W0827 04:31:32.004739  2663 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5070b6c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.004812  2664 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50753540, len: 6912: Operation not permitted [1]
+W0827 04:31:32.144867  2688 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50aed700, len: 6912: Operation not permitted [1]
+W0827 04:31:32.144946  2689 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50792740, len: 6912: Operation not permitted [1]
+W0827 04:31:32.145030  2690 memory_location.cpp:73] Failed to get NUMA node, addr: 0x508fca40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.145135  2691 memory_location.cpp:73] Failed to get NUMA node, addr: 0x505d7c40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.145318  2693 memory_location.cpp:73] Failed to get NUMA node, addr: 0x505ec340, len: 55296: Operation not permitted [1]
+W0827 04:31:32.145387  2692 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50602cc0, len: 55296: Operation not permitted [1]
+W0827 04:31:32.145479  2694 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50748c00, len: 6912: Operation not permitted [1]
+W0827 04:31:32.145632  2695 memory_location.cpp:73] Failed to get NUMA node, addr: 0x507e3480, len: 13824: Operation not permitted [1]
+W0827 04:31:32.145767  2696 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50316400, len: 6912: Operation not permitted [1]
+W0827 04:31:32.145864  2697 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50788600, len: 6912: Operation not permitted [1]
+W0827 04:31:32.177534  2726 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fff3c80, len: 6912: Operation not permitted [1]
+W0827 04:31:32.177623  2728 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fad6780, len: 6912: Operation not permitted [1]
+W0827 04:31:32.177625  2727 memory_location.cpp:73] Failed to get NUMA node, addr: 0x501dc2c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.178279  2729 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50596480, len: 6912: Operation not permitted [1]
+W0827 04:31:32.178463  2730 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50222dc0, len: 55296: Operation not permitted [1]
+W0827 04:31:32.178498  2731 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5038f6c0, len: 55296: Operation not permitted [1]
+W0827 04:31:32.178581  2732 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4ffd7ec0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.178695  2733 memory_location.cpp:73] Failed to get NUMA node, addr: 0x506d84c0, len: 13824: Operation not permitted [1]
+W0827 04:31:32.178884  2734 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5078efc0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.178922  2735 memory_location.cpp:73] Failed to get NUMA node, addr: 0x502c7b00, len: 6912: Operation not permitted [1]
+[2026-08-27 04:31:32] Dummy health check server started in background thread at 0.0.0.0:30000
+[rank10]:[W827 04:32:32.605391940 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank11]:[W827 04:32:32.605815131 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank9]:[W827 04:32:32.606041691 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank8]:[W827 04:32:32.608313727 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank11]:[W827 04:33:19.698792619 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank8]:[W827 04:33:19.699392665 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank9]:[W827 04:33:19.699501937 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank10]:[W827 04:33:19.699589651 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank14]:[W827 04:33:19.728965278 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank12]:[W827 04:33:19.729026815 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank13]:[W827 04:33:19.729092243 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank15]:[W827 04:33:19.730463091 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank14]:[W827 04:33:25.763484084 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank13]:[W827 04:33:25.764518868 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank15]:[W827 04:33:25.764776298 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank12]:[W827 04:33:25.765271843 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[2026-08-27 04:35:10 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.29
+[2026-08-27 04:35:10 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.29
+[2026-08-27 04:35:29 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.39
+[2026-08-27 04:35:29 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.39
+[2026-08-27 04:35:31 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25.86
+[2026-08-27 04:35:31 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25.87
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/d_healthy_p_node2.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/d_healthy_p_node2.log
new file mode 100644
index 0000000..df0f924
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/d_healthy_p_node2.log
@@ -0,0 +1,754 @@
+/usr/local/lib/python3.12/dist-packages/torchao/quantization/quant_api.py:1731: SyntaxWarning: invalid escape sequence '\.'
+  """Configuration class for applying different quantization configs to modules or parameters based on their fully qualified names (FQNs).
+/sgl-workspace/sglang/python/sglang/launch_server.py:56: UserWarning: 'python -m sglang.launch_server' is still supported, but 'sglang serve' is the recommended entrypoint.
+  Example: sglang serve --model-path <model> [options]
+  warnings.warn(
+[2026-08-27 04:28:43] Breakable CUDA graph is incompatible with MLA attention; disabling prefill CUDA graph.
+[2026-08-27 04:28:43] Attention backend not specified. Use triton backend by default.
+[2026-08-27 04:28:43] Pipeline parallelism is incompatible with overlap schedule.
+[2026-08-27 04:28:45] server_args=ServerArgs(model_path='/data/hf_models/Kimi-K3', tokenizer_path='/data/hf_models/Kimi-K3', tokenizer_mode='auto', tokenizer_backend='huggingface', tokenizer_worker_num=1, detokenizer_worker_num=1, skip_tokenizer_init=False, load_format='auto', model_loader_extra_config='{}', trust_remote_code=True, context_length=None, is_embedding=False, enable_multimodal=None, revision=None, model_impl='auto', model_config_parser='auto', json_model_override_args='{}', dtype='auto', quantization=None, quantization_param_path=None, kv_cache_dtype='auto', enable_fp32_lm_head=False, modelopt_quant=None, modelopt_checkpoint_restore_path=None, modelopt_checkpoint_save_path=None, modelopt_export_path=None, quantize_and_serve=False, rl_quant_profile=None, enable_tf32_matmul=False, mem_fraction_static=0.88, max_running_requests=None, max_queued_requests=None, max_total_tokens=None, chunked_prefill_size=8192, enable_dynamic_chunking=False, max_prefill_tokens=16384, prefill_max_requests=None, schedule_policy='fcfs', enable_priority_scheduling=False, disable_priority_preemption=False, default_priority_value=None, abort_on_priority_when_disabled=False, schedule_low_priority_values_first=False, priority_scheduling_preemption_threshold=10, retraction_policy='length', schedule_conservativeness=1.0, page_size=64, swa_full_tokens_ratio=0.8, disable_hybrid_swa_memory=False, radix_eviction_policy='lru', prefill_only_disable_kv_cache=False, disable_radix_cache=True, enable_page_major_kv_layout=False, enable_unified_memory=False, disable_chunked_prefix_cache=False, disable_overlap_schedule=True, num_continuous_decode_steps=1, scheduler_recv_interval=1, enable_mixed_chunk=False, nccl_port=None, dist_timeout=3600, dist_init_addr='174.1.60.1:20000', nnodes=4, node_rank=2, tp_size=4, dcp_size=1, pp_size=8, pp_max_micro_batch_size=None, pp_async_batch_depth=0, dp_size=1, load_balance_method='follow_bootstrap_room', attn_cp_size=1, moe_dp_size=1, dcp_comm_backend='ag_rs', dcp_replicate_q_proj=None, dwdp_size=1, enable_prefill_cp=False, cp_strategy=None, enable_dsa_cache_layer_split=False, enable_dsa_prefill_context_parallel=False, dsa_prefill_cp_mode='round-robin-split', enable_prefill_context_parallel=False, prefill_cp_mode='in-seq-split', enable_cp_decode_attn_tp=False, enable_dp_attention=False, enable_dp_attention_local_control_broadcast=False, enable_dp_lm_head=False, enable_attn_tp_input_scattered=False, disable_attn_tp_gather=False, enable_p2p_check=False, device='cuda', base_gpu_id=0, gpu_id_step=1, random_seed=336932344, watchdog_timeout=300, soft_watchdog_timeout=None, sleep_on_idle=False, use_ray=False, custom_sigquit_handler=None, numa_node=None, gc_threshold=None, host='0.0.0.0', port=30000, fastapi_root_path='', smg_grpc_mode=False, grpc_mode=False, grpc_port=None, sidecar=None, sidecar_args=None, skip_server_warmup=False, warmups=None, enable_http2=False, ssl_keyfile=None, ssl_certfile=None, ssl_ca_certs=None, ssl_keyfile_password=None, enable_ssl_refresh=False, api_key=None, admin_api_key=None, served_model_name='kimi-k3', weight_version='default', chat_template=None, hf_chat_template_name=None, completion_template=None, file_storage_path='sglang_storage', enable_cache_report=False, reasoning_parser=None, default_chat_template_kwargs=None, strip_thinking_cache=False, enable_strict_thinking=False, tool_call_parser=None, tool_server=None, sampling_defaults='model', asr_max_buffer_seconds=60, asr_max_concurrent_sessions=32, preferred_sampling_params=None, allow_auto_truncate=False, stream_interval=1, batch_notify_size=16, stream_response_default_include_usage=False, incremental_streaming_output=False, enable_streaming_session=False, enable_session_radix_cache=False, log_level='info', log_level_http=None, log_requests=False, log_requests_level=2, log_requests_format='text', log_requests_target=None, uvicorn_access_log_exclude_prefixes=[], crash_dump_folder=None, show_time_cost=False, enable_metrics=False, smg_http_sidecar_port=None, enable_mfu_metrics=False, enable_metrics_for_all_schedulers=False, load_snapshot_publish_interval=15, tokenizer_metrics_custom_labels_header='x-custom-labels', tokenizer_metrics_allowed_custom_labels=None, extra_metric_labels=None, bucket_time_to_first_token=None, bucket_inter_token_latency=None, bucket_e2e_request_latency=None, prompt_tokens_buckets=None, generation_tokens_buckets=None, gc_warning_threshold_secs=0.0, decode_log_interval=40, enable_request_time_stats_logging=False, kv_events_config=None, enable_forward_pass_metrics=False, forward_pass_metrics_worker_id='', forward_pass_metrics_ipc_name=None, enable_trace=False, trace_modules='request', otlp_traces_endpoint='localhost:4317', export_metrics_to_file=False, export_metrics_to_file_dir=None, stat_loggers=None, constrained_json_whitespace_pattern=None, constrained_json_disable_any_whitespace=False, attention_backend='triton', decode_attention_backend=None, prefill_attention_backend=None, sampling_backend='flashinfer', grammar_backend='xgrammar', radix_cache_backend=None, mm_attention_backend=None, fp8_gemm_runner_backend='auto', fp4_gemm_runner_backend='auto', bf16_gemm_backend='auto', dsa_prefill_backend=None, dsa_decode_backend=None, dsa_paged_mqa_logits_backend='auto', dsa_topk_backend='sgl-kernel', disable_flashinfer_autotune=False, flashinfer_autotune_skip_ops=None, mamba_backend='triton', cuda_graph_config=CudaGraphConfig(decode=PhaseConfig(backend='disabled', max_bs=16, bs=[1, 2, 4, 8, 12, 16], tc_compiler='eager', full_prefill_max_req=None), prefill=PhaseConfig(backend='disabled', max_bs=2048, bs=[4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], tc_compiler='eager', full_prefill_max_req=None)), cuda_graph_backend_decode=None, cuda_graph_backend_prefill=None, cuda_graph_max_bs_decode=16, cuda_graph_max_bs_prefill=None, cuda_graph_bs_decode=None, cuda_graph_bs_prefill=None, cuda_graph_tc_compiler=None, disable_prefill_cuda_graph=False, disable_decode_cuda_graph=False, disable_cuda_graph=False, disable_cuda_graph_padding=False, enable_profile_cuda_graph=False, enable_cudagraph_gc=False, debug_cuda_graph=False, enable_layerwise_nvtx_marker=False, enable_nccl_nvls=False, enable_symm_mem=False, triton_attention_reduce_in_fp32=False, triton_attention_num_kv_splits=8, triton_attention_split_tile_size=None, flashinfer_mla_disable_ragged=False, enable_fused_qk_norm_rope=False, enable_precise_embedding_interpolation=False, enable_fused_moe_sum_all_reduce=False, enable_deepseek_v4_fp4_indexer=False, disable_custom_all_reduce=False, enable_mscclpp=False, enable_torch_symm_mem=False, enable_scattered_sconv=False, pre_warm_nccl=False, enable_quant_communications=False, enable_flashinfer_allreduce_fusion=False, enforce_disable_flashinfer_allreduce_fusion=False, flashinfer_allreduce_fusion_backend=None, enable_aiter_allreduce_fusion=False, enable_torch_compile=False, enable_torch_compile_debug_mode=False, torch_compile_max_bs=32, torchao_config='', speculative_algorithm=None, speculative_draft_model_path=None, speculative_draft_model_revision=None, speculative_draft_load_format=None, speculative_num_steps=None, speculative_eagle_topk=None, speculative_num_draft_tokens=None, speculative_dflash_block_size=None, speculative_dspark_block_size=None, speculative_dspark_sps_table_path=None, speculative_dspark_confidence_sts_path=None, speculative_dspark_align_verify_tokens_to_graph_tier=False, speculative_accept_threshold_single=1.0, speculative_accept_threshold_acc=1.0, speculative_use_rejection_sampling=False, speculative_token_map=None, speculative_attention_mode='prefill', speculative_draft_attention_backend=None, speculative_draft_window_size=None, speculative_moe_runner_backend='flashinfer_mxfp4', speculative_moe_a2a_backend=None, speculative_draft_model_quantization=None, speculative_skip_dp_mlp_sync=False, enable_multi_layer_eagle=False, speculative_adaptive=False, speculative_adaptive_config=None, decoupled_spec_bind_endpoint=None, decoupled_spec_connect_endpoints=None, decoupled_spec_rank=None, decoupled_spec_role='null', spec_trace_dir=None, speculative_ngram_min_bfs_breadth=1, speculative_ngram_max_bfs_breadth=10, speculative_ngram_match_type='BFS', speculative_ngram_max_trie_depth=18, speculative_ngram_capacity=10000000, speculative_ngram_external_corpus_path=None, speculative_ngram_external_sam_budget=0, speculative_ngram_external_corpus_max_tokens=10000000, ep_size=4, moe_a2a_backend='none', moe_runner_backend='flashinfer_mxfp4', flashinfer_mxfp4_moe_precision='default', deepep_mode='auto', fuseep_mode=2, deepep_dispatcher_output_dtype='auto', ep_num_redundant_experts=0, ep_dispatch_algorithm=None, init_expert_location='trivial', enable_eplb=False, eplb_algorithm='auto', eplb_rebalance_num_iterations=1000, eplb_rebalance_layers_per_chunk=None, eplb_min_rebalancing_utilization_threshold=1.0, expert_distribution_recorder_mode=None, expert_distribution_recorder_buffer_size=1000, enable_expert_distribution_metrics=False, deepep_config=None, moe_dense_tp_size=None, elastic_ep_backend=None, enable_elastic_expert_backup=False, mooncake_ib_device=None, enable_waterfill=False, ep_join_mode=None, ep_join_rank_offset=0, elastic_ep_initial_size=None, max_ep_size=None, elastic_ep_scale_timeout=600, elastic_ep_rejoin=False, disable_flashinfer_cutlass_moe_fp4_allgather=False, disable_shared_experts_fusion=False, enforce_shared_experts_fusion=False, max_mamba_cache_size=None, mamba_ssm_dtype=None, mamba_max_states_per_path=-1, enable_mamba_cache_stochastic_rounding=False, mamba_cache_philox_rounds=0, mamba_full_memory_ratio=0.36, mamba_radix_cache_strategy='extra_buffer_lazy', uses_mamba_radix_cache=False, mamba_track_interval=256, enable_int8_mamba_checkpoint=False, int8_mamba_ckpt_size=None, linear_attn_backend='triton', linear_attn_decode_backend=None, linear_attn_prefill_backend=None, linear_attn_verify_backend=None, enable_linear_replayssm=False, linear_replayssm_cache_len=16, enable_linear_replayssm_spec=False, enable_hierarchical_cache=False, hicache_ratio=2.0, hicache_size=0, hicache_write_policy='write_through', hicache_io_backend='kernel', hicache_mem_layout='page_first', hicache_storage_backend=None, hicache_storage_prefetch_policy='timeout', hicache_storage_backend_extra_config=None, enable_hisparse=False, hisparse_config=None, enable_broadcast_mm_inputs_process=False, enable_prefix_mm_cache=False, mm_enable_dp_encoder=False, mm_process_config={}, mm_processor_worker_num=0, mm_io_worker_num=0, limit_mm_data_per_request=None, enable_mm_global_cache=False, disable_fast_image_processor=False, mm_feature_transport='cpu', keep_mm_feature_on_device=False, enable_lora=None, enable_lora_overlap_loading=None, max_lora_rank=None, lora_target_modules=None, lora_paths=None, max_loaded_loras=None, max_loras_per_batch=8, lora_eviction_policy='lru', lora_backend='csgmv', max_lora_chunk_size=16, experts_shared_outer_loras=None, lora_use_virtual_experts=False, lora_strict_loading=False, lora_drain_wait_threshold=0.0, enable_two_batch_overlap=False, enable_single_batch_overlap=False, tbo_token_distribution_threshold=0.48, cpu_offload_gb=0, offload_group_size=-1, offload_num_in_group=1, offload_prefetch_step=1, offload_mode='cpu', enable_lmcache=False, lmcache_config_file=None, enable_flexkv=False, flexkv_config_file=None, kt_weight_path=None, kt_method='AMXINT4', kt_cpuinfer=None, kt_threadpool_count=2, kt_num_gpu_experts=None, kt_max_deferred_experts_per_token=None, dllm_algorithm=None, dllm_algorithm_config=None, dllm_fdfo=True, disaggregation_mode='prefill', disaggregation_transfer_backend='mooncake', disaggregation_bootstrap_port=28800, disaggregation_ib_device='mlx5_0,mlx5_1,mlx5_2,mlx5_3', disaggregation_decode_enable_radix_cache=False, disaggregation_decode_enable_offload_kvcache=False, num_reserved_decode_tokens=512, disaggregation_decode_extra_slots=None, disaggregation_decode_polling_interval=1, optimistic_prefill_attempts=0, encoder_only=False, language_only=False, encoder_transfer_backend='zmq_to_tokenizer', encoder_urls=[], encoder_bootstrap_port=8997, encoder_register_urls=[], enable_adaptive_dispatch_to_encoder=False, enable_pdmux=False, pdmux_config_path=None, sm_group_num=8, custom_weight_loader=[], weight_loader_disable_mmap=False, weight_loader_prefetch_checkpoints=False, weight_loader_prefetch_num_threads=4, weight_loader_drop_cache_after_load=False, remote_instance_weight_loader_seed_instance_ip=None, remote_instance_weight_loader_seed_instance_service_port=None, remote_instance_weight_loader_send_weights_group_ports=None, remote_instance_weight_loader_backend='nccl', remote_instance_weight_loader_start_seed_via_transfer_engine=False, engine_info_bootstrap_port=6789, modelexpress_config=None, download_dir=None, model_checksum=None, delete_ckpt_after_loading=False, decrypted_config_file=None, decrypted_draft_config_file=None, checkpoint_engine_wait_weights_before_ready=False, enable_prefill_delayer=False, prefill_delayer_max_delay_passes=30, prefill_delayer_token_usage_low_watermark=None, prefill_delayer_forward_passes_buckets=None, prefill_delayer_wait_seconds_buckets=None, prefill_delayer_queue_min_ratio=None, prefill_delayer_max_delay_ms=None, min_free_slots_delay=None, enable_deterministic_inference=False, rl_on_policy_target=None, kv_canary='none', kv_canary_real_data='none', kv_canary_sweep_interval=0, enable_dynamic_batch_tokenizer=False, dynamic_batch_tokenizer_batch_size=32, dynamic_batch_tokenizer_batch_timeout=0.002, enable_tokenizer_batch_encode=False, disable_tokenizer_batch_decode=False, debug_tensor_dump_output_folder=None, debug_tensor_dump_layers=None, debug_tensor_dump_input_file=None, enable_memory_saver=False, enable_weights_cpu_backup=False, enable_draft_weights_cpu_backup=False, enable_custom_logit_processor=False, enable_return_hidden_states=False, enable_return_routed_experts=False, enable_return_indexer_topk=False, disable_outlines_disk_cache=False, enable_mis=False, weight_cache_mode='off', weight_cache_socket=None, weight_cache_timeout=1800, forward_hooks=None, msprobe_dump_config=None)
+[2026-08-27 04:28:45] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:45] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:45] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:45] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:45] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:45] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:45] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:45] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:01 PP4 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:01 PP4 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:01 PP5 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:01 PP5 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:01 PP4 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:01 PP4 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:01 PP5 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:01 PP5 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:01 PP4 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:01 PP4 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:01 PP4 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:01 PP4 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:02 PP5 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:02 PP5 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:02 PP5 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:02 PP5 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:03 PP5 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:03 PP5 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:03 PP4 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:03 PP4 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:03 PP5 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:03 PP5 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:03 PP4 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:03 PP4 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:05.195727   240 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:05.195828   240 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:05.195843   240 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.3 port: 12001
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:05.195847   236 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:05.195899   240 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.3:15652
+I0827 04:29:05.195935   236 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:05.195948   236 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.3 port: 12001
+I0827 04:29:05.195989   236 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.3:15032
+I0827 04:29:05.196039   240 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:05.196110   236 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:05.206050   241 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:05.206164   241 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:05.206182   241 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.3 port: 12001
+I0827 04:29:05.206233   241 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.3:16487
+I0827 04:29:05.206414   241 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:05.206894   238 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:05.207010   238 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:05.207029   238 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.3 port: 12001
+I0827 04:29:05.207082   238 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.3:15213
+I0827 04:29:05.207248   238 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:05.213657   238 topology.cpp:127] Device mlx5_0 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:05.215921   237 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:05.216045   237 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:05.216063   237 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.3 port: 12001
+I0827 04:29:05.216125   237 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.3:16450
+I0827 04:29:05.216308   237 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:05.219630   235 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:05.219745   235 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:05.219763   235 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.3 port: 12001
+I0827 04:29:05.219815   235 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.3:15658
+I0827 04:29:05.219810   236 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:05.220000   235 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:05.222787   237 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:05.223301   240 topology.cpp:127] Device mlx5_0 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:05.223796   239 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+W0827 04:29:05.223798   242 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:05.223948   242 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:05.223948   239 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:05.223969   242 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.3 port: 12001
+I0827 04:29:05.223973   239 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.3 port: 12001
+I0827 04:29:05.224030   242 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.3:16182
+I0827 04:29:05.224030   239 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.3:15581
+I0827 04:29:05.224237   242 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:05.224242   239 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:05.226227   235 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:05.227162   236 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:05.228888   237 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:05.229261   240 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:05.232141   235 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:05.232787   239 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:05.232861   242 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:05.233434   241 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:05.233938   236 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:05.236400   237 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:05.236641   240 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:05.238986   235 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:05.241036   236 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:05.241355   238 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:05.241638   241 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:05.242061   239 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:05.242151   242 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:05.244179   236 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:05.244223   236 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:05.244287   237 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:05.244297   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.244899   240 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:05.245494   235 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:05.249099   237 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:05.249130   237 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:05.249187   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.251125   235 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:05.251137   240 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:05.251168   235 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:05.251168   240 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:05.251241   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.251241   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.251624   239 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:05.251729   242 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:05.251823   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:05.253651   236 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:03
+I0827 04:29:05.253687   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.256857   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:05.257697   239 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:05.258617   237 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:03
+I0827 04:29:05.258652   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.258757   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:05.258822   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:05.260108   239 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:05.260136   239 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:05.260191   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.260488   235 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:03
+I0827 04:29:05.260502   240 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:03
+I0827 04:29:05.260530   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.260533   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.262187   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:05.264094   236 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:03
+I0827 04:29:05.264137   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.266639   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:05.268234   237 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:03
+I0827 04:29:05.268261   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.268549   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:05.269187   241 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:05.269232   238 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:05.270207   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:05.270285   240 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:03
+I0827 04:29:05.270316   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.271827   235 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:03
+I0827 04:29:05.271884   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.275274   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:05.276836   238 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:05.276907   242 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:05.276996   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:05.277058   236 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:03
+I0827 04:29:05.277101   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.278084   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:05.278632   237 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:03
+I0827 04:29:05.278667   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.279789   240 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:03
+I0827 04:29:05.279817   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.282152   238 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:05.282193   238 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:05.282260   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.283128   242 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:05.283200   242 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:05.283263   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.286593   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:05.287469   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:05.288283   239 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:03
+I0827 04:29:05.288322   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.289064   240 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:03
+I0827 04:29:05.289194   240 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:05 PP5 TP1 EP1] Init torch distributed begin.
+I0827 04:29:05.290841   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:05.292457   242 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:03
+I0827 04:29:05.292488   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.292999   241 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:05.296203   241 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:05.296272   241 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:05.296341   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.300206   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:05.301280   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:05.301775   242 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:03
+I0827 04:29:05.301801   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.303004   235 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:03
+I0827 04:29:05.303042   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.303999   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:05.305142   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:05.305728   241 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:03
+I0827 04:29:05.305768   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.305974   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:05.306758   237 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:03
+I0827 04:29:05.306921   237 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:05 PP4 TP2 EP2] Init torch distributed begin.
+I0827 04:29:05.307497   236 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:03
+I0827 04:29:05.307627   236 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:05 PP4 TP1 EP1] Init torch distributed begin.
+I0827 04:29:05.309624   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:05.309773   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:05.311504   238 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:03
+I0827 04:29:05.311542   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.311590   242 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:03
+I0827 04:29:05.311633   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.313076   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:05.314762   239 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:03
+I0827 04:29:05.314813   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.319522   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:05.319566   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:05.321249   242 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:03
+I0827 04:29:05.321398   242 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:05 PP5 TP3 EP3] Init torch distributed begin.
+I0827 04:29:05.326727   238 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:03
+I0827 04:29:05.326759   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.329339   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:05.331007   235 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:03
+I0827 04:29:05.331142   235 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:05 PP4 TP0 EP0] Init torch distributed begin.
+I0827 04:29:05.334666   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:05.336285   238 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:03
+I0827 04:29:05.336313   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.337296   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:05.338828   241 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:03
+I0827 04:29:05.338877   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.344906   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:05.346731   239 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:03
+I0827 04:29:05.346776   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.365299   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:05.366928   238 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:03
+I0827 04:29:05.367062   238 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:05 PP4 TP3 EP3] Init torch distributed begin.
+I0827 04:29:05.368464   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:05.369644   241 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:03
+I0827 04:29:05.369663   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.377357   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:05.379019   239 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:03
+I0827 04:29:05.379146   239 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:05 PP5 TP0 EP0] Init torch distributed begin.
+I0827 04:29:05.396831   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:05.398531   241 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:03
+I0827 04:29:05.398662   241 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:05 PP5 TP2 EP2] Init torch distributed begin.
+[2026-08-27 04:29:26 PP5 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:26 PP4 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:27 PP4 TP1 EP1] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP4 TP3 EP3] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP4 TP2 EP2] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP4 TP0 EP0] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP5 TP2 EP2] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP5 TP3 EP3] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP5 TP0 EP0] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP5 TP1 EP1] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP5 TP0 EP0] Init torch distributed ends. elapsed=22.37 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP5 TP3 EP3] Init torch distributed ends. elapsed=22.42 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP5 TP2 EP2] Init torch distributed ends. elapsed=22.35 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP4 TP1 EP1] Init torch distributed ends. elapsed=22.44 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP5 TP1 EP1] Init torch distributed ends. elapsed=22.46 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP4 TP3 EP3] Init torch distributed ends. elapsed=22.38 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP4 TP2 EP2] Init torch distributed ends. elapsed=22.44 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP4 TP0 EP0] Init torch distributed ends. elapsed=22.41 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP5 TP0 EP0] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP4 TP1 EP1] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP4 TP3 EP3] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP4 TP2 EP2] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP5 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP4 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP4 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP5 TP2 EP2] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP4 TP0 EP0] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP4 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP5 TP3 EP3] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP5 TP1 EP1] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP4 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP5 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP5 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP5 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP4 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP5 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP4 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP5 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP5 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP4 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP4 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP5 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP5 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP4 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP4 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP4 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP5 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP5 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP4 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP5 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:30:33 PP4 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:33 PP4 TP2 EP2] Load weight end. elapsed=65.50 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:34 PP4 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:34 PP4 TP3 EP3] Load weight end. elapsed=66.61 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP4 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:36 PP4 TP1 EP1] Load weight end. elapsed=68.45 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:36 PP5 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:36 PP5 TP0 EP0] Load weight end. elapsed=69.00 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:37 PP5 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:37 PP5 TP3 EP3] Load weight end. elapsed=69.63 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:37 PP5 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:37 PP5 TP1 EP1] Load weight end. elapsed=69.72 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:38 PP5 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:38 PP5 TP2 EP2] Load weight end. elapsed=70.08 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:38 PP5 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:38 PP5 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:38 PP5 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:38 PP5 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:38 PP5 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:38 PP5 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:38 PP5 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:38 PP5 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:47 PP4 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:47 PP4 TP0 EP0] Load weight end. elapsed=79.31 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:47 PP4 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:47 PP4 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:47 PP4 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:47 PP4 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:47 PP4 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:47 PP4 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:47 PP4 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:47 PP4 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:54 PP5 TP3 EP3] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP5 TP1 EP1] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP5 TP2 EP2] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP5 TP0 EP0] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP4 TP3 EP3] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP4 TP2 EP2] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP4 TP1 EP1] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP4 TP0 EP0] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP5 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP5 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP4 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP4 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP4 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP5 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP4 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP5 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP5 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP5 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP4 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP5 TP3 EP3] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP4 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP4 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP4 TP1 EP1] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP5 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP5 TP2 EP2] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP4 TP3 EP3] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP4 TP0 EP0] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP5 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP5 TP1 EP1] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP5 TP0 EP0] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP4 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP4 TP2 EP2] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP4 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:30:54 PP4 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:30:54 PP4 TP0 EP0] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp0_pp4_dp0.json
+[2026-08-27 04:30:54 PP4 TP1 EP1] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp1_pp4_dp0.json
+[2026-08-27 04:30:54 PP5 TP2 EP2] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp2_pp5_dp0.json
+[2026-08-27 04:30:54 PP5 TP3 EP3] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp3_pp5_dp0.json
+[2026-08-27 04:30:54 PP5 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:30:54 PP5 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:30:54 PP5 TP0 EP0] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp0_pp5_dp0.json
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO[TensorRT-LLM][INFO] Set logger level to INFO
+
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[2026-08-27 04:30:54 PP4 TP2 EP2] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp2_pp4_dp0.json
+[2026-08-27 04:30:54 PP4 TP3 EP3] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp3_pp4_dp0.json
+[2026-08-27 04:30:54 PP5 TP1 EP1] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp1_pp5_dp0.json
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.37profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.12profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.87profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.32profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.91profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.77profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.94profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.88profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.20profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.87profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  7.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 10.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.46profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.67profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.91profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.29profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.49profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  7.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.30profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.94profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.47profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.12profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.06profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.35profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  9.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:01<00:02,  2.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.89profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.58profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.32profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.70profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.20profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.27profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  6.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  7.03profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.56profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.73profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.88profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.37profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:05,  1.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:01<00:02,  2.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.05profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.19profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.77profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.18profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.25profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  3.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.18profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.20profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.37profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  9.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.27profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  8.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.43profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.89profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.35profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.96profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.90profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.30profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.72profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.18profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.33profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.51profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  2.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.03profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.68profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.16profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.01profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.24profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 10.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:02,  2.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.06profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.83profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.35profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.87profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.19profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  3.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.06profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.86profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 13.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 13.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.13profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:01,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:01,  3.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.47profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  6.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.47profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.57profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.22profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.17profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.50profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.71profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.19profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.55profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.90profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:05,  1.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:01<00:06,  1.00s/profile]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.33profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.19profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.03profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.76profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.28profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.32profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.15profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.79profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.31profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.18profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 11.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:01,  3.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.06profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:01,  3.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  6.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  3.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.05profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.60profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.50profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.92profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.00profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.29profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 11.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.09profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.43profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.11profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.91profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  7.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:02,  2.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.18profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.78profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.75profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.89profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.64profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.05profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.47profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.47profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.65profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:05,  1.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.85profile/s]
+[2026-08-27 04:31:28 PP5 TP1 EP1] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.21profile/s]
+[2026-08-27 04:31:28 PP5 TP0 EP0] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.74profile/s]
+[2026-08-27 04:31:28 PP5 TP2 EP2] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.77profile/s]
+[2026-08-27 04:31:28 PP5 TP3 EP3] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.19profile/s][2026-08-27 04:31:28 PP5 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:28 PP5 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:28 PP5 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:28 PP5 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:28 PP5 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:28 PP5 TP0 EP0] max_total_num_tokens=3801408, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=54, context_len=1048576, available_gpu_mem=16.30 GB
+[2026-08-27 04:31:28 PP5 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:28 PP5 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:28 PP5 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.87profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.90profile/s]
+[2026-08-27 04:31:29 PP4 TP3 EP3] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.87profile/s]
+[2026-08-27 04:31:29 PP4 TP2 EP2] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.93profile/s]
+[2026-08-27 04:31:29 PP4 TP0 EP0] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.43profile/s]
+[2026-08-27 04:31:29 PP4 TP1 EP1] FlashInfer autotune completed.
+[2026-08-27 04:31:29 PP4 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:29 PP4 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:29 PP4 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:29 PP4 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:29 PP4 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:29 PP4 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:29 PP4 TP0 EP0] max_total_num_tokens=3801408, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=54, context_len=1048576, available_gpu_mem=16.30 GB
+[2026-08-27 04:31:29 PP4 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:29 PP4 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+W0827 04:31:30.513926  2527 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50a93b80, len: 6912: Operation not permitted [1]
+W0827 04:31:30.514082  2528 memory_location.cpp:73] Failed to get NUMA node, addr: 0x507df700, len: 6912: Operation not permitted [1]
+W0827 04:31:30.514106  2529 memory_location.cpp:73] Failed to get NUMA node, addr: 0x508aa940, len: 6912: Operation not permitted [1]
+W0827 04:31:30.514214  2530 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50f67bc0, len: 6912: Operation not permitted [1]
+W0827 04:31:30.514272  2531 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50a2f680, len: 55296: Operation not permitted [1]
+W0827 04:31:30.514393  2532 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50afecc0, len: 55296: Operation not permitted [1]
+W0827 04:31:30.514506  2533 memory_location.cpp:73] Failed to get NUMA node, addr: 0x502be580, len: 6912: Operation not permitted [1]
+W0827 04:31:30.514607  2534 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5054d3c0, len: 13824: Operation not permitted [1]
+W0827 04:31:30.514652  2535 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50ad4580, len: 6912: Operation not permitted [1]
+W0827 04:31:30.514743  2536 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50eb9880, len: 6912: Operation not permitted [1]
+W0827 04:31:31.680995  2556 memory_location.cpp:73] Failed to get NUMA node, addr: 0x501c3300, len: 6912: Operation not permitted [1]
+W0827 04:31:31.680985  2557 memory_location.cpp:73] Failed to get NUMA node, addr: 0x504ecec0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.680985  2555 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd4bac0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681169  2558 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50706900, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681272  2559 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5006d700, len: 55296: Operation not permitted [1]
+W0827 04:31:31.681360  2560 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50041cc0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.681468  2561 memory_location.cpp:73] Failed to get NUMA node, addr: 0x506f5a40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681562  2562 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fdfdf40, len: 13824: Operation not permitted [1]
+W0827 04:31:31.681607  2563 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd16080, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681685  2564 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fcfa300, len: 6912: Operation not permitted [1]
+W0827 04:31:31.733497  2583 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51554040, len: 6912: Operation not permitted [1]
+W0827 04:31:31.733552  2584 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5110a8c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.733618  2585 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51ca8e80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.733795  2586 memory_location.cpp:73] Failed to get NUMA node, addr: 0x52017680, len: 6912: Operation not permitted [1]
+W0827 04:31:31.733909  2587 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5174af80, len: 55296: Operation not permitted [1]
+W0827 04:31:31.734004  2588 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51770440, len: 55296: Operation not permitted [1]
+W0827 04:31:31.734061  2589 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51118000, len: 6912: Operation not permitted [1]
+W0827 04:31:31.734179  2590 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51564f40, len: 13824: Operation not permitted [1]
+W0827 04:31:31.734277  2591 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51f69480, len: 6912: Operation not permitted [1]
+W0827 04:31:31.734400  2592 memory_location.cpp:73] Failed to get NUMA node, addr: 0x517c2940, len: 6912: Operation not permitted [1]
+W0827 04:31:31.741586  2611 memory_location.cpp:73] Failed to get NUMA node, addr: 0x519520c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.741609  2612 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50ff8c40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.741679  2613 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51a2ad80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.741792  2614 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51b5b300, len: 6912: Operation not permitted [1]
+W0827 04:31:31.741938  2615 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50fe5b80, len: 55296: Operation not permitted [1]
+W0827 04:31:31.742043  2616 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51625140, len: 55296: Operation not permitted [1]
+W0827 04:31:31.742168  2617 memory_location.cpp:73] Failed to get NUMA node, addr: 0x516bf640, len: 6912: Operation not permitted [1]
+W0827 04:31:31.742278  2618 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51b73e80, len: 13824: Operation not permitted [1]
+W0827 04:31:31.742414  2619 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51ef60c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.742429  2620 memory_location.cpp:73] Failed to get NUMA node, addr: 0x513e95c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.900663  2641 memory_location.cpp:73] Failed to get NUMA node, addr: 0x517e8c80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.900679  2640 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5170b840, len: 6912: Operation not permitted [1]
+W0827 04:31:31.900738  2642 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5131d0c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.900669  2639 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51372400, len: 6912: Operation not permitted [1]
+W0827 04:31:31.900923  2643 memory_location.cpp:73] Failed to get NUMA node, addr: 0x512b7e00, len: 55296: Operation not permitted [1]
+W0827 04:31:31.900976  2644 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51380200, len: 55296: Operation not permitted [1]
+W0827 04:31:31.901000  2646 memory_location.cpp:73] Failed to get NUMA node, addr: 0x513151c0, len: 13824: Operation not permitted [1]
+W0827 04:31:31.901011  2645 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51194f40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.901038  2647 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51520740, len: 6912: Operation not permitted [1]
+W0827 04:31:31.901073  2648 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51724980, len: 6912: Operation not permitted [1]
+W0827 04:31:31.931138  2668 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51061000, len: 6912: Operation not permitted [1]
+W0827 04:31:31.931178  2669 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50d74c00, len: 6912: Operation not permitted [1]
+W0827 04:31:31.931098  2667 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50b2fb40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.931272  2670 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50de9000, len: 6912: Operation not permitted [1]
+W0827 04:31:31.931349  2671 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5097dc80, len: 55296: Operation not permitted [1]
+W0827 04:31:31.931519  2672 memory_location.cpp:73] Failed to get NUMA node, addr: 0x515aa000, len: 55296: Operation not permitted [1]
+W0827 04:31:31.931557  2675 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51269700, len: 6912: Operation not permitted [1]
+W0827 04:31:31.931556  2673 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5097bd00, len: 6912: Operation not permitted [1]
+W0827 04:31:31.931576  2674 memory_location.cpp:73] Failed to get NUMA node, addr: 0x514ee800, len: 13824: Operation not permitted [1]
+W0827 04:31:31.931759  2676 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50deac80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.983201  2696 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50819380, len: 6912: Operation not permitted [1]
+W0827 04:31:31.983201  2695 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50600a80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.983378  2697 memory_location.cpp:73] Failed to get NUMA node, addr: 0x506ddf40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.983438  2698 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50716800, len: 6912: Operation not permitted [1]
+W0827 04:31:31.983490  2699 memory_location.cpp:73] Failed to get NUMA node, addr: 0x501575c0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.983624  2700 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50863340, len: 55296: Operation not permitted [1]
+W0827 04:31:31.983737  2701 memory_location.cpp:73] Failed to get NUMA node, addr: 0x508d7a40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.983930  2702 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50668c40, len: 13824: Operation not permitted [1]
+W0827 04:31:31.984035  2703 memory_location.cpp:73] Failed to get NUMA node, addr: 0x506841c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.984107  2704 memory_location.cpp:73] Failed to get NUMA node, addr: 0x508f16c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.001853  2725 memory_location.cpp:73] Failed to get NUMA node, addr: 0x507c8d40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.001852  2724 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50b39240, len: 6912: Operation not permitted [1]
+W0827 04:31:32.002048  2727 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50a8ecc0, len: 55296: Operation not permitted [1]
+W0827 04:31:32.001888  2726 memory_location.cpp:73] Failed to get NUMA node, addr: 0x512a8140, len: 6912: Operation not permitted [1]
+W0827 04:31:32.002044  2728 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50a68880, len: 55296: Operation not permitted [1]
+W0827 04:31:32.002257  2729 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5087b880, len: 6912: Operation not permitted [1]
+W0827 04:31:32.002264  2731 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50fc6980, len: 6912: Operation not permitted [1]
+W0827 04:31:32.002266  2730 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51331100, len: 13824: Operation not permitted [1]
+W0827 04:31:32.002470  2732 memory_location.cpp:73] Failed to get NUMA node, addr: 0x513384c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.002780  2723 memory_location.cpp:73] Failed to get NUMA node, addr: 0x512c6940, len: 6912: Operation not permitted [1]
+[2026-08-27 04:31:32] Dummy health check server started in background thread at 0.0.0.0:30000
+[rank16]:[W827 04:33:25.972222579 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank17]:[W827 04:33:25.980800100 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank19]:[W827 04:33:25.981413562 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank18]:[W827 04:33:26.644294178 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank16]:[W827 04:34:10.467297472 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank18]:[W827 04:34:10.467837454 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank17]:[W827 04:34:10.468151853 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank19]:[W827 04:34:10.468427857 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank22]:[W827 04:34:10.495396864 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank23]:[W827 04:34:10.495456053 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank21]:[W827 04:34:10.495795969 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank20]:[W827 04:34:10.495814123 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank22]:[W827 04:34:16.703081720 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank20]:[W827 04:34:16.703915664 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank21]:[W827 04:34:16.704033167 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank23]:[W827 04:34:16.704314444 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[2026-08-27 04:35:10 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.29
+[2026-08-27 04:35:10 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.29
+[2026-08-27 04:35:29 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.39
+[2026-08-27 04:35:29 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.39
+[2026-08-27 04:35:31 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25.89
+[2026-08-27 04:35:31 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25.91
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/d_healthy_p_node3.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/d_healthy_p_node3.log
new file mode 100644
index 0000000..09a7f9c
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/d_healthy_p_node3.log
@@ -0,0 +1,757 @@
+/usr/local/lib/python3.12/dist-packages/torchao/quantization/quant_api.py:1731: SyntaxWarning: invalid escape sequence '\.'
+  """Configuration class for applying different quantization configs to modules or parameters based on their fully qualified names (FQNs).
+/sgl-workspace/sglang/python/sglang/launch_server.py:56: UserWarning: 'python -m sglang.launch_server' is still supported, but 'sglang serve' is the recommended entrypoint.
+  Example: sglang serve --model-path <model> [options]
+  warnings.warn(
+[2026-08-27 04:28:46] Breakable CUDA graph is incompatible with MLA attention; disabling prefill CUDA graph.
+[2026-08-27 04:28:47] Attention backend not specified. Use triton backend by default.
+[2026-08-27 04:28:47] Pipeline parallelism is incompatible with overlap schedule.
+[2026-08-27 04:28:49] server_args=ServerArgs(model_path='/data/hf_models/Kimi-K3', tokenizer_path='/data/hf_models/Kimi-K3', tokenizer_mode='auto', tokenizer_backend='huggingface', tokenizer_worker_num=1, detokenizer_worker_num=1, skip_tokenizer_init=False, load_format='auto', model_loader_extra_config='{}', trust_remote_code=True, context_length=None, is_embedding=False, enable_multimodal=None, revision=None, model_impl='auto', model_config_parser='auto', json_model_override_args='{}', dtype='auto', quantization=None, quantization_param_path=None, kv_cache_dtype='auto', enable_fp32_lm_head=False, modelopt_quant=None, modelopt_checkpoint_restore_path=None, modelopt_checkpoint_save_path=None, modelopt_export_path=None, quantize_and_serve=False, rl_quant_profile=None, enable_tf32_matmul=False, mem_fraction_static=0.88, max_running_requests=None, max_queued_requests=None, max_total_tokens=None, chunked_prefill_size=8192, enable_dynamic_chunking=False, max_prefill_tokens=16384, prefill_max_requests=None, schedule_policy='fcfs', enable_priority_scheduling=False, disable_priority_preemption=False, default_priority_value=None, abort_on_priority_when_disabled=False, schedule_low_priority_values_first=False, priority_scheduling_preemption_threshold=10, retraction_policy='length', schedule_conservativeness=1.0, page_size=64, swa_full_tokens_ratio=0.8, disable_hybrid_swa_memory=False, radix_eviction_policy='lru', prefill_only_disable_kv_cache=False, disable_radix_cache=True, enable_page_major_kv_layout=False, enable_unified_memory=False, disable_chunked_prefix_cache=False, disable_overlap_schedule=True, num_continuous_decode_steps=1, scheduler_recv_interval=1, enable_mixed_chunk=False, nccl_port=None, dist_timeout=3600, dist_init_addr='174.1.60.1:20000', nnodes=4, node_rank=3, tp_size=4, dcp_size=1, pp_size=8, pp_max_micro_batch_size=None, pp_async_batch_depth=0, dp_size=1, load_balance_method='follow_bootstrap_room', attn_cp_size=1, moe_dp_size=1, dcp_comm_backend='ag_rs', dcp_replicate_q_proj=None, dwdp_size=1, enable_prefill_cp=False, cp_strategy=None, enable_dsa_cache_layer_split=False, enable_dsa_prefill_context_parallel=False, dsa_prefill_cp_mode='round-robin-split', enable_prefill_context_parallel=False, prefill_cp_mode='in-seq-split', enable_cp_decode_attn_tp=False, enable_dp_attention=False, enable_dp_attention_local_control_broadcast=False, enable_dp_lm_head=False, enable_attn_tp_input_scattered=False, disable_attn_tp_gather=False, enable_p2p_check=False, device='cuda', base_gpu_id=0, gpu_id_step=1, random_seed=927276655, watchdog_timeout=300, soft_watchdog_timeout=None, sleep_on_idle=False, use_ray=False, custom_sigquit_handler=None, numa_node=None, gc_threshold=None, host='0.0.0.0', port=30000, fastapi_root_path='', smg_grpc_mode=False, grpc_mode=False, grpc_port=None, sidecar=None, sidecar_args=None, skip_server_warmup=False, warmups=None, enable_http2=False, ssl_keyfile=None, ssl_certfile=None, ssl_ca_certs=None, ssl_keyfile_password=None, enable_ssl_refresh=False, api_key=None, admin_api_key=None, served_model_name='kimi-k3', weight_version='default', chat_template=None, hf_chat_template_name=None, completion_template=None, file_storage_path='sglang_storage', enable_cache_report=False, reasoning_parser=None, default_chat_template_kwargs=None, strip_thinking_cache=False, enable_strict_thinking=False, tool_call_parser=None, tool_server=None, sampling_defaults='model', asr_max_buffer_seconds=60, asr_max_concurrent_sessions=32, preferred_sampling_params=None, allow_auto_truncate=False, stream_interval=1, batch_notify_size=16, stream_response_default_include_usage=False, incremental_streaming_output=False, enable_streaming_session=False, enable_session_radix_cache=False, log_level='info', log_level_http=None, log_requests=False, log_requests_level=2, log_requests_format='text', log_requests_target=None, uvicorn_access_log_exclude_prefixes=[], crash_dump_folder=None, show_time_cost=False, enable_metrics=False, smg_http_sidecar_port=None, enable_mfu_metrics=False, enable_metrics_for_all_schedulers=False, load_snapshot_publish_interval=15, tokenizer_metrics_custom_labels_header='x-custom-labels', tokenizer_metrics_allowed_custom_labels=None, extra_metric_labels=None, bucket_time_to_first_token=None, bucket_inter_token_latency=None, bucket_e2e_request_latency=None, prompt_tokens_buckets=None, generation_tokens_buckets=None, gc_warning_threshold_secs=0.0, decode_log_interval=40, enable_request_time_stats_logging=False, kv_events_config=None, enable_forward_pass_metrics=False, forward_pass_metrics_worker_id='', forward_pass_metrics_ipc_name=None, enable_trace=False, trace_modules='request', otlp_traces_endpoint='localhost:4317', export_metrics_to_file=False, export_metrics_to_file_dir=None, stat_loggers=None, constrained_json_whitespace_pattern=None, constrained_json_disable_any_whitespace=False, attention_backend='triton', decode_attention_backend=None, prefill_attention_backend=None, sampling_backend='flashinfer', grammar_backend='xgrammar', radix_cache_backend=None, mm_attention_backend=None, fp8_gemm_runner_backend='auto', fp4_gemm_runner_backend='auto', bf16_gemm_backend='auto', dsa_prefill_backend=None, dsa_decode_backend=None, dsa_paged_mqa_logits_backend='auto', dsa_topk_backend='sgl-kernel', disable_flashinfer_autotune=False, flashinfer_autotune_skip_ops=None, mamba_backend='triton', cuda_graph_config=CudaGraphConfig(decode=PhaseConfig(backend='disabled', max_bs=16, bs=[1, 2, 4, 8, 12, 16], tc_compiler='eager', full_prefill_max_req=None), prefill=PhaseConfig(backend='disabled', max_bs=2048, bs=[4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], tc_compiler='eager', full_prefill_max_req=None)), cuda_graph_backend_decode=None, cuda_graph_backend_prefill=None, cuda_graph_max_bs_decode=16, cuda_graph_max_bs_prefill=None, cuda_graph_bs_decode=None, cuda_graph_bs_prefill=None, cuda_graph_tc_compiler=None, disable_prefill_cuda_graph=False, disable_decode_cuda_graph=False, disable_cuda_graph=False, disable_cuda_graph_padding=False, enable_profile_cuda_graph=False, enable_cudagraph_gc=False, debug_cuda_graph=False, enable_layerwise_nvtx_marker=False, enable_nccl_nvls=False, enable_symm_mem=False, triton_attention_reduce_in_fp32=False, triton_attention_num_kv_splits=8, triton_attention_split_tile_size=None, flashinfer_mla_disable_ragged=False, enable_fused_qk_norm_rope=False, enable_precise_embedding_interpolation=False, enable_fused_moe_sum_all_reduce=False, enable_deepseek_v4_fp4_indexer=False, disable_custom_all_reduce=False, enable_mscclpp=False, enable_torch_symm_mem=False, enable_scattered_sconv=False, pre_warm_nccl=False, enable_quant_communications=False, enable_flashinfer_allreduce_fusion=False, enforce_disable_flashinfer_allreduce_fusion=False, flashinfer_allreduce_fusion_backend=None, enable_aiter_allreduce_fusion=False, enable_torch_compile=False, enable_torch_compile_debug_mode=False, torch_compile_max_bs=32, torchao_config='', speculative_algorithm=None, speculative_draft_model_path=None, speculative_draft_model_revision=None, speculative_draft_load_format=None, speculative_num_steps=None, speculative_eagle_topk=None, speculative_num_draft_tokens=None, speculative_dflash_block_size=None, speculative_dspark_block_size=None, speculative_dspark_sps_table_path=None, speculative_dspark_confidence_sts_path=None, speculative_dspark_align_verify_tokens_to_graph_tier=False, speculative_accept_threshold_single=1.0, speculative_accept_threshold_acc=1.0, speculative_use_rejection_sampling=False, speculative_token_map=None, speculative_attention_mode='prefill', speculative_draft_attention_backend=None, speculative_draft_window_size=None, speculative_moe_runner_backend='flashinfer_mxfp4', speculative_moe_a2a_backend=None, speculative_draft_model_quantization=None, speculative_skip_dp_mlp_sync=False, enable_multi_layer_eagle=False, speculative_adaptive=False, speculative_adaptive_config=None, decoupled_spec_bind_endpoint=None, decoupled_spec_connect_endpoints=None, decoupled_spec_rank=None, decoupled_spec_role='null', spec_trace_dir=None, speculative_ngram_min_bfs_breadth=1, speculative_ngram_max_bfs_breadth=10, speculative_ngram_match_type='BFS', speculative_ngram_max_trie_depth=18, speculative_ngram_capacity=10000000, speculative_ngram_external_corpus_path=None, speculative_ngram_external_sam_budget=0, speculative_ngram_external_corpus_max_tokens=10000000, ep_size=4, moe_a2a_backend='none', moe_runner_backend='flashinfer_mxfp4', flashinfer_mxfp4_moe_precision='default', deepep_mode='auto', fuseep_mode=2, deepep_dispatcher_output_dtype='auto', ep_num_redundant_experts=0, ep_dispatch_algorithm=None, init_expert_location='trivial', enable_eplb=False, eplb_algorithm='auto', eplb_rebalance_num_iterations=1000, eplb_rebalance_layers_per_chunk=None, eplb_min_rebalancing_utilization_threshold=1.0, expert_distribution_recorder_mode=None, expert_distribution_recorder_buffer_size=1000, enable_expert_distribution_metrics=False, deepep_config=None, moe_dense_tp_size=None, elastic_ep_backend=None, enable_elastic_expert_backup=False, mooncake_ib_device=None, enable_waterfill=False, ep_join_mode=None, ep_join_rank_offset=0, elastic_ep_initial_size=None, max_ep_size=None, elastic_ep_scale_timeout=600, elastic_ep_rejoin=False, disable_flashinfer_cutlass_moe_fp4_allgather=False, disable_shared_experts_fusion=False, enforce_shared_experts_fusion=False, max_mamba_cache_size=None, mamba_ssm_dtype=None, mamba_max_states_per_path=-1, enable_mamba_cache_stochastic_rounding=False, mamba_cache_philox_rounds=0, mamba_full_memory_ratio=0.36, mamba_radix_cache_strategy='extra_buffer_lazy', uses_mamba_radix_cache=False, mamba_track_interval=256, enable_int8_mamba_checkpoint=False, int8_mamba_ckpt_size=None, linear_attn_backend='triton', linear_attn_decode_backend=None, linear_attn_prefill_backend=None, linear_attn_verify_backend=None, enable_linear_replayssm=False, linear_replayssm_cache_len=16, enable_linear_replayssm_spec=False, enable_hierarchical_cache=False, hicache_ratio=2.0, hicache_size=0, hicache_write_policy='write_through', hicache_io_backend='kernel', hicache_mem_layout='page_first', hicache_storage_backend=None, hicache_storage_prefetch_policy='timeout', hicache_storage_backend_extra_config=None, enable_hisparse=False, hisparse_config=None, enable_broadcast_mm_inputs_process=False, enable_prefix_mm_cache=False, mm_enable_dp_encoder=False, mm_process_config={}, mm_processor_worker_num=0, mm_io_worker_num=0, limit_mm_data_per_request=None, enable_mm_global_cache=False, disable_fast_image_processor=False, mm_feature_transport='cpu', keep_mm_feature_on_device=False, enable_lora=None, enable_lora_overlap_loading=None, max_lora_rank=None, lora_target_modules=None, lora_paths=None, max_loaded_loras=None, max_loras_per_batch=8, lora_eviction_policy='lru', lora_backend='csgmv', max_lora_chunk_size=16, experts_shared_outer_loras=None, lora_use_virtual_experts=False, lora_strict_loading=False, lora_drain_wait_threshold=0.0, enable_two_batch_overlap=False, enable_single_batch_overlap=False, tbo_token_distribution_threshold=0.48, cpu_offload_gb=0, offload_group_size=-1, offload_num_in_group=1, offload_prefetch_step=1, offload_mode='cpu', enable_lmcache=False, lmcache_config_file=None, enable_flexkv=False, flexkv_config_file=None, kt_weight_path=None, kt_method='AMXINT4', kt_cpuinfer=None, kt_threadpool_count=2, kt_num_gpu_experts=None, kt_max_deferred_experts_per_token=None, dllm_algorithm=None, dllm_algorithm_config=None, dllm_fdfo=True, disaggregation_mode='prefill', disaggregation_transfer_backend='mooncake', disaggregation_bootstrap_port=28800, disaggregation_ib_device='mlx5_0,mlx5_1,mlx5_2,mlx5_3', disaggregation_decode_enable_radix_cache=False, disaggregation_decode_enable_offload_kvcache=False, num_reserved_decode_tokens=512, disaggregation_decode_extra_slots=None, disaggregation_decode_polling_interval=1, optimistic_prefill_attempts=0, encoder_only=False, language_only=False, encoder_transfer_backend='zmq_to_tokenizer', encoder_urls=[], encoder_bootstrap_port=8997, encoder_register_urls=[], enable_adaptive_dispatch_to_encoder=False, enable_pdmux=False, pdmux_config_path=None, sm_group_num=8, custom_weight_loader=[], weight_loader_disable_mmap=False, weight_loader_prefetch_checkpoints=False, weight_loader_prefetch_num_threads=4, weight_loader_drop_cache_after_load=False, remote_instance_weight_loader_seed_instance_ip=None, remote_instance_weight_loader_seed_instance_service_port=None, remote_instance_weight_loader_send_weights_group_ports=None, remote_instance_weight_loader_backend='nccl', remote_instance_weight_loader_start_seed_via_transfer_engine=False, engine_info_bootstrap_port=6789, modelexpress_config=None, download_dir=None, model_checksum=None, delete_ckpt_after_loading=False, decrypted_config_file=None, decrypted_draft_config_file=None, checkpoint_engine_wait_weights_before_ready=False, enable_prefill_delayer=False, prefill_delayer_max_delay_passes=30, prefill_delayer_token_usage_low_watermark=None, prefill_delayer_forward_passes_buckets=None, prefill_delayer_wait_seconds_buckets=None, prefill_delayer_queue_min_ratio=None, prefill_delayer_max_delay_ms=None, min_free_slots_delay=None, enable_deterministic_inference=False, rl_on_policy_target=None, kv_canary='none', kv_canary_real_data='none', kv_canary_sweep_interval=0, enable_dynamic_batch_tokenizer=False, dynamic_batch_tokenizer_batch_size=32, dynamic_batch_tokenizer_batch_timeout=0.002, enable_tokenizer_batch_encode=False, disable_tokenizer_batch_decode=False, debug_tensor_dump_output_folder=None, debug_tensor_dump_layers=None, debug_tensor_dump_input_file=None, enable_memory_saver=False, enable_weights_cpu_backup=False, enable_draft_weights_cpu_backup=False, enable_custom_logit_processor=False, enable_return_hidden_states=False, enable_return_routed_experts=False, enable_return_indexer_topk=False, disable_outlines_disk_cache=False, enable_mis=False, weight_cache_mode='off', weight_cache_socket=None, weight_cache_timeout=1800, forward_hooks=None, msprobe_dump_config=None)
+[2026-08-27 04:28:49] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:49] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:49] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:49] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:49] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:49] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:49] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:49] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:05 PP6 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:05 PP6 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:07 PP6 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:07 PP6 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:07 PP6 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:07 PP7 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:07 PP7 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:07 PP7 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:07 PP7 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:07 PP6 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:07 PP6 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:07.662256   238 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:07.662379   238 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:07.662392   238 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.4 port: 12001
+I0827 04:29:07.662429   238 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.4:15914
+I0827 04:29:07.662514   238 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:07.685189   238 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:07.689950   238 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:07.695539   238 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:07.701294   238 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:07.703691   238 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:07.703715   238 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:07.703760   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:07.710492   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:07.711953   238 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:04
+I0827 04:29:07.711989   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:07.735716   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:07.737195   238 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:04
+I0827 04:29:07.737223   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:07.763653   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:07.765146   238 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:04
+I0827 04:29:07.765172   238 rdma_context.cpp:205] Using SIEVE endpoint store
+[2026-08-27 04:29:07 PP7 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:07 PP7 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:07 PP7 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:07 PP7 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+I0827 04:29:07.791589   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:07.793025   238 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:04
+I0827 04:29:07.793139   238 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:07 PP6 TP3 EP3] Init torch distributed begin.
+[2026-08-27 04:29:07 PP6 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:07 PP6 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:08 PP6 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:08.916008   235 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:08.916072   235 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:08.916082   235 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.4 port: 12001
+I0827 04:29:08.916121   235 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.4:15737
+I0827 04:29:08.916215   235 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:08.941840   235 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:08.950320   235 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:08.979278   235 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:08.988204   235 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:08.991855   235 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:08.991879   235 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:08.991928   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:08.999985   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:09.001416   235 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:04
+I0827 04:29:09.001446   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:09.027477   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:09.028954   235 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:04
+I0827 04:29:09.028980   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:09.055524   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:09.056981   235 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:04
+I0827 04:29:09.057020   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:09.083544   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:09.084911   235 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:04
+I0827 04:29:09.085031   235 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:09 PP6 TP0 EP0] Init torch distributed begin.
+[2026-08-27 04:29:09 PP7 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:09 PP6 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:09 PP7 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:09 PP6 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:09 PP7 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:09 PP7 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:10.000486   242 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:10.000609   242 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:10.000628   242 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.4 port: 12001
+I0827 04:29:10.000689   242 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.4:16890
+I0827 04:29:10.000851   242 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:10.006592   236 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:10.006711   236 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:10.006728   236 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.4 port: 12001
+I0827 04:29:10.006786   236 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.4:15893
+I0827 04:29:10.006963   236 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:10.022238   241 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:10.022354   241 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:10.022370   241 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.4 port: 12001
+I0827 04:29:10.022423   241 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.4:16277
+I0827 04:29:10.022593   241 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:10.032052   242 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:10.032101   236 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:10.039772   236 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:10.039858   242 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:10.047292   236 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:10.047341   242 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:10.047785   241 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:10.054806   242 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:10.054838   236 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:10.055280   241 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:10.059140   236 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:10.059171   236 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:10.059175   242 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:10.059201   242 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:10.059227   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.059260   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.062610   241 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:10.066787   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:10.066826   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:10.068500   242 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:04
+I0827 04:29:10.068532   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.068558   236 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:04
+I0827 04:29:10.068590   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.069866   241 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:10.072566   241 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:10.072592   241 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:10.072649   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.080149   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:10.081809   241 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:04
+I0827 04:29:10.081843   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.095701   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:10.097368   236 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:04
+I0827 04:29:10.097400   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.099974   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:10.101672   242 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:04
+I0827 04:29:10.101717   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.107705   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:10.109316   241 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:04
+I0827 04:29:10.109347   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.123868   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:10.125504   236 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:04
+I0827 04:29:10.125536   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.127957   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:10.129673   242 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:04
+I0827 04:29:10.129703   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.139813   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:10.141404   241 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:04
+I0827 04:29:10.141431   241 rdma_context.cpp:205] Using SIEVE endpoint store
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:10.150050   237 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:10.150159   237 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:10.150173   237 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.4 port: 12001
+I0827 04:29:10.150219   237 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.4:15413
+I0827 04:29:10.150358   237 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:10.151880   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:10.153475   236 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:04
+I0827 04:29:10.153601   236 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:10 PP6 TP1 EP1] Init torch distributed begin.
+I0827 04:29:10.155849   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:10.155864   237 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:10.157462   242 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:04
+I0827 04:29:10.157585   242 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:10 PP7 TP3 EP3] Init torch distributed begin.
+I0827 04:29:10.171923   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:10.172516   240 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:10.172667   240 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:10.172688   240 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.4 port: 12001
+I0827 04:29:10.172762   240 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.4:15705
+I0827 04:29:10.172977   240 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:10.173573   241 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:04
+I0827 04:29:10.173692   241 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:10 PP7 TP2 EP2] Init torch distributed begin.
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:10.177717   239 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:10.177865   239 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:10.177886   239 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.4 port: 12001
+I0827 04:29:10.177956   239 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.4:16192
+I0827 04:29:10.178146   239 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:10.183638   237 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:10.187090   239 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:10.191001   237 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:10.195258   239 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:10.198271   237 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:10.200984   237 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:10.201025   237 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:10.201079   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.203014   239 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:10.203156   240 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:10.209589   240 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:10.210305   239 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:10.212793   239 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:10.212821   239 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:10.212878   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.216092   240 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:10.219946   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:10.221297   239 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:04
+I0827 04:29:10.221323   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.222187   240 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:10.224284   240 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:10.224304   240 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:10.224344   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.230640   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:10.231778   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:10.231937   240 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:04
+I0827 04:29:10.231962   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.233362   237 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:04
+I0827 04:29:10.233397   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.236478   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:10.237651   240 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:04
+I0827 04:29:10.237670   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.250197   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:10.251711   239 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:04
+I0827 04:29:10.251737   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.263743   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:10.265311   237 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:04
+I0827 04:29:10.265339   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.267139   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:10.268607   240 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:04
+I0827 04:29:10.268632   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.273075   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:10.274682   237 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:04
+I0827 04:29:10.274710   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.276602   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:10.278275   239 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:04
+I0827 04:29:10.278314   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.295709   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:10.297385   240 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:04
+I0827 04:29:10.297510   240 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:10 PP7 TP1 EP1] Init torch distributed begin.
+I0827 04:29:10.299893   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:10.301563   237 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:04
+I0827 04:29:10.301689   237 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:10 PP6 TP2 EP2] Init torch distributed begin.
+I0827 04:29:10.303737   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:10.305307   239 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:04
+I0827 04:29:10.305433   239 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:10 PP7 TP0 EP0] Init torch distributed begin.
+[2026-08-27 04:29:26 PP7 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:26 PP6 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:27 PP6 TP0 EP0] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP6 TP3 EP3] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP6 TP2 EP2] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP6 TP1 EP1] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP7 TP1 EP1] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP7 TP0 EP0] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP7 TP3 EP3] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP7 TP2 EP2] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP7 TP1 EP1] Init torch distributed ends. elapsed=17.45 s, mem usage=0.46 GB
+[2026-08-27 04:29:27 PP7 TP3 EP3] Init torch distributed ends. elapsed=17.59 s, mem usage=0.46 GB
+[2026-08-27 04:29:27 PP7 TP2 EP2] Init torch distributed ends. elapsed=17.57 s, mem usage=0.46 GB
+[2026-08-27 04:29:27 PP7 TP0 EP0] Init torch distributed ends. elapsed=17.44 s, mem usage=0.46 GB
+[2026-08-27 04:29:27 PP6 TP1 EP1] Init torch distributed ends. elapsed=17.59 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP6 TP3 EP3] Init torch distributed ends. elapsed=19.95 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP6 TP0 EP0] Init torch distributed ends. elapsed=18.66 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP6 TP2 EP2] Init torch distributed ends. elapsed=17.44 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP6 TP1 EP1] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP7 TP1 EP1] Load weight begin. avail mem=82.12 GB
+[2026-08-27 04:29:27 PP6 TP0 EP0] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP6 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP7 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP7 TP3 EP3] Load weight begin. avail mem=82.12 GB
+[2026-08-27 04:29:27 PP6 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP7 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP6 TP2 EP2] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP6 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP7 TP2 EP2] Load weight begin. avail mem=82.12 GB
+[2026-08-27 04:29:28 PP7 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP6 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP7 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP6 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP7 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP7 TP0 EP0] Load weight begin. avail mem=82.12 GB
+[2026-08-27 04:29:28 PP6 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP7 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP7 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP6 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP7 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP6 TP3 EP3] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP6 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP6 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP6 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP7 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP7 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP7 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP6 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP7 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP6 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:30:40 PP7 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:40 PP7 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:40 PP7 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:40 PP7 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:40 PP7 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:40 PP7 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:40 PP7 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:40 PP7 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:41 PP7 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:41 PP7 TP1 EP1] Load weight end. elapsed=73.17 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=31.91 GB, mem usage=50.21 GB.
+[2026-08-27 04:30:41 PP7 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:41 PP7 TP0 EP0] Load weight end. elapsed=73.80 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=31.91 GB, mem usage=50.21 GB.
+[2026-08-27 04:30:42 PP6 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:42 PP6 TP1 EP1] Load weight end. elapsed=74.47 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:45 PP6 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:45 PP6 TP2 EP2] Load weight end. elapsed=77.20 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:50 PP7 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:50 PP7 TP3 EP3] Load weight end. elapsed=82.79 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=31.91 GB, mem usage=50.21 GB.
+[2026-08-27 04:30:51 PP6 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:51 PP6 TP3 EP3] Load weight end. elapsed=83.19 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:51 PP7 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:51 PP7 TP2 EP2] Load weight end. elapsed=83.68 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=31.91 GB, mem usage=50.21 GB.
+[2026-08-27 04:30:51 PP7 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:51 PP7 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:52 PP7 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:52 PP7 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:52 PP7 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:52 PP7 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:52 PP7 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:52 PP7 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:53 PP6 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:53 PP6 TP0 EP0] Load weight end. elapsed=85.47 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:53 PP6 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:53 PP6 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:53 PP6 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:53 PP6 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:53 PP6 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:53 PP6 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:53 PP6 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:53 PP6 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:54 PP7 TP3 EP3] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP7 TP2 EP2] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP7 TP1 EP1] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP7 TP0 EP0] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP6 TP3 EP3] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP6 TP2 EP2] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP6 TP1 EP1] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP6 TP0 EP0] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP6 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP6 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP7 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:54 PP6 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP7 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:54 PP7 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:54 PP7 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:54 PP6 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP6 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP6 TP3 EP3] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP6 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP6 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP6 TP2 EP2] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP6 TP1 EP1] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP7 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 16.31 GB
+[2026-08-27 04:30:54 PP7 TP2 EP2] Memory pool end. avail mem=14.95 GB
+[2026-08-27 04:30:54 PP6 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP6 TP0 EP0] Memory pool end. avail mem=19.33 GB
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[2026-08-27 04:30:54 PP7 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 16.31 GB
+[2026-08-27 04:30:54 PP7 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 16.31 GB
+[2026-08-27 04:30:54 PP7 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 16.31 GB
+[2026-08-27 04:30:54 PP7 TP3 EP3] Memory pool end. avail mem=14.95 GB
+[2026-08-27 04:30:54 PP7 TP0 EP0] Memory pool end. avail mem=14.95 GB
+[2026-08-27 04:30:54 PP7 TP1 EP1] Memory pool end. avail mem=14.95 GB
+[2026-08-27 04:30:54 PP7 TP2 EP2] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp2_pp7_dp0.json
+[2026-08-27 04:30:54 PP6 TP3 EP3] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp3_pp6_dp0.json
+[2026-08-27 04:30:54 PP6 TP1 EP1] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp1_pp6_dp0.json
+[2026-08-27 04:30:54 PP7 TP3 EP3] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp3_pp7_dp0.json
+[2026-08-27 04:30:54 PP6 TP2 EP2] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp2_pp6_dp0.json
+[2026-08-27 04:30:54 PP6 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:30:54 PP6 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:30:54 PP6 TP0 EP0] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp0_pp6_dp0.json
+[2026-08-27 04:30:54 PP7 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:30:54 PP7 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:30:54 PP7 TP0 EP0] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp0_pp7_dp0.json
+[2026-08-27 04:30:54 PP7 TP1 EP1] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp1_pp7_dp0.json
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 11.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 13.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.76profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.62profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.66profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.72profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.68profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.55profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.12profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  6.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  7.16profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.92profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.15profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.59profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.56profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 19.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  8.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.67profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.40profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.88profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  8.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  6.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.83profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.50profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.32profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.92profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.77profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.94profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.57profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 12.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 12.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 13.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.36profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.66profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  6.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.72profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  6.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.22profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.65profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.13profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.03profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.46profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.59profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.16profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  6.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.96profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.76profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.67profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  8.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.27profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.88profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.01profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.57profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.12profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 11.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  2.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.68profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.90profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.43profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.92profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  8.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.66profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  6.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  7.13profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.32profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.71profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:05,  1.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.16profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.31profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.77profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.28profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.48profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.87profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.40profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 13.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.35profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.98profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.88profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.73profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  3.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.88profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.09profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.20profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.97profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.83profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.47profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.40profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.60profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  8.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  7.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.97profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.05profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.05profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  6.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.49profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.72profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.19profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 10.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.91profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.64profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.58profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  3.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.35profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  9.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 12.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.09profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:02,  2.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 12.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.15profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.76profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.49profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.03profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.79profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.96profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.47profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.65profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.55profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.03profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  9.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.63profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.98profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:05,  1.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:01,  3.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.50profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  6.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.94profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.13profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.78profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.53profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.35profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.56profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.36profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:01<00:06,  1.06s/profile]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  8.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.91profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.66profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  8.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  7.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:02,  2.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.83profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.01profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.66profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.13profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.59profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  8.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  8.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.71profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.29profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.97profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.88profile/s]
+[2026-08-27 04:31:27 PP6 TP3 EP3] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.86profile/s]
+[2026-08-27 04:31:27 PP6 TP1 EP1] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.24profile/s]
+[2026-08-27 04:31:27 PP6 TP2 EP2] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.33profile/s]
+[2026-08-27 04:31:27 PP6 TP0 EP0] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.56profile/s][2026-08-27 04:31:27 PP6 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:27 PP6 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:27 PP6 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:27 PP6 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:27 PP6 TP0 EP0] max_total_num_tokens=3801408, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=54, context_len=1048576, available_gpu_mem=16.52 GB
+[2026-08-27 04:31:27 PP6 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:27 PP6 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:27 PP6 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:27 PP6 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
+[2026-08-27 04:31:28 PP7 TP2 EP2] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.32profile/s]
+[2026-08-27 04:31:28 PP7 TP3 EP3] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.60profile/s]
+[2026-08-27 04:31:28 PP7 TP0 EP0] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.60profile/s]
+[2026-08-27 04:31:28 PP7 TP1 EP1] FlashInfer autotune completed.
+[2026-08-27 04:31:28 PP7 TP2 EP2] multimem all-gather disabled (no multicast for world_size=4)
+[2026-08-27 04:31:28 PP7 TP3 EP3] multimem all-gather disabled (no multicast for world_size=4)
+[2026-08-27 04:31:28 PP7 TP1 EP1] multimem all-gather disabled (no multicast for world_size=4)
+[2026-08-27 04:31:28 PP7 TP0 EP0] multimem all-gather disabled (no multicast for world_size=4)
+[2026-08-27 04:31:28 PP7 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:28 PP7 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:28 PP7 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:28 PP7 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:28 PP7 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:28 PP7 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:28 PP7 TP0 EP0] max_total_num_tokens=3801408, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=54, context_len=1048576, available_gpu_mem=12.14 GB
+[2026-08-27 04:31:28 PP7 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:28 PP7 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+W0827 04:31:31.959518  2596 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5efc8ec0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.959637  2598 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ee654c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.959640  2597 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5eda3c80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.959542  2595 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5efc6f80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.959741  2599 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5eeaf500, len: 55296: Operation not permitted [1]
+W0827 04:31:31.959794  2600 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ef4e380, len: 55296: Operation not permitted [1]
+W0827 04:31:31.959827  2601 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5f548500, len: 6912: Operation not permitted [1]
+W0827 04:31:31.959918  2602 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5eec3bc0, len: 13824: Operation not permitted [1]
+W0827 04:31:31.960085  2603 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5f01bcc0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.960222  2604 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5f4c8ac0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.980162  2623 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60ba35c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.980152  2624 memory_location.cpp:73] Failed to get NUMA node, addr: 0x609ccf40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.980242  2625 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60d38600, len: 6912: Operation not permitted [1]
+W0827 04:31:31.980361  2626 memory_location.cpp:73] Failed to get NUMA node, addr: 0x609546c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.980516  2628 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60c49a00, len: 55296: Operation not permitted [1]
+W0827 04:31:31.980537  2627 memory_location.cpp:73] Failed to get NUMA node, addr: 0x607c6800, len: 55296: Operation not permitted [1]
+W0827 04:31:31.980547  2629 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60d11800, len: 6912: Operation not permitted [1]
+W0827 04:31:31.980834  2630 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60a1b740, len: 13824: Operation not permitted [1]
+W0827 04:31:31.981053  2632 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60ce5000, len: 6912: Operation not permitted [1]
+W0827 04:31:31.981062  2631 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60aef300, len: 6912: Operation not permitted [1]
+W0827 04:31:32.117022  2651 memory_location.cpp:73] Failed to get NUMA node, addr: 0x74c79d40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.117132  2652 memory_location.cpp:73] Failed to get NUMA node, addr: 0x74f99700, len: 6912: Operation not permitted [1]
+W0827 04:31:32.117317  2653 memory_location.cpp:73] Failed to get NUMA node, addr: 0x7546c740, len: 6912: Operation not permitted [1]
+W0827 04:31:32.117331  2654 memory_location.cpp:73] Failed to get NUMA node, addr: 0x74f4f140, len: 6912: Operation not permitted [1]
+W0827 04:31:32.117414  2656 memory_location.cpp:73] Failed to get NUMA node, addr: 0x750037c0, len: 55296: Operation not permitted [1]
+W0827 04:31:32.117445  2655 memory_location.cpp:73] Failed to get NUMA node, addr: 0x74bf1200, len: 55296: Operation not permitted [1]
+W0827 04:31:32.117705  2657 memory_location.cpp:73] Failed to get NUMA node, addr: 0x74f80080, len: 6912: Operation not permitted [1]
+W0827 04:31:32.117738  2658 memory_location.cpp:73] Failed to get NUMA node, addr: 0x74fbd2c0, len: 13824: Operation not permitted [1]
+W0827 04:31:32.117790  2659 memory_location.cpp:73] Failed to get NUMA node, addr: 0x74fa5e40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.118012  2660 memory_location.cpp:73] Failed to get NUMA node, addr: 0x74c24880, len: 6912: Operation not permitted [1]
+W0827 04:31:32.137427  2677 memory_location.cpp:73] Failed to get NUMA node, addr: 0x685bea40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.137485  2678 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6865c240, len: 6912: Operation not permitted [1]
+W0827 04:31:32.137612  2680 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6832f280, len: 6912: Operation not permitted [1]
+W0827 04:31:32.137683  2679 memory_location.cpp:73] Failed to get NUMA node, addr: 0x68511ec0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.137791  2681 memory_location.cpp:73] Failed to get NUMA node, addr: 0x687ae280, len: 55296: Operation not permitted [1]
+W0827 04:31:32.137800  2682 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6863aec0, len: 55296: Operation not permitted [1]
+W0827 04:31:32.137928  2683 memory_location.cpp:73] Failed to get NUMA node, addr: 0x685f4540, len: 6912: Operation not permitted [1]
+W0827 04:31:32.138145  2685 memory_location.cpp:73] Failed to get NUMA node, addr: 0x68b70f80, len: 6912: Operation not permitted [1]
+W0827 04:31:32.138144  2684 memory_location.cpp:73] Failed to get NUMA node, addr: 0x686e87c0, len: 13824: Operation not permitted [1]
+W0827 04:31:32.138219  2686 memory_location.cpp:73] Failed to get NUMA node, addr: 0x685491c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.187127  2706 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8eaeff40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.187187  2707 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8eaa1740, len: 6912: Operation not permitted [1]
+W0827 04:31:32.187136  2705 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8e5f2600, len: 6912: Operation not permitted [1]
+W0827 04:31:32.187413  2709 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ec4eb00, len: 55296: Operation not permitted [1]
+W0827 04:31:32.187405  2708 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ea3b800, len: 6912: Operation not permitted [1]
+W0827 04:31:32.187651  2710 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ed92ec0, len: 55296: Operation not permitted [1]
+W0827 04:31:32.187799  2711 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8eb3fbc0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.188133  2712 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ed0e1c0, len: 13824: Operation not permitted [1]
+W0827 04:31:32.188231  2713 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8e9d20c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.188272  2714 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8f2c7a80, len: 6912: Operation not permitted [1]
+W0827 04:31:32.252096  2732 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72b6b180, len: 6912: Operation not permitted [1]
+W0827 04:31:32.252050  2731 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72be0600, len: 6912: Operation not permitted [1]
+W0827 04:31:32.252147  2733 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72716c00, len: 6912: Operation not permitted [1]
+W0827 04:31:32.252276  2734 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72739940, len: 6912: Operation not permitted [1]
+W0827 04:31:32.252379  2735 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72da8a80, len: 55296: Operation not permitted [1]
+W0827 04:31:32.252439  2736 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72d15800, len: 55296: Operation not permitted [1]
+W0827 04:31:32.252678  2737 memory_location.cpp:73] Failed to get NUMA node, addr: 0x7315db40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.252701  2738 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72e0e700, len: 13824: Operation not permitted [1]
+W0827 04:31:32.252745  2739 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72f5f480, len: 6912: Operation not permitted [1]
+W0827 04:31:32.252832  2740 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72b295c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.268508  2760 memory_location.cpp:73] Failed to get NUMA node, addr: 0x69eefc00, len: 6912: Operation not permitted [1]
+W0827 04:31:32.268532  2759 memory_location.cpp:73] Failed to get NUMA node, addr: 0x69ec8680, len: 6912: Operation not permitted [1]
+W0827 04:31:32.268577  2761 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6a02e280, len: 6912: Operation not permitted [1]
+W0827 04:31:32.268708  2762 memory_location.cpp:73] Failed to get NUMA node, addr: 0x69d75dc0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.268766  2763 memory_location.cpp:73] Failed to get NUMA node, addr: 0x69fc4880, len: 55296: Operation not permitted [1]
+W0827 04:31:32.268886  2764 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6a0c8540, len: 55296: Operation not permitted [1]
+W0827 04:31:32.268975  2765 memory_location.cpp:73] Failed to get NUMA node, addr: 0x69c75940, len: 6912: Operation not permitted [1]
+W0827 04:31:32.269125  2766 memory_location.cpp:73] Failed to get NUMA node, addr: 0x698ae1c0, len: 13824: Operation not permitted [1]
+W0827 04:31:32.269232  2767 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6a5457c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.269354  2768 memory_location.cpp:73] Failed to get NUMA node, addr: 0x69c8fe80, len: 6912: Operation not permitted [1]
+W0827 04:31:32.281724  2785 memory_location.cpp:73] Failed to get NUMA node, addr: 0x678e47c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.281751  2786 memory_location.cpp:73] Failed to get NUMA node, addr: 0x67a39b80, len: 6912: Operation not permitted [1]
+W0827 04:31:32.281852  2787 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6798d9c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.281994  2788 memory_location.cpp:73] Failed to get NUMA node, addr: 0x67981dc0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.282085  2789 memory_location.cpp:73] Failed to get NUMA node, addr: 0x67823300, len: 55296: Operation not permitted [1]
+W0827 04:31:32.282146  2790 memory_location.cpp:73] Failed to get NUMA node, addr: 0x67a3bb00, len: 55296: Operation not permitted [1]
+W0827 04:31:32.282214  2791 memory_location.cpp:73] Failed to get NUMA node, addr: 0x67cdad40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.282285  2792 memory_location.cpp:73] Failed to get NUMA node, addr: 0x68028e40, len: 13824: Operation not permitted [1]
+W0827 04:31:32.282544  2794 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6796cf40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.282545  2793 memory_location.cpp:73] Failed to get NUMA node, addr: 0x67fa6600, len: 6912: Operation not permitted [1]
+[2026-08-27 04:31:32] Dummy health check server started in background thread at 0.0.0.0:30000
+[rank27]:[W827 04:34:16.127713860 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank25]:[W827 04:34:16.128748517 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank24]:[W827 04:34:16.129831915 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank26]:[W827 04:34:16.136239904 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank26]:[W827 04:35:04.964905528 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank25]:[W827 04:35:04.966225881 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank27]:[W827 04:35:04.966239242 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank24]:[W827 04:35:04.966301664 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank31]:[W827 04:35:04.990701389 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank28]:[W827 04:35:04.991914270 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank29]:[W827 04:35:04.994546241 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank30]:[W827 04:35:04.999020079 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank31]:[W827 04:35:10.873647776 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank30]:[W827 04:35:10.873901815 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank28]:[W827 04:35:10.874227564 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank29]:[W827 04:35:10.885318244 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[2026-08-27 04:35:10 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.29
+[2026-08-27 04:35:10 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.29
+[2026-08-27 04:35:29 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.39
+[2026-08-27 04:35:29 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.39
+[2026-08-27 04:35:31 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25.93
+[2026-08-27 04:35:31 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25.95
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/d_healthy_router.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/d_healthy_router.log
new file mode 100644
index 0000000..1190be5
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/d_healthy_router.log
@@ -0,0 +1 @@
+Error response from daemon: No such container: kimi3_pro6000_pd_pp8_standard_router
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/d_starting_d_node0.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/d_starting_d_node0.log
new file mode 100644
index 0000000..ae78e78
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/d_starting_d_node0.log
@@ -0,0 +1,487 @@
+/usr/local/lib/python3.12/dist-packages/torchao/quantization/quant_api.py:1731: SyntaxWarning: invalid escape sequence '\.'
+  """Configuration class for applying different quantization configs to modules or parameters based on their fully qualified names (FQNs).
+/sgl-workspace/sglang/python/sglang/launch_server.py:56: UserWarning: 'python -m sglang.launch_server' is still supported, but 'sglang serve' is the recommended entrypoint.
+  Example: sglang serve --model-path <model> [options]
+  warnings.warn(
+[2026-08-27 04:36:08] KV cache is forced as chunk cache for decode server
+[2026-08-27 04:36:11] Breakable CUDA graph is incompatible with MLA attention; disabling prefill CUDA graph.
+[2026-08-27 04:36:12] Attention backend not specified. Use triton backend by default.
+[2026-08-27 04:36:14] server_args=ServerArgs(model_path='/data/hf_models/Kimi-K3', tokenizer_path='/data/hf_models/Kimi-K3', tokenizer_mode='auto', tokenizer_backend='huggingface', tokenizer_worker_num=1, detokenizer_worker_num=1, skip_tokenizer_init=False, load_format='auto', model_loader_extra_config='{}', trust_remote_code=True, context_length=None, is_embedding=False, enable_multimodal=None, revision=None, model_impl='auto', model_config_parser='auto', json_model_override_args='{}', dtype='auto', quantization=None, quantization_param_path=None, kv_cache_dtype='auto', enable_fp32_lm_head=False, modelopt_quant=None, modelopt_checkpoint_restore_path=None, modelopt_checkpoint_save_path=None, modelopt_export_path=None, quantize_and_serve=False, rl_quant_profile=None, enable_tf32_matmul=False, mem_fraction_static=0.88, max_running_requests=None, max_queued_requests=None, max_total_tokens=None, chunked_prefill_size=8192, enable_dynamic_chunking=False, max_prefill_tokens=16384, prefill_max_requests=None, schedule_policy='fcfs', enable_priority_scheduling=False, disable_priority_preemption=False, default_priority_value=None, abort_on_priority_when_disabled=False, schedule_low_priority_values_first=False, priority_scheduling_preemption_threshold=10, retraction_policy='length', schedule_conservativeness=1.0, page_size=64, swa_full_tokens_ratio=0.8, disable_hybrid_swa_memory=False, radix_eviction_policy='lru', prefill_only_disable_kv_cache=False, disable_radix_cache=True, enable_page_major_kv_layout=False, enable_unified_memory=False, disable_chunked_prefix_cache=False, disable_overlap_schedule=False, num_continuous_decode_steps=1, scheduler_recv_interval=1, enable_mixed_chunk=False, nccl_port=None, dist_timeout=3600, dist_init_addr='174.1.60.5:20000', nnodes=4, node_rank=0, tp_size=32, dcp_size=1, pp_size=1, pp_max_micro_batch_size=None, pp_async_batch_depth=0, dp_size=1, load_balance_method='round_robin', attn_cp_size=1, moe_dp_size=1, dcp_comm_backend='ag_rs', dcp_replicate_q_proj=None, dwdp_size=1, enable_prefill_cp=False, cp_strategy=None, enable_dsa_cache_layer_split=False, enable_dsa_prefill_context_parallel=False, dsa_prefill_cp_mode='round-robin-split', enable_prefill_context_parallel=False, prefill_cp_mode='in-seq-split', enable_cp_decode_attn_tp=False, enable_dp_attention=False, enable_dp_attention_local_control_broadcast=False, enable_dp_lm_head=False, enable_attn_tp_input_scattered=False, disable_attn_tp_gather=False, enable_p2p_check=False, device='cuda', base_gpu_id=0, gpu_id_step=1, random_seed=369235288, watchdog_timeout=300, soft_watchdog_timeout=None, sleep_on_idle=False, use_ray=False, custom_sigquit_handler=None, numa_node=None, gc_threshold=None, host='0.0.0.0', port=30000, fastapi_root_path='', smg_grpc_mode=False, grpc_mode=False, grpc_port=None, sidecar=None, sidecar_args=None, skip_server_warmup=False, warmups=None, enable_http2=False, ssl_keyfile=None, ssl_certfile=None, ssl_ca_certs=None, ssl_keyfile_password=None, enable_ssl_refresh=False, api_key=None, admin_api_key=None, served_model_name='kimi-k3', weight_version='default', chat_template=None, hf_chat_template_name=None, completion_template=None, file_storage_path='sglang_storage', enable_cache_report=False, reasoning_parser=None, default_chat_template_kwargs=None, strip_thinking_cache=False, enable_strict_thinking=False, tool_call_parser=None, tool_server=None, sampling_defaults='model', asr_max_buffer_seconds=60, asr_max_concurrent_sessions=32, preferred_sampling_params=None, allow_auto_truncate=False, stream_interval=1, batch_notify_size=16, stream_response_default_include_usage=False, incremental_streaming_output=False, enable_streaming_session=False, enable_session_radix_cache=False, log_level='info', log_level_http=None, log_requests=False, log_requests_level=2, log_requests_format='text', log_requests_target=None, uvicorn_access_log_exclude_prefixes=[], crash_dump_folder=None, show_time_cost=False, enable_metrics=False, smg_http_sidecar_port=None, enable_mfu_metrics=False, enable_metrics_for_all_schedulers=False, load_snapshot_publish_interval=15, tokenizer_metrics_custom_labels_header='x-custom-labels', tokenizer_metrics_allowed_custom_labels=None, extra_metric_labels=None, bucket_time_to_first_token=None, bucket_inter_token_latency=None, bucket_e2e_request_latency=None, prompt_tokens_buckets=None, generation_tokens_buckets=None, gc_warning_threshold_secs=0.0, decode_log_interval=40, enable_request_time_stats_logging=False, kv_events_config=None, enable_forward_pass_metrics=False, forward_pass_metrics_worker_id='', forward_pass_metrics_ipc_name=None, enable_trace=False, trace_modules='request', otlp_traces_endpoint='localhost:4317', export_metrics_to_file=False, export_metrics_to_file_dir=None, stat_loggers=None, constrained_json_whitespace_pattern=None, constrained_json_disable_any_whitespace=False, attention_backend='triton', decode_attention_backend=None, prefill_attention_backend=None, sampling_backend='flashinfer', grammar_backend='xgrammar', radix_cache_backend=None, mm_attention_backend=None, fp8_gemm_runner_backend='auto', fp4_gemm_runner_backend='auto', bf16_gemm_backend='auto', dsa_prefill_backend=None, dsa_decode_backend=None, dsa_paged_mqa_logits_backend='auto', dsa_topk_backend='sgl-kernel', disable_flashinfer_autotune=False, flashinfer_autotune_skip_ops=None, mamba_backend='triton', cuda_graph_config=CudaGraphConfig(decode=PhaseConfig(backend='full', max_bs=16, bs=[1, 2, 4, 8, 12, 16], tc_compiler='eager', full_prefill_max_req=None), prefill=PhaseConfig(backend='disabled', max_bs=2048, bs=[4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], tc_compiler='eager', full_prefill_max_req=None)), cuda_graph_backend_decode=None, cuda_graph_backend_prefill=None, cuda_graph_max_bs_decode=16, cuda_graph_max_bs_prefill=None, cuda_graph_bs_decode=None, cuda_graph_bs_prefill=None, cuda_graph_tc_compiler=None, disable_prefill_cuda_graph=False, disable_decode_cuda_graph=False, disable_cuda_graph=False, disable_cuda_graph_padding=False, enable_profile_cuda_graph=False, enable_cudagraph_gc=False, debug_cuda_graph=False, enable_layerwise_nvtx_marker=False, enable_nccl_nvls=False, enable_symm_mem=False, triton_attention_reduce_in_fp32=False, triton_attention_num_kv_splits=8, triton_attention_split_tile_size=None, flashinfer_mla_disable_ragged=False, enable_fused_qk_norm_rope=False, enable_precise_embedding_interpolation=False, enable_fused_moe_sum_all_reduce=False, enable_deepseek_v4_fp4_indexer=False, disable_custom_all_reduce=False, enable_mscclpp=False, enable_torch_symm_mem=False, enable_scattered_sconv=False, pre_warm_nccl=False, enable_quant_communications=False, enable_flashinfer_allreduce_fusion=False, enforce_disable_flashinfer_allreduce_fusion=False, flashinfer_allreduce_fusion_backend=None, enable_aiter_allreduce_fusion=False, enable_torch_compile=False, enable_torch_compile_debug_mode=False, torch_compile_max_bs=32, torchao_config='', speculative_algorithm=None, speculative_draft_model_path=None, speculative_draft_model_revision=None, speculative_draft_load_format=None, speculative_num_steps=None, speculative_eagle_topk=None, speculative_num_draft_tokens=None, speculative_dflash_block_size=None, speculative_dspark_block_size=None, speculative_dspark_sps_table_path=None, speculative_dspark_confidence_sts_path=None, speculative_dspark_align_verify_tokens_to_graph_tier=False, speculative_accept_threshold_single=1.0, speculative_accept_threshold_acc=1.0, speculative_use_rejection_sampling=False, speculative_token_map=None, speculative_attention_mode='prefill', speculative_draft_attention_backend=None, speculative_draft_window_size=None, speculative_moe_runner_backend='marlin', speculative_moe_a2a_backend=None, speculative_draft_model_quantization=None, speculative_skip_dp_mlp_sync=False, enable_multi_layer_eagle=False, speculative_adaptive=False, speculative_adaptive_config=None, decoupled_spec_bind_endpoint=None, decoupled_spec_connect_endpoints=None, decoupled_spec_rank=None, decoupled_spec_role='null', spec_trace_dir=None, speculative_ngram_min_bfs_breadth=1, speculative_ngram_max_bfs_breadth=10, speculative_ngram_match_type='BFS', speculative_ngram_max_trie_depth=18, speculative_ngram_capacity=10000000, speculative_ngram_external_corpus_path=None, speculative_ngram_external_sam_budget=0, speculative_ngram_external_corpus_max_tokens=10000000, ep_size=32, moe_a2a_backend='none', moe_runner_backend='marlin', flashinfer_mxfp4_moe_precision='default', deepep_mode='auto', fuseep_mode=2, deepep_dispatcher_output_dtype='auto', ep_num_redundant_experts=0, ep_dispatch_algorithm=None, init_expert_location='trivial', enable_eplb=False, eplb_algorithm='auto', eplb_rebalance_num_iterations=1000, eplb_rebalance_layers_per_chunk=None, eplb_min_rebalancing_utilization_threshold=1.0, expert_distribution_recorder_mode=None, expert_distribution_recorder_buffer_size=1000, enable_expert_distribution_metrics=False, deepep_config=None, moe_dense_tp_size=None, elastic_ep_backend=None, enable_elastic_expert_backup=False, mooncake_ib_device=None, enable_waterfill=False, ep_join_mode=None, ep_join_rank_offset=0, elastic_ep_initial_size=None, max_ep_size=None, elastic_ep_scale_timeout=600, elastic_ep_rejoin=False, disable_flashinfer_cutlass_moe_fp4_allgather=False, disable_shared_experts_fusion=False, enforce_shared_experts_fusion=False, max_mamba_cache_size=None, mamba_ssm_dtype=None, mamba_max_states_per_path=-1, enable_mamba_cache_stochastic_rounding=False, mamba_cache_philox_rounds=0, mamba_full_memory_ratio=0.21, mamba_radix_cache_strategy='extra_buffer_lazy', uses_mamba_radix_cache=False, mamba_track_interval=256, enable_int8_mamba_checkpoint=False, int8_mamba_ckpt_size=None, linear_attn_backend='triton', linear_attn_decode_backend=None, linear_attn_prefill_backend=None, linear_attn_verify_backend=None, enable_linear_replayssm=False, linear_replayssm_cache_len=16, enable_linear_replayssm_spec=False, enable_hierarchical_cache=False, hicache_ratio=2.0, hicache_size=0, hicache_write_policy='write_through', hicache_io_backend='kernel', hicache_mem_layout='page_first', hicache_storage_backend=None, hicache_storage_prefetch_policy='timeout', hicache_storage_backend_extra_config=None, enable_hisparse=False, hisparse_config=None, enable_broadcast_mm_inputs_process=False, enable_prefix_mm_cache=False, mm_enable_dp_encoder=False, mm_process_config={}, mm_processor_worker_num=0, mm_io_worker_num=0, limit_mm_data_per_request=None, enable_mm_global_cache=False, disable_fast_image_processor=False, mm_feature_transport='cpu', keep_mm_feature_on_device=False, enable_lora=None, enable_lora_overlap_loading=None, max_lora_rank=None, lora_target_modules=None, lora_paths=None, max_loaded_loras=None, max_loras_per_batch=8, lora_eviction_policy='lru', lora_backend='csgmv', max_lora_chunk_size=16, experts_shared_outer_loras=None, lora_use_virtual_experts=False, lora_strict_loading=False, lora_drain_wait_threshold=0.0, enable_two_batch_overlap=False, enable_single_batch_overlap=False, tbo_token_distribution_threshold=0.48, cpu_offload_gb=0, offload_group_size=-1, offload_num_in_group=1, offload_prefetch_step=1, offload_mode='cpu', enable_lmcache=False, lmcache_config_file=None, enable_flexkv=False, flexkv_config_file=None, kt_weight_path=None, kt_method='AMXINT4', kt_cpuinfer=None, kt_threadpool_count=2, kt_num_gpu_experts=None, kt_max_deferred_experts_per_token=None, dllm_algorithm=None, dllm_algorithm_config=None, dllm_fdfo=True, disaggregation_mode='decode', disaggregation_transfer_backend='mooncake', disaggregation_bootstrap_port=28800, disaggregation_ib_device='mlx5_0,mlx5_1,mlx5_2,mlx5_3', disaggregation_decode_enable_radix_cache=False, disaggregation_decode_enable_offload_kvcache=False, num_reserved_decode_tokens=512, disaggregation_decode_extra_slots=0, disaggregation_decode_polling_interval=1, optimistic_prefill_attempts=0, encoder_only=False, language_only=False, encoder_transfer_backend='zmq_to_tokenizer', encoder_urls=[], encoder_bootstrap_port=8997, encoder_register_urls=[], enable_adaptive_dispatch_to_encoder=False, enable_pdmux=False, pdmux_config_path=None, sm_group_num=8, custom_weight_loader=[], weight_loader_disable_mmap=False, weight_loader_prefetch_checkpoints=False, weight_loader_prefetch_num_threads=4, weight_loader_drop_cache_after_load=False, remote_instance_weight_loader_seed_instance_ip=None, remote_instance_weight_loader_seed_instance_service_port=None, remote_instance_weight_loader_send_weights_group_ports=None, remote_instance_weight_loader_backend='nccl', remote_instance_weight_loader_start_seed_via_transfer_engine=False, engine_info_bootstrap_port=6789, modelexpress_config=None, download_dir=None, model_checksum=None, delete_ckpt_after_loading=False, decrypted_config_file=None, decrypted_draft_config_file=None, checkpoint_engine_wait_weights_before_ready=False, enable_prefill_delayer=False, prefill_delayer_max_delay_passes=30, prefill_delayer_token_usage_low_watermark=None, prefill_delayer_forward_passes_buckets=None, prefill_delayer_wait_seconds_buckets=None, prefill_delayer_queue_min_ratio=None, prefill_delayer_max_delay_ms=None, min_free_slots_delay=None, enable_deterministic_inference=False, rl_on_policy_target=None, kv_canary='none', kv_canary_real_data='none', kv_canary_sweep_interval=0, enable_dynamic_batch_tokenizer=False, dynamic_batch_tokenizer_batch_size=32, dynamic_batch_tokenizer_batch_timeout=0.002, enable_tokenizer_batch_encode=False, disable_tokenizer_batch_decode=False, debug_tensor_dump_output_folder=None, debug_tensor_dump_layers=None, debug_tensor_dump_input_file=None, enable_memory_saver=False, enable_weights_cpu_backup=False, enable_draft_weights_cpu_backup=False, enable_custom_logit_processor=False, enable_return_hidden_states=False, enable_return_routed_experts=False, enable_return_indexer_topk=False, disable_outlines_disk_cache=False, enable_mis=False, weight_cache_mode='off', weight_cache_socket=None, weight_cache_timeout=1800, forward_hooks=None, msprobe_dump_config=None)
+[2026-08-27 04:36:14] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:14] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:14] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:14] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:14] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:14] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:14] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:14] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:17] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:17] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:18] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:19] Multimodal processor concurrency enabled with 2 isolated worker threads (auto).
+[2026-08-27 04:36:19] No HuggingFace chat template found
+[2026-08-27 04:36:19] No chat template found, defaulting to 'string' content format
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:34 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:34 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:34 TP4 EP4] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:34 TP4 EP4] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:34 TP6 EP6] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:34 TP6 EP6] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:34] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:34] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:34 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:34 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+/sgl-workspace/sglang/python/sglang/srt/utils/hf_transformers/tokenizer.py:423: UserWarning: Using a slow tokenizer. This might cause a significant slowdown. Consider using a fast tokenizer instead.
+  warnings.warn(
+[2026-08-27 04:36:34] Applying special tokens cache patch for Kimi tokenizer: <class 'transformers_modules.Kimi_hyphen_K3.caf40b2773f9029e.tokenization_kimi.TikTokenTokenizer'>
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:34 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:34 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:34 TP7 EP7] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:34 TP7 EP7] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:34 TP5 EP5] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:34 TP5 EP5] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:35 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:35 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:36 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:36 TP6 EP6] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:36 TP4 EP4] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:36 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:36 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:36 TP7 EP7] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:36 TP5 EP5] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:37 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:37.190651   236 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:37.190778   236 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:37.190796   236 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.5 port: 12001
+I0827 04:36:37.190856   236 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.5:16500
+I0827 04:36:37.191038   236 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:37.194572   241 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:37.194692   241 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:37.194712   241 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.5 port: 12001
+I0827 04:36:37.194767   241 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.5:15678
+I0827 04:36:37.194965   241 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:37.219420   236 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:37.223201   241 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:37.227941   236 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:37.231628   241 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:37.236233   236 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:37.239850   241 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:37.244647   236 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:37.248145   241 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:37.248418   236 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:37.248464   236 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:37.248525   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.250921   241 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:37.250947   241 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:37.251004   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.255973   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:37.257668   236 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:05
+I0827 04:36:37.257705   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.283210   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:37.283238   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:37.284873   236 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:05
+I0827 04:36:37.284902   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.284904   241 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:05
+I0827 04:36:37.284934   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.293713   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:37.295406   241 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:05
+I0827 04:36:37.295433   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.311285   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:37.312896   236 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:05
+I0827 04:36:37.312925   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.327399   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:37.328944   241 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:05
+I0827 04:36:37.328972   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.339579   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:37.341301   236 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:05
+I0827 04:36:37.341434   236 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:37 TP1 EP1] Init torch distributed begin.
+I0827 04:36:37.359290   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:37.360814   241 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:05
+I0827 04:36:37.360936   241 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:37 TP6 EP6] Init torch distributed begin.
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:37.366210   239 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:37.366365   239 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:37.366390   239 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.5 port: 12001
+I0827 04:36:37.366467   239 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.5:15787
+I0827 04:36:37.366704   239 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:37.395485   239 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:37.402652   239 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:37.409775   239 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:37.434130   239 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:37.436257   239 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:37.436282   239 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:37.436331   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.442425   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:37.443729   239 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:05
+I0827 04:36:37.443749   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.448421   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:37.449750   239 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:05
+I0827 04:36:37.449769   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.454622   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:37.455952   239 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:05
+I0827 04:36:37.455972   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.463356   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:37.464902   239 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:05
+I0827 04:36:37.465021   239 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:37 TP4 EP4] Init torch distributed begin.
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:37.510618   238 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:37.510744   238 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:37.510763   238 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.5 port: 12001
+I0827 04:36:37.510824   238 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.5:16205
+I0827 04:36:37.511003   238 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:37.524268   242 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:37.524391   242 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:37.524411   242 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.5 port: 12001
+I0827 04:36:37.524524   242 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.5:16605
+I0827 04:36:37.524709   242 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:37.528693   235 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:37.528817   235 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:37.528836   235 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.5 port: 12001
+I0827 04:36:37.528893   235 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.5:16672
+I0827 04:36:37.529069   235 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:37.535192   238 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:37.535425   235 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:37.541376   235 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:37.542625   238 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:37.547586   235 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:37.549854   238 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:37.551366   242 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:37.553905   235 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:37.556406   235 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:37.556468   235 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:37.556523   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.558794   242 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:37.563313   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:37.566102   242 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:37.573271   242 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:37.574895   238 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:37.576385   242 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:37.576412   242 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:37.576478   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.577947   238 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:37.577976   238 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:37.578033   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.583841   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:37.585472   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:37.585623   242 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:05
+I0827 04:36:37.585656   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.587072   238 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:05
+I0827 04:36:37.587105   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.594069   235 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:05
+I0827 04:36:37.594128   235 rdma_context.cpp:205] Using SIEVE endpoint store
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:37.598918   240 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:37.599033   240 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:37.599050   240 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.5 port: 12001
+I0827 04:36:37.599107   240 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.5:15428
+I0827 04:36:37.599270   240 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:37.605341   240 topology.cpp:127] Device mlx5_0 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:37.606741   237 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:37.606841   237 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:37.606855   237 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.5 port: 12001
+I0827 04:36:37.606901   237 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.5:16531
+I0827 04:36:37.607028   237 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:37.611215   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:37.611989   237 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:37.612797   242 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:05
+I0827 04:36:37.612828   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.615110   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:37.616685   238 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:05
+I0827 04:36:37.616714   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.619258   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:37.620560   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:37.620851   235 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:05
+I0827 04:36:37.620885   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.622239   242 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:05
+I0827 04:36:37.622268   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.624459   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:37.625998   238 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:05
+I0827 04:36:37.626025   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.630085   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:37.630686   240 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:37.631680   242 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:05
+I0827 04:36:37.631806   242 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:37 TP7 EP7] Init torch distributed begin.
+I0827 04:36:37.633723   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:37.635268   238 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:05
+I0827 04:36:37.635390   238 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:37 TP3 EP3] Init torch distributed begin.
+I0827 04:36:37.642561   237 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:37.649834   237 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:37.651238   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:37.652786   235 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:05
+I0827 04:36:37.652813   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.657148   237 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:37.658787   240 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:37.659991   237 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:37.660017   237 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:37.660071   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.667523   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:37.669195   240 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:37.669226   237 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:05
+I0827 04:36:37.669257   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.672511   240 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:37.672547   240 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:37.672606   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.677067   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:37.678661   237 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:05
+I0827 04:36:37.678689   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.682178   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:37.683274   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:37.683740   240 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:05
+I0827 04:36:37.683773   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.684834   235 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:05
+I0827 04:36:37.684954   235 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:37 TP0 EP0] Init torch distributed begin.
+I0827 04:36:37.703369   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:37.704975   237 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:05
+I0827 04:36:37.705003   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.711591   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:37.713003   240 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:05
+I0827 04:36:37.713030   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.735574   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:37.737313   237 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:05
+I0827 04:36:37.737439   237 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:37 TP2 EP2] Init torch distributed begin.
+I0827 04:36:37.740485   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:37.741588   240 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:05
+I0827 04:36:37.741605   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.766494   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:37.767982   240 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:05
+I0827 04:36:37.768137   240 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:37 TP5 EP5] Init torch distributed begin.
+[2026-08-27 04:36:46 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:36:48 TP0 EP0] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP1 EP1] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP2 EP2] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP5 EP5] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP4 EP4] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP3 EP3] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP6 EP6] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP7 EP7] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP0 EP0] Init torch distributed ends. elapsed=10.55 s, mem usage=0.40 GB
+[2026-08-27 04:36:48 TP7 EP7] Init torch distributed ends. elapsed=10.61 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP6 EP6] Init torch distributed ends. elapsed=10.88 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP4 EP4] Init torch distributed ends. elapsed=10.77 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP3 EP3] Init torch distributed ends. elapsed=10.60 s, mem usage=0.40 GB
+[2026-08-27 04:36:48 TP1 EP1] Init torch distributed ends. elapsed=10.90 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP5 EP5] Init torch distributed ends. elapsed=10.47 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP2 EP2] Init torch distributed ends. elapsed=10.50 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP7 EP7] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP4 EP4] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP1 EP1] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP6 EP6] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP7 EP7] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP4 EP4] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP5 EP5] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP6 EP6] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP3 EP3] Load weight begin. avail mem=82.19 GB
+[2026-08-27 04:36:48 TP0 EP0] Load weight begin. avail mem=82.19 GB
+[2026-08-27 04:36:48 TP5 EP5] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP2 EP2] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP4 EP4] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP7 EP7] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP6 EP6] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP4 EP4] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP7 EP7] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP6 EP6] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP5 EP5] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP5 EP5] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:50 TP1 EP1] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP0 EP0] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP2 EP2] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP3 EP3] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP5 EP5] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP4 EP4] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP6 EP6] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP7 EP7] multimem all-gather disabled because the TP group spans across nodes.
+
Multi-thread loading shards:   0% Completed | 0/96 [00:00<?, ?it/s]
Multi-thread loading shards:   1% Completed | 1/96 [00:00<00:24,  3.85it/s]
Multi-thread loading shards:   2% Completed | 2/96 [00:07<07:13,  4.61s/it]
Multi-thread loading shards:   3% Completed | 3/96 [00:09<04:44,  3.06s/it]
Multi-thread loading shards:   4% Completed | 4/96 [00:10<03:33,  2.32s/it]
Multi-thread loading shards:   5% Completed | 5/96 [00:11<02:54,  1.92s/it]
Multi-thread loading shards:   6% Completed | 6/96 [00:12<02:29,  1.66s/it]
Multi-thread loading shards:   7% Completed | 7/96 [00:13<02:13,  1.50s/it]
Multi-thread loading shards:   8% Completed | 8/96 [00:15<02:04,  1.41s/it]
Multi-thread loading shards:   9% Completed | 9/96 [00:16<01:56,  1.34s/it]
Multi-thread loading shards:  10% Completed | 10/96 [00:17<01:47,  1.25s/it]
Multi-thread loading shards:  11% Completed | 11/96 [00:18<01:44,  1.23s/it]
Multi-thread loading shards:  12% Completed | 12/96 [00:19<01:43,  1.23s/it]
Multi-thread loading shards:  14% Completed | 13/96 [00:21<01:59,  1.44s/it]
Multi-thread loading shards:  15% Completed | 14/96 [00:22<01:51,  1.36s/it]
Multi-thread loading shards:  16% Completed | 15/96 [00:24<01:45,  1.30s/it]
Multi-thread loading shards:  17% Completed | 16/96 [00:25<01:42,  1.28s/it]
Multi-thread loading shards:  18% Completed | 17/96 [00:26<01:39,  1.26s/it]
Multi-thread loading shards:  19% Completed | 18/96 [00:27<01:36,  1.24s/it]
Multi-thread loading shards:  20% Completed | 19/96 [00:28<01:32,  1.20s/it]
Multi-thread loading shards:  21% Completed | 20/96 [00:29<01:30,  1.19s/it]
Multi-thread loading shards:  22% Completed | 21/96 [00:31<01:29,  1.19s/it]
Multi-thread loading shards:  23% Completed | 22/96 [00:32<01:27,  1.18s/it]
Multi-thread loading shards:  24% Completed | 23/96 [00:33<01:26,  1.18s/it]
Multi-thread loading shards:  25% Completed | 24/96 [00:34<01:23,  1.16s/it]
Multi-thread loading shards:  26% Completed | 25/96 [00:35<01:20,  1.14s/it]
Multi-thread loading shards:  27% Completed | 26/96 [00:36<01:18,  1.13s/it]
Multi-thread loading shards:  28% Completed | 27/96 [00:37<01:19,  1.15s/it]
Multi-thread loading shards:  29% Completed | 28/96 [00:39<01:18,  1.16s/it]
Multi-thread loading shards:  30% Completed | 29/96 [00:40<01:17,  1.16s/it]
Multi-thread loading shards:  31% Completed | 30/96 [00:41<01:17,  1.18s/it]
Multi-thread loading shards:  32% Completed | 31/96 [00:42<01:16,  1.18s/it]
Multi-thread loading shards:  33% Completed | 32/96 [00:44<01:30,  1.42s/it]
Multi-thread loading shards:  34% Completed | 33/96 [00:45<01:23,  1.33s/it]
Multi-thread loading shards:  35% Completed | 34/96 [00:46<01:17,  1.26s/it]
Multi-thread loading shards:  36% Completed | 35/96 [00:48<01:14,  1.23s/it]
Multi-thread loading shards:  38% Completed | 36/96 [00:49<01:13,  1.22s/it]
Multi-thread loading shards:  39% Completed | 37/96 [00:50<01:11,  1.21s/it]
Multi-thread loading shards:  40% Completed | 38/96 [00:51<01:09,  1.20s/it]
Multi-thread loading shards:  41% Completed | 39/96 [00:52<01:07,  1.18s/it]
Multi-thread loading shards:  42% Completed | 40/96 [00:53<01:05,  1.18s/it]
Multi-thread loading shards:  43% Completed | 41/96 [00:55<01:04,  1.17s/it]
Multi-thread loading shards:  44% Completed | 42/96 [00:56<01:02,  1.17s/it]
Multi-thread loading shards:  45% Completed | 43/96 [00:57<01:03,  1.19s/it]
Multi-thread loading shards:  46% Completed | 44/96 [00:58<01:01,  1.18s/it]
Multi-thread loading shards:  47% Completed | 45/96 [00:59<00:57,  1.13s/it]
Multi-thread loading shards:  48% Completed | 46/96 [01:00<00:57,  1.15s/it]
Multi-thread loading shards:  49% Completed | 47/96 [01:02<00:57,  1.17s/it]
Multi-thread loading shards:  50% Completed | 48/96 [01:03<00:55,  1.15s/it]
Multi-thread loading shards:  51% Completed | 49/96 [01:04<00:53,  1.13s/it]
Multi-thread loading shards:  52% Completed | 50/96 [01:06<01:01,  1.34s/it]
Multi-thread loading shards:  53% Completed | 51/96 [01:07<00:58,  1.29s/it]
Multi-thread loading shards:  54% Completed | 52/96 [01:08<00:55,  1.27s/it]
Multi-thread loading shards:  55% Completed | 53/96 [01:09<00:53,  1.24s/it]
Multi-thread loading shards:  56% Completed | 54/96 [01:10<00:50,  1.21s/it]
Multi-thread loading shards:  57% Completed | 55/96 [01:12<00:49,  1.22s/it]
Multi-thread loading shards:  58% Completed | 56/96 [01:13<00:48,  1.22s/it]
Multi-thread loading shards:  59% Completed | 57/96 [01:14<00:47,  1.21s/it]
Multi-thread loading shards:  60% Completed | 58/96 [01:15<00:45,  1.20s/it]
Multi-thread loading shards:  61% Completed | 59/96 [01:16<00:44,  1.19s/it]
Multi-thread loading shards:  62% Completed | 60/96 [01:17<00:43,  1.20s/it]
Multi-thread loading shards:  64% Completed | 61/96 [01:19<00:42,  1.21s/it]
Multi-thread loading shards:  65% Completed | 62/96 [01:20<00:40,  1.19s/it]
Multi-thread loading shards:  66% Completed | 63/96 [01:21<00:40,  1.21s/it]
Multi-thread loading shards:  67% Completed | 64/96 [01:22<00:38,  1.20s/it]
Multi-thread loading shards:  68% Completed | 65/96 [01:24<00:37,  1.20s/it]
Multi-thread loading shards:  69% Completed | 66/96 [01:25<00:36,  1.20s/it]
Multi-thread loading shards:  70% Completed | 67/96 [01:26<00:35,  1.22s/it]
Multi-thread loading shards:  71% Completed | 68/96 [01:28<00:41,  1.48s/it]
Multi-thread loading shards:  72% Completed | 69/96 [01:29<00:37,  1.39s/it]
Multi-thread loading shards:  73% Completed | 70/96 [01:30<00:34,  1.34s/it]
Multi-thread loading shards:  74% Completed | 71/96 [01:32<00:32,  1.29s/it]
Multi-thread loading shards:  75% Completed | 72/96 [01:33<00:31,  1.30s/it]
Multi-thread loading shards:  76% Completed | 73/96 [01:34<00:29,  1.29s/it]
Multi-thread loading shards:  77% Completed | 74/96 [01:35<00:27,  1.27s/it]
Multi-thread loading shards:  78% Completed | 75/96 [01:37<00:26,  1.26s/it]
Multi-thread loading shards:  79% Completed | 76/96 [01:38<00:24,  1.24s/it]
Multi-thread loading shards:  80% Completed | 77/96 [01:39<00:23,  1.23s/it]
Multi-thread loading shards:  81% Completed | 78/96 [01:40<00:21,  1.19s/it]
Multi-thread loading shards:  82% Completed | 79/96 [01:41<00:20,  1.20s/it]
Multi-thread loading shards:  83% Completed | 80/96 [01:43<00:19,  1.20s/it]
Multi-thread loading shards:  84% Completed | 81/96 [01:44<00:18,  1.21s/it]
Multi-thread loading shards:  85% Completed | 82/96 [01:45<00:16,  1.20s/it]
Multi-thread loading shards:  86% Completed | 83/96 [01:46<00:15,  1.17s/it]
Multi-thread loading shards:  88% Completed | 84/96 [01:47<00:14,  1.17s/it]
Multi-thread loading shards:  89% Completed | 85/96 [01:48<00:12,  1.10s/it]
Multi-thread loading shards:  90% Completed | 86/96 [01:49<00:10,  1.05s/it]
Multi-thread loading shards:  91% Completed | 87/96 [01:50<00:09,  1.02s/it]
Multi-thread loading shards:  92% Completed | 88/96 [01:51<00:07,  1.01it/s]
Multi-thread loading shards:  93% Completed | 89/96 [01:52<00:06,  1.03it/s]
Multi-thread loading shards:  94% Completed | 90/96 [01:53<00:05,  1.04it/s]
Multi-thread loading shards:  95% Completed | 91/96 [01:54<00:04,  1.06it/s]
Multi-thread loading shards:  96% Completed | 92/96 [01:55<00:03,  1.06it/s]
Multi-thread loading shards:  97% Completed | 93/96 [01:56<00:02,  1.08it/s]
Multi-thread loading shards:  98% Completed | 94/96 [01:56<00:01,  1.46it/s]
Multi-thread loading shards: 100% Completed | 96/96 [01:56<00:00,  2.28it/s]
Multi-thread loading shards: 100% Completed | 96/96 [01:56<00:00,  1.21s/it]
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:58 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP6 EP6] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP0 EP0] Load weight end. elapsed=130.28 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP6 EP6] Load weight end. elapsed=130.29 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP3 EP3] Load weight end. elapsed=130.28 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP2 EP2] Load weight end. elapsed=130.32 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP5 EP5] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP5 EP5] Load weight end. elapsed=130.38 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP4 EP4] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP4 EP4] Load weight end. elapsed=130.48 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP7 EP7] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP7 EP7] Load weight end. elapsed=130.48 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:39:01 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:39:01 TP1 EP1] Load weight end. elapsed=132.81 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:39:01 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP6 EP6] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP6 EP6] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP5 EP5] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP5 EP5] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP7 EP7] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP7 EP7] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP4 EP4] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP4 EP4] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:02 TP0 EP0] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP7 EP7] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP6 EP6] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP5 EP5] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP1 EP1] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP4 EP4] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP3 EP3] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP2 EP2] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP7 EP7] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP4 EP4] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP6 EP6] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP5 EP5] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP0 EP0] Memory pool end. avail mem=9.18 GB
+[2026-08-27 04:39:02 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP1 EP1] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP6 EP6] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP5 EP5] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP7 EP7] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP6 EP6] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP5 EP5] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP7 EP7] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP4 EP4] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP4 EP4] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP2 EP2] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP3 EP3] Memory pool end. avail mem=9.18 GB
+[2026-08-27 04:39:02 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:39:02 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:39:02 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP0 EP0] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.17 GB
+[2026-08-27 04:39:02 TP6 EP6] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP6 EP6] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP1 EP1] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP5 EP5] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP7 EP7] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP5 EP5] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP7 EP7] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:39:02 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP3 EP3] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.17 GB
+[2026-08-27 04:39:02 TP4 EP4] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP4 EP4] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:39:02 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP2 EP2] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/d_starting_d_node1.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/d_starting_d_node1.log
new file mode 100644
index 0000000..128ac56
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/d_starting_d_node1.log
@@ -0,0 +1,401 @@
+/usr/local/lib/python3.12/dist-packages/torchao/quantization/quant_api.py:1731: SyntaxWarning: invalid escape sequence '\.'
+  """Configuration class for applying different quantization configs to modules or parameters based on their fully qualified names (FQNs).
+/sgl-workspace/sglang/python/sglang/launch_server.py:56: UserWarning: 'python -m sglang.launch_server' is still supported, but 'sglang serve' is the recommended entrypoint.
+  Example: sglang serve --model-path <model> [options]
+  warnings.warn(
+[2026-08-27 04:35:58] KV cache is forced as chunk cache for decode server
+[2026-08-27 04:36:02] Breakable CUDA graph is incompatible with MLA attention; disabling prefill CUDA graph.
+[2026-08-27 04:36:02] Attention backend not specified. Use triton backend by default.
+[2026-08-27 04:36:05] server_args=ServerArgs(model_path='/data/hf_models/Kimi-K3', tokenizer_path='/data/hf_models/Kimi-K3', tokenizer_mode='auto', tokenizer_backend='huggingface', tokenizer_worker_num=1, detokenizer_worker_num=1, skip_tokenizer_init=False, load_format='auto', model_loader_extra_config='{}', trust_remote_code=True, context_length=None, is_embedding=False, enable_multimodal=None, revision=None, model_impl='auto', model_config_parser='auto', json_model_override_args='{}', dtype='auto', quantization=None, quantization_param_path=None, kv_cache_dtype='auto', enable_fp32_lm_head=False, modelopt_quant=None, modelopt_checkpoint_restore_path=None, modelopt_checkpoint_save_path=None, modelopt_export_path=None, quantize_and_serve=False, rl_quant_profile=None, enable_tf32_matmul=False, mem_fraction_static=0.88, max_running_requests=None, max_queued_requests=None, max_total_tokens=None, chunked_prefill_size=8192, enable_dynamic_chunking=False, max_prefill_tokens=16384, prefill_max_requests=None, schedule_policy='fcfs', enable_priority_scheduling=False, disable_priority_preemption=False, default_priority_value=None, abort_on_priority_when_disabled=False, schedule_low_priority_values_first=False, priority_scheduling_preemption_threshold=10, retraction_policy='length', schedule_conservativeness=1.0, page_size=64, swa_full_tokens_ratio=0.8, disable_hybrid_swa_memory=False, radix_eviction_policy='lru', prefill_only_disable_kv_cache=False, disable_radix_cache=True, enable_page_major_kv_layout=False, enable_unified_memory=False, disable_chunked_prefix_cache=False, disable_overlap_schedule=False, num_continuous_decode_steps=1, scheduler_recv_interval=1, enable_mixed_chunk=False, nccl_port=None, dist_timeout=3600, dist_init_addr='174.1.60.5:20000', nnodes=4, node_rank=1, tp_size=32, dcp_size=1, pp_size=1, pp_max_micro_batch_size=None, pp_async_batch_depth=0, dp_size=1, load_balance_method='round_robin', attn_cp_size=1, moe_dp_size=1, dcp_comm_backend='ag_rs', dcp_replicate_q_proj=None, dwdp_size=1, enable_prefill_cp=False, cp_strategy=None, enable_dsa_cache_layer_split=False, enable_dsa_prefill_context_parallel=False, dsa_prefill_cp_mode='round-robin-split', enable_prefill_context_parallel=False, prefill_cp_mode='in-seq-split', enable_cp_decode_attn_tp=False, enable_dp_attention=False, enable_dp_attention_local_control_broadcast=False, enable_dp_lm_head=False, enable_attn_tp_input_scattered=False, disable_attn_tp_gather=False, enable_p2p_check=False, device='cuda', base_gpu_id=0, gpu_id_step=1, random_seed=434112381, watchdog_timeout=300, soft_watchdog_timeout=None, sleep_on_idle=False, use_ray=False, custom_sigquit_handler=None, numa_node=None, gc_threshold=None, host='0.0.0.0', port=30000, fastapi_root_path='', smg_grpc_mode=False, grpc_mode=False, grpc_port=None, sidecar=None, sidecar_args=None, skip_server_warmup=False, warmups=None, enable_http2=False, ssl_keyfile=None, ssl_certfile=None, ssl_ca_certs=None, ssl_keyfile_password=None, enable_ssl_refresh=False, api_key=None, admin_api_key=None, served_model_name='kimi-k3', weight_version='default', chat_template=None, hf_chat_template_name=None, completion_template=None, file_storage_path='sglang_storage', enable_cache_report=False, reasoning_parser=None, default_chat_template_kwargs=None, strip_thinking_cache=False, enable_strict_thinking=False, tool_call_parser=None, tool_server=None, sampling_defaults='model', asr_max_buffer_seconds=60, asr_max_concurrent_sessions=32, preferred_sampling_params=None, allow_auto_truncate=False, stream_interval=1, batch_notify_size=16, stream_response_default_include_usage=False, incremental_streaming_output=False, enable_streaming_session=False, enable_session_radix_cache=False, log_level='info', log_level_http=None, log_requests=False, log_requests_level=2, log_requests_format='text', log_requests_target=None, uvicorn_access_log_exclude_prefixes=[], crash_dump_folder=None, show_time_cost=False, enable_metrics=False, smg_http_sidecar_port=None, enable_mfu_metrics=False, enable_metrics_for_all_schedulers=False, load_snapshot_publish_interval=15, tokenizer_metrics_custom_labels_header='x-custom-labels', tokenizer_metrics_allowed_custom_labels=None, extra_metric_labels=None, bucket_time_to_first_token=None, bucket_inter_token_latency=None, bucket_e2e_request_latency=None, prompt_tokens_buckets=None, generation_tokens_buckets=None, gc_warning_threshold_secs=0.0, decode_log_interval=40, enable_request_time_stats_logging=False, kv_events_config=None, enable_forward_pass_metrics=False, forward_pass_metrics_worker_id='', forward_pass_metrics_ipc_name=None, enable_trace=False, trace_modules='request', otlp_traces_endpoint='localhost:4317', export_metrics_to_file=False, export_metrics_to_file_dir=None, stat_loggers=None, constrained_json_whitespace_pattern=None, constrained_json_disable_any_whitespace=False, attention_backend='triton', decode_attention_backend=None, prefill_attention_backend=None, sampling_backend='flashinfer', grammar_backend='xgrammar', radix_cache_backend=None, mm_attention_backend=None, fp8_gemm_runner_backend='auto', fp4_gemm_runner_backend='auto', bf16_gemm_backend='auto', dsa_prefill_backend=None, dsa_decode_backend=None, dsa_paged_mqa_logits_backend='auto', dsa_topk_backend='sgl-kernel', disable_flashinfer_autotune=False, flashinfer_autotune_skip_ops=None, mamba_backend='triton', cuda_graph_config=CudaGraphConfig(decode=PhaseConfig(backend='full', max_bs=16, bs=[1, 2, 4, 8, 12, 16], tc_compiler='eager', full_prefill_max_req=None), prefill=PhaseConfig(backend='disabled', max_bs=2048, bs=[4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], tc_compiler='eager', full_prefill_max_req=None)), cuda_graph_backend_decode=None, cuda_graph_backend_prefill=None, cuda_graph_max_bs_decode=16, cuda_graph_max_bs_prefill=None, cuda_graph_bs_decode=None, cuda_graph_bs_prefill=None, cuda_graph_tc_compiler=None, disable_prefill_cuda_graph=False, disable_decode_cuda_graph=False, disable_cuda_graph=False, disable_cuda_graph_padding=False, enable_profile_cuda_graph=False, enable_cudagraph_gc=False, debug_cuda_graph=False, enable_layerwise_nvtx_marker=False, enable_nccl_nvls=False, enable_symm_mem=False, triton_attention_reduce_in_fp32=False, triton_attention_num_kv_splits=8, triton_attention_split_tile_size=None, flashinfer_mla_disable_ragged=False, enable_fused_qk_norm_rope=False, enable_precise_embedding_interpolation=False, enable_fused_moe_sum_all_reduce=False, enable_deepseek_v4_fp4_indexer=False, disable_custom_all_reduce=False, enable_mscclpp=False, enable_torch_symm_mem=False, enable_scattered_sconv=False, pre_warm_nccl=False, enable_quant_communications=False, enable_flashinfer_allreduce_fusion=False, enforce_disable_flashinfer_allreduce_fusion=False, flashinfer_allreduce_fusion_backend=None, enable_aiter_allreduce_fusion=False, enable_torch_compile=False, enable_torch_compile_debug_mode=False, torch_compile_max_bs=32, torchao_config='', speculative_algorithm=None, speculative_draft_model_path=None, speculative_draft_model_revision=None, speculative_draft_load_format=None, speculative_num_steps=None, speculative_eagle_topk=None, speculative_num_draft_tokens=None, speculative_dflash_block_size=None, speculative_dspark_block_size=None, speculative_dspark_sps_table_path=None, speculative_dspark_confidence_sts_path=None, speculative_dspark_align_verify_tokens_to_graph_tier=False, speculative_accept_threshold_single=1.0, speculative_accept_threshold_acc=1.0, speculative_use_rejection_sampling=False, speculative_token_map=None, speculative_attention_mode='prefill', speculative_draft_attention_backend=None, speculative_draft_window_size=None, speculative_moe_runner_backend='marlin', speculative_moe_a2a_backend=None, speculative_draft_model_quantization=None, speculative_skip_dp_mlp_sync=False, enable_multi_layer_eagle=False, speculative_adaptive=False, speculative_adaptive_config=None, decoupled_spec_bind_endpoint=None, decoupled_spec_connect_endpoints=None, decoupled_spec_rank=None, decoupled_spec_role='null', spec_trace_dir=None, speculative_ngram_min_bfs_breadth=1, speculative_ngram_max_bfs_breadth=10, speculative_ngram_match_type='BFS', speculative_ngram_max_trie_depth=18, speculative_ngram_capacity=10000000, speculative_ngram_external_corpus_path=None, speculative_ngram_external_sam_budget=0, speculative_ngram_external_corpus_max_tokens=10000000, ep_size=32, moe_a2a_backend='none', moe_runner_backend='marlin', flashinfer_mxfp4_moe_precision='default', deepep_mode='auto', fuseep_mode=2, deepep_dispatcher_output_dtype='auto', ep_num_redundant_experts=0, ep_dispatch_algorithm=None, init_expert_location='trivial', enable_eplb=False, eplb_algorithm='auto', eplb_rebalance_num_iterations=1000, eplb_rebalance_layers_per_chunk=None, eplb_min_rebalancing_utilization_threshold=1.0, expert_distribution_recorder_mode=None, expert_distribution_recorder_buffer_size=1000, enable_expert_distribution_metrics=False, deepep_config=None, moe_dense_tp_size=None, elastic_ep_backend=None, enable_elastic_expert_backup=False, mooncake_ib_device=None, enable_waterfill=False, ep_join_mode=None, ep_join_rank_offset=0, elastic_ep_initial_size=None, max_ep_size=None, elastic_ep_scale_timeout=600, elastic_ep_rejoin=False, disable_flashinfer_cutlass_moe_fp4_allgather=False, disable_shared_experts_fusion=False, enforce_shared_experts_fusion=False, max_mamba_cache_size=None, mamba_ssm_dtype=None, mamba_max_states_per_path=-1, enable_mamba_cache_stochastic_rounding=False, mamba_cache_philox_rounds=0, mamba_full_memory_ratio=0.21, mamba_radix_cache_strategy='extra_buffer_lazy', uses_mamba_radix_cache=False, mamba_track_interval=256, enable_int8_mamba_checkpoint=False, int8_mamba_ckpt_size=None, linear_attn_backend='triton', linear_attn_decode_backend=None, linear_attn_prefill_backend=None, linear_attn_verify_backend=None, enable_linear_replayssm=False, linear_replayssm_cache_len=16, enable_linear_replayssm_spec=False, enable_hierarchical_cache=False, hicache_ratio=2.0, hicache_size=0, hicache_write_policy='write_through', hicache_io_backend='kernel', hicache_mem_layout='page_first', hicache_storage_backend=None, hicache_storage_prefetch_policy='timeout', hicache_storage_backend_extra_config=None, enable_hisparse=False, hisparse_config=None, enable_broadcast_mm_inputs_process=False, enable_prefix_mm_cache=False, mm_enable_dp_encoder=False, mm_process_config={}, mm_processor_worker_num=0, mm_io_worker_num=0, limit_mm_data_per_request=None, enable_mm_global_cache=False, disable_fast_image_processor=False, mm_feature_transport='cpu', keep_mm_feature_on_device=False, enable_lora=None, enable_lora_overlap_loading=None, max_lora_rank=None, lora_target_modules=None, lora_paths=None, max_loaded_loras=None, max_loras_per_batch=8, lora_eviction_policy='lru', lora_backend='csgmv', max_lora_chunk_size=16, experts_shared_outer_loras=None, lora_use_virtual_experts=False, lora_strict_loading=False, lora_drain_wait_threshold=0.0, enable_two_batch_overlap=False, enable_single_batch_overlap=False, tbo_token_distribution_threshold=0.48, cpu_offload_gb=0, offload_group_size=-1, offload_num_in_group=1, offload_prefetch_step=1, offload_mode='cpu', enable_lmcache=False, lmcache_config_file=None, enable_flexkv=False, flexkv_config_file=None, kt_weight_path=None, kt_method='AMXINT4', kt_cpuinfer=None, kt_threadpool_count=2, kt_num_gpu_experts=None, kt_max_deferred_experts_per_token=None, dllm_algorithm=None, dllm_algorithm_config=None, dllm_fdfo=True, disaggregation_mode='decode', disaggregation_transfer_backend='mooncake', disaggregation_bootstrap_port=28800, disaggregation_ib_device='mlx5_0,mlx5_1,mlx5_2,mlx5_3', disaggregation_decode_enable_radix_cache=False, disaggregation_decode_enable_offload_kvcache=False, num_reserved_decode_tokens=512, disaggregation_decode_extra_slots=0, disaggregation_decode_polling_interval=1, optimistic_prefill_attempts=0, encoder_only=False, language_only=False, encoder_transfer_backend='zmq_to_tokenizer', encoder_urls=[], encoder_bootstrap_port=8997, encoder_register_urls=[], enable_adaptive_dispatch_to_encoder=False, enable_pdmux=False, pdmux_config_path=None, sm_group_num=8, custom_weight_loader=[], weight_loader_disable_mmap=False, weight_loader_prefetch_checkpoints=False, weight_loader_prefetch_num_threads=4, weight_loader_drop_cache_after_load=False, remote_instance_weight_loader_seed_instance_ip=None, remote_instance_weight_loader_seed_instance_service_port=None, remote_instance_weight_loader_send_weights_group_ports=None, remote_instance_weight_loader_backend='nccl', remote_instance_weight_loader_start_seed_via_transfer_engine=False, engine_info_bootstrap_port=6789, modelexpress_config=None, download_dir=None, model_checksum=None, delete_ckpt_after_loading=False, decrypted_config_file=None, decrypted_draft_config_file=None, checkpoint_engine_wait_weights_before_ready=False, enable_prefill_delayer=False, prefill_delayer_max_delay_passes=30, prefill_delayer_token_usage_low_watermark=None, prefill_delayer_forward_passes_buckets=None, prefill_delayer_wait_seconds_buckets=None, prefill_delayer_queue_min_ratio=None, prefill_delayer_max_delay_ms=None, min_free_slots_delay=None, enable_deterministic_inference=False, rl_on_policy_target=None, kv_canary='none', kv_canary_real_data='none', kv_canary_sweep_interval=0, enable_dynamic_batch_tokenizer=False, dynamic_batch_tokenizer_batch_size=32, dynamic_batch_tokenizer_batch_timeout=0.002, enable_tokenizer_batch_encode=False, disable_tokenizer_batch_decode=False, debug_tensor_dump_output_folder=None, debug_tensor_dump_layers=None, debug_tensor_dump_input_file=None, enable_memory_saver=False, enable_weights_cpu_backup=False, enable_draft_weights_cpu_backup=False, enable_custom_logit_processor=False, enable_return_hidden_states=False, enable_return_routed_experts=False, enable_return_indexer_topk=False, disable_outlines_disk_cache=False, enable_mis=False, weight_cache_mode='off', weight_cache_socket=None, weight_cache_timeout=1800, forward_hooks=None, msprobe_dump_config=None)
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:22 TP15 EP15] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:22 TP15 EP15] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:23 TP9 EP9] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:23 TP9 EP9] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:23 TP12 EP12] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:23 TP12 EP12] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:23 TP13 EP13] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:23 TP13 EP13] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:23 TP14 EP14] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:23 TP14 EP14] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:23 TP8 EP8] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:23 TP8 EP8] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:23 TP10 EP10] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:23 TP10 EP10] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:23 TP11 EP11] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:23 TP11 EP11] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:24 TP15 EP15] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:24 TP9 EP9] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:24 TP13 EP13] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:24 TP12 EP12] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:24 TP14 EP14] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:25 TP10 EP10] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:25 TP8 EP8] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:25 TP11 EP11] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:26.087226   242 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:26.087333   242 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:26.087350   242 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.6 port: 12001
+I0827 04:36:26.087409   242 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.6:16298
+I0827 04:36:26.087569   242 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:26.095995   236 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:26.096091   236 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:26.096107   236 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.6 port: 12001
+I0827 04:36:26.096150   236 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.6:15958
+I0827 04:36:26.096269   236 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:26.096457   240 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:26.096560   240 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:26.096577   240 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.6 port: 12001
+I0827 04:36:26.096626   240 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.6:16447
+I0827 04:36:26.096773   240 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:26.096876   241 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:26.096975   241 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:26.096992   241 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.6 port: 12001
+I0827 04:36:26.097035   241 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.6:16035
+I0827 04:36:26.097170   241 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:26.103180   241 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:26.107820   240 topology.cpp:127] Device mlx5_0 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:26.109326   239 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:26.109416   239 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:26.109428   239 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.6 port: 12001
+I0827 04:36:26.109464   239 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.6:16843
+I0827 04:36:26.109562   239 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:26.114297   242 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:26.114653   239 topology.cpp:127] Device mlx5_0 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:26.115540   237 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:26.115630   237 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:26.115645   237 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.6 port: 12001
+I0827 04:36:26.115686   237 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.6:15544
+I0827 04:36:26.115813   237 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:26.118220   240 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:26.121481   242 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:26.122418   237 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:26.126497   236 topology.cpp:127] Device mlx5_0 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:26.126574   235 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:26.126684   235 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:26.126703   235 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.6 port: 12001
+I0827 04:36:26.126758   235 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.6:15006
+I0827 04:36:26.126933   235 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:26.127116   240 topology.cpp:127] Device mlx5_2 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:26.127162   238 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:26.127270   238 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:26.127285   238 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.6 port: 12001
+I0827 04:36:26.127342   238 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.6:16140
+I0827 04:36:26.127489   238 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:26.128777   237 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:26.129559   242 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:26.130432   241 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:26.133821   238 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:26.134943   235 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:26.135799   240 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:26.136080   237 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:26.138710   241 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:26.139173   242 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:26.139978   240 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:26.140033   240 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:26.140090   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.140504   235 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:26.142936   242 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:26.142961   242 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:26.143014   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.144655   237 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:26.146834   235 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:26.147805   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:26.147867   241 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:26.148357   239 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:26.148370   237 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:26.148394   237 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:26.148444   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.149503   240 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:06
+I0827 04:36:26.149536   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.150290   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:26.150381   236 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:26.151773   242 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:06
+I0827 04:36:26.151803   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.152055   241 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:26.152087   241 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:26.152148   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.153594   235 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:26.156044   235 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:26.156075   235 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:26.156124   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.157910   238 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:26.159448   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:26.159547   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:26.160912   242 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:06
+I0827 04:36:26.160940   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.161144   241 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:06
+I0827 04:36:26.161176   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.162882   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:26.164348   235 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:06
+I0827 04:36:26.164378   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.165159   238 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:26.168772   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:26.170400   241 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:06
+I0827 04:36:26.170428   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.172459   238 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:26.174376   239 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:26.175243   238 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:26.175272   238 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:26.175325   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.178375   236 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:26.178763   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:26.178785   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:26.180411   237 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:06
+I0827 04:36:26.180462   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.180604   240 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:06
+I0827 04:36:26.180672   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.181643   239 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:26.182818   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:26.184404   238 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:06
+I0827 04:36:26.184437   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.184495   239 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:26.184537   239 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:26.184595   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.185601   236 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:26.186753   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:26.188311   242 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:06
+I0827 04:36:26.188344   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.188390   236 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:26.188416   236 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:26.188470   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.190752   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:26.192051   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:26.192406   235 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:06
+I0827 04:36:26.192444   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.193856   240 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:06
+I0827 04:36:26.193897   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.194197   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:26.195964   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:26.195997   238 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:06
+I0827 04:36:26.196039   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.196040   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:26.197574   242 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:06
+I0827 04:36:26.197599   236 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:06
+I0827 04:36:26.197629   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.197691   242 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:26 TP15 EP15] Init torch distributed begin.
+I0827 04:36:26.198544   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:26.200238   241 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:06
+I0827 04:36:26.200268   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.204000   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:26.205737   238 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:06
+I0827 04:36:26.205766   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.210784   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:26.212373   237 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:06
+I0827 04:36:26.212417   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.218751   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:26.220386   235 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:06
+I0827 04:36:26.220422   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.222854   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:26.223088   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:26.223217   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:26.223345   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:26.224500   241 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:06
+I0827 04:36:26.224536   240 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:06
+I0827 04:36:26.224612   241 transfer_engine_impl.cpp:388] installTransport, type=rdma
+I0827 04:36:26.224682   240 transfer_engine_impl.cpp:388] installTransport, type=rdma
+I0827 04:36:26.224973   236 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:06
+I0827 04:36:26.225018   236 rdma_context.cpp:205] Using SIEVE endpoint store
+[2026-08-27 04:36:26 TP14 EP14] Init torch distributed begin.
+[2026-08-27 04:36:26 TP13 EP13] Init torch distributed begin.
+I0827 04:36:26.225459   239 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:06
+I0827 04:36:26.225523   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.230858   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:26.232468   238 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:06
+I0827 04:36:26.232614   238 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:26 TP11 EP11] Init torch distributed begin.
+I0827 04:36:26.233065   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:26.234685   236 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:06
+I0827 04:36:26.234727   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.238961   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:26.240710   237 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:06
+I0827 04:36:26.240758   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.242511   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:26.244089   236 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:06
+I0827 04:36:26.244236   236 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:26 TP9 EP9] Init torch distributed begin.
+I0827 04:36:26.247059   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:26.248764   235 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:06
+I0827 04:36:26.248920   235 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:26 TP8 EP8] Init torch distributed begin.
+I0827 04:36:26.250787   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:26.252344   239 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:06
+I0827 04:36:26.252393   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.266752   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:26.268360   237 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:06
+I0827 04:36:26.268492   237 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:26 TP10 EP10] Init torch distributed begin.
+I0827 04:36:26.278649   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:26.280236   239 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:06
+I0827 04:36:26.280287   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.306850   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:26.308480   239 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:06
+I0827 04:36:26.308605   239 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:26 TP12 EP12] Init torch distributed begin.
+[2026-08-27 04:36:48 TP14 EP14] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP15 EP15] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP8 EP8] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP11 EP11] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP9 EP9] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP13 EP13] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP10 EP10] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP12 EP12] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP15 EP15] Init torch distributed ends. elapsed=22.03 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP14 EP14] Init torch distributed ends. elapsed=22.01 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP13 EP13] Init torch distributed ends. elapsed=22.01 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP12 EP12] Init torch distributed ends. elapsed=21.92 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP11 EP11] Init torch distributed ends. elapsed=22.00 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP9 EP9] Init torch distributed ends. elapsed=21.99 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP10 EP10] Init torch distributed ends. elapsed=21.96 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP8 EP8] Init torch distributed ends. elapsed=21.98 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP14 EP14] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP15 EP15] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP9 EP9] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP13 EP13] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP8 EP8] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP10 EP10] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP12 EP12] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP14 EP14] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP15 EP15] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP9 EP9] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP10 EP10] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP13 EP13] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP8 EP8] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP12 EP12] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP11 EP11] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP11 EP11] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP13 EP13] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP14 EP14] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP15 EP15] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP12 EP12] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP10 EP10] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP8 EP8] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP9 EP9] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP13 EP13] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP12 EP12] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP14 EP14] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP15 EP15] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP10 EP10] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP8 EP8] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP9 EP9] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP11 EP11] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP11 EP11] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:50 TP8 EP8] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP9 EP9] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP11 EP11] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP12 EP12] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP13 EP13] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP14 EP14] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP10 EP10] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP15 EP15] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:38:57 TP14 EP14] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:57 TP14 EP14] Load weight end. elapsed=129.27 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP9 EP9] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP9 EP9] Load weight end. elapsed=129.52 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP10 EP10] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP10 EP10] Load weight end. elapsed=129.56 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP8 EP8] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP8 EP8] Load weight end. elapsed=129.57 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP13 EP13] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP13 EP13] Load weight end. elapsed=129.58 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP11 EP11] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP11 EP11] Load weight end. elapsed=129.64 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP15 EP15] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP15 EP15] Load weight end. elapsed=129.77 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP12 EP12] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP12 EP12] Load weight end. elapsed=129.90 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:39:01 TP14 EP14] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP14 EP14] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP11 EP11] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP11 EP11] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP8 EP8] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP8 EP8] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP13 EP13] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP13 EP13] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP9 EP9] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP9 EP9] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP15 EP15] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP15 EP15] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP12 EP12] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP12 EP12] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP10 EP10] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP10 EP10] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:02 TP14 EP14] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP15 EP15] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP12 EP12] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP13 EP13] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP10 EP10] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP9 EP9] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP11 EP11] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP8 EP8] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP15 EP15] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP12 EP12] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP10 EP10] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP8 EP8] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP11 EP11] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP13 EP13] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP9 EP9] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP14 EP14] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP9 EP9] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP12 EP12] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP9 EP9] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP10 EP10] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP12 EP12] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP13 EP13] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP15 EP15] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP10 EP10] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP11 EP11] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP13 EP13] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP14 EP14] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP15 EP15] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP11 EP11] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP8 EP8] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP14 EP14] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP8 EP8] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP14 EP14] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP14 EP14] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP9 EP9] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP9 EP9] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP12 EP12] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP12 EP12] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:39:02 TP10 EP10] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP10 EP10] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP8 EP8] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP8 EP8] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:39:02 TP11 EP11] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP11 EP11] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:39:02 TP13 EP13] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP13 EP13] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP15 EP15] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP15 EP15] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/d_starting_d_node2.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/d_starting_d_node2.log
new file mode 100644
index 0000000..800cf83
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/d_starting_d_node2.log
@@ -0,0 +1,401 @@
+/usr/local/lib/python3.12/dist-packages/torchao/quantization/quant_api.py:1731: SyntaxWarning: invalid escape sequence '\.'
+  """Configuration class for applying different quantization configs to modules or parameters based on their fully qualified names (FQNs).
+/sgl-workspace/sglang/python/sglang/launch_server.py:56: UserWarning: 'python -m sglang.launch_server' is still supported, but 'sglang serve' is the recommended entrypoint.
+  Example: sglang serve --model-path <model> [options]
+  warnings.warn(
+[2026-08-27 04:35:59] KV cache is forced as chunk cache for decode server
+[2026-08-27 04:36:03] Breakable CUDA graph is incompatible with MLA attention; disabling prefill CUDA graph.
+[2026-08-27 04:36:03] Attention backend not specified. Use triton backend by default.
+[2026-08-27 04:36:06] server_args=ServerArgs(model_path='/data/hf_models/Kimi-K3', tokenizer_path='/data/hf_models/Kimi-K3', tokenizer_mode='auto', tokenizer_backend='huggingface', tokenizer_worker_num=1, detokenizer_worker_num=1, skip_tokenizer_init=False, load_format='auto', model_loader_extra_config='{}', trust_remote_code=True, context_length=None, is_embedding=False, enable_multimodal=None, revision=None, model_impl='auto', model_config_parser='auto', json_model_override_args='{}', dtype='auto', quantization=None, quantization_param_path=None, kv_cache_dtype='auto', enable_fp32_lm_head=False, modelopt_quant=None, modelopt_checkpoint_restore_path=None, modelopt_checkpoint_save_path=None, modelopt_export_path=None, quantize_and_serve=False, rl_quant_profile=None, enable_tf32_matmul=False, mem_fraction_static=0.88, max_running_requests=None, max_queued_requests=None, max_total_tokens=None, chunked_prefill_size=8192, enable_dynamic_chunking=False, max_prefill_tokens=16384, prefill_max_requests=None, schedule_policy='fcfs', enable_priority_scheduling=False, disable_priority_preemption=False, default_priority_value=None, abort_on_priority_when_disabled=False, schedule_low_priority_values_first=False, priority_scheduling_preemption_threshold=10, retraction_policy='length', schedule_conservativeness=1.0, page_size=64, swa_full_tokens_ratio=0.8, disable_hybrid_swa_memory=False, radix_eviction_policy='lru', prefill_only_disable_kv_cache=False, disable_radix_cache=True, enable_page_major_kv_layout=False, enable_unified_memory=False, disable_chunked_prefix_cache=False, disable_overlap_schedule=False, num_continuous_decode_steps=1, scheduler_recv_interval=1, enable_mixed_chunk=False, nccl_port=None, dist_timeout=3600, dist_init_addr='174.1.60.5:20000', nnodes=4, node_rank=2, tp_size=32, dcp_size=1, pp_size=1, pp_max_micro_batch_size=None, pp_async_batch_depth=0, dp_size=1, load_balance_method='round_robin', attn_cp_size=1, moe_dp_size=1, dcp_comm_backend='ag_rs', dcp_replicate_q_proj=None, dwdp_size=1, enable_prefill_cp=False, cp_strategy=None, enable_dsa_cache_layer_split=False, enable_dsa_prefill_context_parallel=False, dsa_prefill_cp_mode='round-robin-split', enable_prefill_context_parallel=False, prefill_cp_mode='in-seq-split', enable_cp_decode_attn_tp=False, enable_dp_attention=False, enable_dp_attention_local_control_broadcast=False, enable_dp_lm_head=False, enable_attn_tp_input_scattered=False, disable_attn_tp_gather=False, enable_p2p_check=False, device='cuda', base_gpu_id=0, gpu_id_step=1, random_seed=212084647, watchdog_timeout=300, soft_watchdog_timeout=None, sleep_on_idle=False, use_ray=False, custom_sigquit_handler=None, numa_node=None, gc_threshold=None, host='0.0.0.0', port=30000, fastapi_root_path='', smg_grpc_mode=False, grpc_mode=False, grpc_port=None, sidecar=None, sidecar_args=None, skip_server_warmup=False, warmups=None, enable_http2=False, ssl_keyfile=None, ssl_certfile=None, ssl_ca_certs=None, ssl_keyfile_password=None, enable_ssl_refresh=False, api_key=None, admin_api_key=None, served_model_name='kimi-k3', weight_version='default', chat_template=None, hf_chat_template_name=None, completion_template=None, file_storage_path='sglang_storage', enable_cache_report=False, reasoning_parser=None, default_chat_template_kwargs=None, strip_thinking_cache=False, enable_strict_thinking=False, tool_call_parser=None, tool_server=None, sampling_defaults='model', asr_max_buffer_seconds=60, asr_max_concurrent_sessions=32, preferred_sampling_params=None, allow_auto_truncate=False, stream_interval=1, batch_notify_size=16, stream_response_default_include_usage=False, incremental_streaming_output=False, enable_streaming_session=False, enable_session_radix_cache=False, log_level='info', log_level_http=None, log_requests=False, log_requests_level=2, log_requests_format='text', log_requests_target=None, uvicorn_access_log_exclude_prefixes=[], crash_dump_folder=None, show_time_cost=False, enable_metrics=False, smg_http_sidecar_port=None, enable_mfu_metrics=False, enable_metrics_for_all_schedulers=False, load_snapshot_publish_interval=15, tokenizer_metrics_custom_labels_header='x-custom-labels', tokenizer_metrics_allowed_custom_labels=None, extra_metric_labels=None, bucket_time_to_first_token=None, bucket_inter_token_latency=None, bucket_e2e_request_latency=None, prompt_tokens_buckets=None, generation_tokens_buckets=None, gc_warning_threshold_secs=0.0, decode_log_interval=40, enable_request_time_stats_logging=False, kv_events_config=None, enable_forward_pass_metrics=False, forward_pass_metrics_worker_id='', forward_pass_metrics_ipc_name=None, enable_trace=False, trace_modules='request', otlp_traces_endpoint='localhost:4317', export_metrics_to_file=False, export_metrics_to_file_dir=None, stat_loggers=None, constrained_json_whitespace_pattern=None, constrained_json_disable_any_whitespace=False, attention_backend='triton', decode_attention_backend=None, prefill_attention_backend=None, sampling_backend='flashinfer', grammar_backend='xgrammar', radix_cache_backend=None, mm_attention_backend=None, fp8_gemm_runner_backend='auto', fp4_gemm_runner_backend='auto', bf16_gemm_backend='auto', dsa_prefill_backend=None, dsa_decode_backend=None, dsa_paged_mqa_logits_backend='auto', dsa_topk_backend='sgl-kernel', disable_flashinfer_autotune=False, flashinfer_autotune_skip_ops=None, mamba_backend='triton', cuda_graph_config=CudaGraphConfig(decode=PhaseConfig(backend='full', max_bs=16, bs=[1, 2, 4, 8, 12, 16], tc_compiler='eager', full_prefill_max_req=None), prefill=PhaseConfig(backend='disabled', max_bs=2048, bs=[4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], tc_compiler='eager', full_prefill_max_req=None)), cuda_graph_backend_decode=None, cuda_graph_backend_prefill=None, cuda_graph_max_bs_decode=16, cuda_graph_max_bs_prefill=None, cuda_graph_bs_decode=None, cuda_graph_bs_prefill=None, cuda_graph_tc_compiler=None, disable_prefill_cuda_graph=False, disable_decode_cuda_graph=False, disable_cuda_graph=False, disable_cuda_graph_padding=False, enable_profile_cuda_graph=False, enable_cudagraph_gc=False, debug_cuda_graph=False, enable_layerwise_nvtx_marker=False, enable_nccl_nvls=False, enable_symm_mem=False, triton_attention_reduce_in_fp32=False, triton_attention_num_kv_splits=8, triton_attention_split_tile_size=None, flashinfer_mla_disable_ragged=False, enable_fused_qk_norm_rope=False, enable_precise_embedding_interpolation=False, enable_fused_moe_sum_all_reduce=False, enable_deepseek_v4_fp4_indexer=False, disable_custom_all_reduce=False, enable_mscclpp=False, enable_torch_symm_mem=False, enable_scattered_sconv=False, pre_warm_nccl=False, enable_quant_communications=False, enable_flashinfer_allreduce_fusion=False, enforce_disable_flashinfer_allreduce_fusion=False, flashinfer_allreduce_fusion_backend=None, enable_aiter_allreduce_fusion=False, enable_torch_compile=False, enable_torch_compile_debug_mode=False, torch_compile_max_bs=32, torchao_config='', speculative_algorithm=None, speculative_draft_model_path=None, speculative_draft_model_revision=None, speculative_draft_load_format=None, speculative_num_steps=None, speculative_eagle_topk=None, speculative_num_draft_tokens=None, speculative_dflash_block_size=None, speculative_dspark_block_size=None, speculative_dspark_sps_table_path=None, speculative_dspark_confidence_sts_path=None, speculative_dspark_align_verify_tokens_to_graph_tier=False, speculative_accept_threshold_single=1.0, speculative_accept_threshold_acc=1.0, speculative_use_rejection_sampling=False, speculative_token_map=None, speculative_attention_mode='prefill', speculative_draft_attention_backend=None, speculative_draft_window_size=None, speculative_moe_runner_backend='marlin', speculative_moe_a2a_backend=None, speculative_draft_model_quantization=None, speculative_skip_dp_mlp_sync=False, enable_multi_layer_eagle=False, speculative_adaptive=False, speculative_adaptive_config=None, decoupled_spec_bind_endpoint=None, decoupled_spec_connect_endpoints=None, decoupled_spec_rank=None, decoupled_spec_role='null', spec_trace_dir=None, speculative_ngram_min_bfs_breadth=1, speculative_ngram_max_bfs_breadth=10, speculative_ngram_match_type='BFS', speculative_ngram_max_trie_depth=18, speculative_ngram_capacity=10000000, speculative_ngram_external_corpus_path=None, speculative_ngram_external_sam_budget=0, speculative_ngram_external_corpus_max_tokens=10000000, ep_size=32, moe_a2a_backend='none', moe_runner_backend='marlin', flashinfer_mxfp4_moe_precision='default', deepep_mode='auto', fuseep_mode=2, deepep_dispatcher_output_dtype='auto', ep_num_redundant_experts=0, ep_dispatch_algorithm=None, init_expert_location='trivial', enable_eplb=False, eplb_algorithm='auto', eplb_rebalance_num_iterations=1000, eplb_rebalance_layers_per_chunk=None, eplb_min_rebalancing_utilization_threshold=1.0, expert_distribution_recorder_mode=None, expert_distribution_recorder_buffer_size=1000, enable_expert_distribution_metrics=False, deepep_config=None, moe_dense_tp_size=None, elastic_ep_backend=None, enable_elastic_expert_backup=False, mooncake_ib_device=None, enable_waterfill=False, ep_join_mode=None, ep_join_rank_offset=0, elastic_ep_initial_size=None, max_ep_size=None, elastic_ep_scale_timeout=600, elastic_ep_rejoin=False, disable_flashinfer_cutlass_moe_fp4_allgather=False, disable_shared_experts_fusion=False, enforce_shared_experts_fusion=False, max_mamba_cache_size=None, mamba_ssm_dtype=None, mamba_max_states_per_path=-1, enable_mamba_cache_stochastic_rounding=False, mamba_cache_philox_rounds=0, mamba_full_memory_ratio=0.21, mamba_radix_cache_strategy='extra_buffer_lazy', uses_mamba_radix_cache=False, mamba_track_interval=256, enable_int8_mamba_checkpoint=False, int8_mamba_ckpt_size=None, linear_attn_backend='triton', linear_attn_decode_backend=None, linear_attn_prefill_backend=None, linear_attn_verify_backend=None, enable_linear_replayssm=False, linear_replayssm_cache_len=16, enable_linear_replayssm_spec=False, enable_hierarchical_cache=False, hicache_ratio=2.0, hicache_size=0, hicache_write_policy='write_through', hicache_io_backend='kernel', hicache_mem_layout='page_first', hicache_storage_backend=None, hicache_storage_prefetch_policy='timeout', hicache_storage_backend_extra_config=None, enable_hisparse=False, hisparse_config=None, enable_broadcast_mm_inputs_process=False, enable_prefix_mm_cache=False, mm_enable_dp_encoder=False, mm_process_config={}, mm_processor_worker_num=0, mm_io_worker_num=0, limit_mm_data_per_request=None, enable_mm_global_cache=False, disable_fast_image_processor=False, mm_feature_transport='cpu', keep_mm_feature_on_device=False, enable_lora=None, enable_lora_overlap_loading=None, max_lora_rank=None, lora_target_modules=None, lora_paths=None, max_loaded_loras=None, max_loras_per_batch=8, lora_eviction_policy='lru', lora_backend='csgmv', max_lora_chunk_size=16, experts_shared_outer_loras=None, lora_use_virtual_experts=False, lora_strict_loading=False, lora_drain_wait_threshold=0.0, enable_two_batch_overlap=False, enable_single_batch_overlap=False, tbo_token_distribution_threshold=0.48, cpu_offload_gb=0, offload_group_size=-1, offload_num_in_group=1, offload_prefetch_step=1, offload_mode='cpu', enable_lmcache=False, lmcache_config_file=None, enable_flexkv=False, flexkv_config_file=None, kt_weight_path=None, kt_method='AMXINT4', kt_cpuinfer=None, kt_threadpool_count=2, kt_num_gpu_experts=None, kt_max_deferred_experts_per_token=None, dllm_algorithm=None, dllm_algorithm_config=None, dllm_fdfo=True, disaggregation_mode='decode', disaggregation_transfer_backend='mooncake', disaggregation_bootstrap_port=28800, disaggregation_ib_device='mlx5_0,mlx5_1,mlx5_2,mlx5_3', disaggregation_decode_enable_radix_cache=False, disaggregation_decode_enable_offload_kvcache=False, num_reserved_decode_tokens=512, disaggregation_decode_extra_slots=0, disaggregation_decode_polling_interval=1, optimistic_prefill_attempts=0, encoder_only=False, language_only=False, encoder_transfer_backend='zmq_to_tokenizer', encoder_urls=[], encoder_bootstrap_port=8997, encoder_register_urls=[], enable_adaptive_dispatch_to_encoder=False, enable_pdmux=False, pdmux_config_path=None, sm_group_num=8, custom_weight_loader=[], weight_loader_disable_mmap=False, weight_loader_prefetch_checkpoints=False, weight_loader_prefetch_num_threads=4, weight_loader_drop_cache_after_load=False, remote_instance_weight_loader_seed_instance_ip=None, remote_instance_weight_loader_seed_instance_service_port=None, remote_instance_weight_loader_send_weights_group_ports=None, remote_instance_weight_loader_backend='nccl', remote_instance_weight_loader_start_seed_via_transfer_engine=False, engine_info_bootstrap_port=6789, modelexpress_config=None, download_dir=None, model_checksum=None, delete_ckpt_after_loading=False, decrypted_config_file=None, decrypted_draft_config_file=None, checkpoint_engine_wait_weights_before_ready=False, enable_prefill_delayer=False, prefill_delayer_max_delay_passes=30, prefill_delayer_token_usage_low_watermark=None, prefill_delayer_forward_passes_buckets=None, prefill_delayer_wait_seconds_buckets=None, prefill_delayer_queue_min_ratio=None, prefill_delayer_max_delay_ms=None, min_free_slots_delay=None, enable_deterministic_inference=False, rl_on_policy_target=None, kv_canary='none', kv_canary_real_data='none', kv_canary_sweep_interval=0, enable_dynamic_batch_tokenizer=False, dynamic_batch_tokenizer_batch_size=32, dynamic_batch_tokenizer_batch_timeout=0.002, enable_tokenizer_batch_encode=False, disable_tokenizer_batch_decode=False, debug_tensor_dump_output_folder=None, debug_tensor_dump_layers=None, debug_tensor_dump_input_file=None, enable_memory_saver=False, enable_weights_cpu_backup=False, enable_draft_weights_cpu_backup=False, enable_custom_logit_processor=False, enable_return_hidden_states=False, enable_return_routed_experts=False, enable_return_indexer_topk=False, disable_outlines_disk_cache=False, enable_mis=False, weight_cache_mode='off', weight_cache_socket=None, weight_cache_timeout=1800, forward_hooks=None, msprobe_dump_config=None)
+[2026-08-27 04:36:06] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:06] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:06] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:06] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:06] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:06] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:06] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:06] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:24 TP18 EP18] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:24 TP18 EP18] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:24 TP21 EP21] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:24 TP21 EP21] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:24 TP20 EP20] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:24 TP20 EP20] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:25 TP19 EP19] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:25 TP19 EP19] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:25 TP22 EP22] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:25 TP22 EP22] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:25 TP23 EP23] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:25 TP23 EP23] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:26 TP17 EP17] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:26 TP17 EP17] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:26 TP16 EP16] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:26 TP16 EP16] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:27 TP19 EP19] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:27 TP21 EP21] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:27 TP18 EP18] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:27 TP20 EP20] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:27 TP23 EP23] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:27 TP22 EP22] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:27 TP17 EP17] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:27 TP16 EP16] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:29.187384   239 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:29.187532   239 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:29.187558   239 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.7 port: 12001
+I0827 04:36:29.187624   239 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.7:16248
+I0827 04:36:29.187814   239 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:29.191661   235 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:29.191802   235 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:29.191823   235 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.7 port: 12001
+I0827 04:36:29.191901   235 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.7:16408
+I0827 04:36:29.192099   235 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:29.195215   242 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:29.195319   242 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:29.195333   242 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.7 port: 12001
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:29.195333   238 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:29.195374   242 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.7:15148
+I0827 04:36:29.195490   238 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:29.195513   238 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.7 port: 12001
+I0827 04:36:29.195519   242 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:29.195581   238 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.7:16909
+I0827 04:36:29.195815   238 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:29.198207   236 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:29.198346   236 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:29.198370   236 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.7 port: 12001
+I0827 04:36:29.198427   236 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.7:16738
+I0827 04:36:29.198611   236 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:29.200070   242 topology.cpp:127] Device mlx5_0 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:29.201086   237 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:29.201221   237 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:29.201239   237 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.7 port: 12001
+I0827 04:36:29.201300   237 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.7:16522
+I0827 04:36:29.201495   237 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:29.201511   240 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:29.201632   240 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:29.201648   240 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.7 port: 12001
+I0827 04:36:29.201701   240 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.7:16431
+I0827 04:36:29.201871   240 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:29.204236   241 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:29.204355   242 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:29.204380   241 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:29.204393   241 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.7 port: 12001
+I0827 04:36:29.204454   241 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.7:16508
+I0827 04:36:29.204633   241 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:29.204896   238 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:29.207675   237 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:29.208006   240 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:29.208549   242 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:29.212560   242 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:29.214218   242 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:29.214243   242 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:29.214290   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.214713   240 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:29.214757   237 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:29.219102   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:29.219870   239 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:29.220057   235 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:29.220343   242 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:07
+I0827 04:36:29.220369   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.222033   240 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:29.222069   237 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:29.223913   236 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:29.228577   235 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:29.228634   239 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:29.230391   237 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:29.230446   240 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:29.232455   236 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:29.233546   237 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:29.233578   237 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:29.233631   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.234635   240 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:29.234665   240 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:29.234716   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.235774   238 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:29.235827   241 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:29.237140   235 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:29.241117   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:29.241130   236 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:29.242069   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:29.242725   237 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:07
+I0827 04:36:29.242758   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.243661   240 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:07
+I0827 04:36:29.243691   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.245620   235 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:29.247138   238 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:29.247233   241 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:29.248172   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:29.248514   235 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:29.248550   235 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:29.248612   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.248719   236 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:29.249847   242 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:07
+I0827 04:36:29.249886   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.251354   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:29.251672   236 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:29.251719   236 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:29.251781   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.252889   240 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:07
+I0827 04:36:29.252921   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.255584   239 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:29.256263   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:29.257805   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:29.257980   235 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:07
+I0827 04:36:29.258016   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.258512   238 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:29.258567   241 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:29.259447   242 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:07
+I0827 04:36:29.259477   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.259642   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:29.261375   236 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:07
+I0827 04:36:29.261410   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.261518   238 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:29.261557   238 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:29.261615   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.261615   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:29.262753   239 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:29.263202   240 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:07
+I0827 04:36:29.263229   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.265475   239 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:29.265504   239 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:29.265558   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.269449   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:29.271157   236 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:07
+I0827 04:36:29.271186   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.271761   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:29.272920   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:29.273422   237 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:07
+I0827 04:36:29.273456   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.274566   239 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:07
+I0827 04:36:29.274600   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.281176   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:29.282341   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:29.282708   237 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:07
+I0827 04:36:29.282739   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.284137   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:29.284178   239 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:07
+I0827 04:36:29.284219   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.285904   235 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:07
+I0827 04:36:29.285953   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.288185   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:29.289810   242 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:07
+I0827 04:36:29.289940   242 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:29 TP23 EP23] Init torch distributed begin.
+I0827 04:36:29.291602   241 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:29.291882   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:29.291936   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:29.293576   240 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:07
+I0827 04:36:29.293699   240 transfer_engine_impl.cpp:388] installTransport, type=rdma
+I0827 04:36:29.293921   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+[2026-08-27 04:36:29 TP21 EP21] Init torch distributed begin.
+I0827 04:36:29.294121   238 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:07
+I0827 04:36:29.294169   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.294241   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:29.295475   239 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:07
+I0827 04:36:29.295507   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.295553   241 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:29.295591   241 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:29.295648   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.296002   235 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:07
+I0827 04:36:29.296041   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.296232   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:29.298113   236 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:07
+I0827 04:36:29.298154   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.303915   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:29.305596   235 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:07
+I0827 04:36:29.305743   235 transfer_engine_impl.cpp:388] installTransport, type=rdma
+I0827 04:36:29.306217   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+[2026-08-27 04:36:29 TP16 EP16] Init torch distributed begin.
+I0827 04:36:29.307994   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:29.308105   236 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:07
+I0827 04:36:29.308277   236 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:29 TP17 EP17] Init torch distributed begin.
+I0827 04:36:29.310081   237 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:07
+I0827 04:36:29.310261   237 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:29 TP18 EP18] Init torch distributed begin.
+I0827 04:36:29.320421   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:29.321812   239 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:07
+I0827 04:36:29.321919   239 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:29 TP20 EP20] Init torch distributed begin.
+I0827 04:36:29.323829   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:29.323894   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:29.325821   238 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:07
+I0827 04:36:29.325820   241 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:07
+I0827 04:36:29.325871   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.325872   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.360142   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:29.360253   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:29.362092   241 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:07
+I0827 04:36:29.362147   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.362201   238 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:07
+I0827 04:36:29.362243   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.392091   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:29.392130   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:29.394111   238 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:07
+I0827 04:36:29.394127   241 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:07
+I0827 04:36:29.394166   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.394286   238 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:29 TP19 EP19] Init torch distributed begin.
+I0827 04:36:29.420166   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:29.421864   241 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:07
+I0827 04:36:29.421993   241 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:29 TP22 EP22] Init torch distributed begin.
+[2026-08-27 04:36:48 TP23 EP23] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP22 EP22] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP21 EP21] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP20 EP20] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP19 EP19] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP18 EP18] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP17 EP17] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP16 EP16] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP22 EP22] Init torch distributed ends. elapsed=18.81 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP23 EP23] Init torch distributed ends. elapsed=18.94 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP21 EP21] Init torch distributed ends. elapsed=18.93 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP20 EP20] Init torch distributed ends. elapsed=18.91 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP19 EP19] Init torch distributed ends. elapsed=18.83 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP18 EP18] Init torch distributed ends. elapsed=18.92 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP17 EP17] Init torch distributed ends. elapsed=18.92 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP16 EP16] Init torch distributed ends. elapsed=18.92 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP21 EP21] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP16 EP16] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP22 EP22] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP20 EP20] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP23 EP23] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP18 EP18] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP19 EP19] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP22 EP22] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP20 EP20] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP21 EP21] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP16 EP16] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP23 EP23] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP19 EP19] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP18 EP18] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP17 EP17] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP17 EP17] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP20 EP20] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP22 EP22] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP18 EP18] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP23 EP23] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP19 EP19] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP21 EP21] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP20 EP20] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP16 EP16] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP18 EP18] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP22 EP22] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP19 EP19] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP21 EP21] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP23 EP23] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP16 EP16] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP17 EP17] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP17 EP17] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:50 TP23 EP23] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP22 EP22] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP21 EP21] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP19 EP19] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP18 EP18] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP20 EP20] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP16 EP16] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP17 EP17] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:38:53 TP16 EP16] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:53 TP16 EP16] Load weight end. elapsed=124.66 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:53 TP17 EP17] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:53 TP17 EP17] Load weight end. elapsed=124.69 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:53 TP23 EP23] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:53 TP23 EP23] Load weight end. elapsed=124.71 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:53 TP21 EP21] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:53 TP21 EP21] Load weight end. elapsed=124.74 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:53 TP18 EP18] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:53 TP18 EP18] Load weight end. elapsed=124.82 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:53 TP20 EP20] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:53 TP20 EP20] Load weight end. elapsed=124.82 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:53 TP22 EP22] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:53 TP22 EP22] Load weight end. elapsed=124.92 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:53 TP19 EP19] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:53 TP19 EP19] Load weight end. elapsed=125.03 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:39:01 TP17 EP17] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP17 EP17] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP23 EP23] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP23 EP23] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP16 EP16] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP16 EP16] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP18 EP18] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP18 EP18] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP22 EP22] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP22 EP22] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP21 EP21] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP21 EP21] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP20 EP20] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP20 EP20] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP19 EP19] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP19 EP19] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:02 TP23 EP23] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP18 EP18] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP21 EP21] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP22 EP22] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP20 EP20] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP17 EP17] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP19 EP19] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP16 EP16] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP21 EP21] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP22 EP22] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP23 EP23] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP17 EP17] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP16 EP16] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP20 EP20] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP18 EP18] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP19 EP19] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP21 EP21] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP23 EP23] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP22 EP22] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP17 EP17] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP23 EP23] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP21 EP21] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP20 EP20] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP16 EP16] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP22 EP22] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP17 EP17] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP20 EP20] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP16 EP16] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP18 EP18] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP19 EP19] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP18 EP18] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP19 EP19] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP23 EP23] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP23 EP23] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:39:02 TP17 EP17] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP17 EP17] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP16 EP16] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP16 EP16] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:39:02 TP21 EP21] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP21 EP21] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP22 EP22] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP22 EP22] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP20 EP20] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP20 EP20] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:39:02 TP18 EP18] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP18 EP18] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP19 EP19] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP19 EP19] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/d_starting_d_node3.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/d_starting_d_node3.log
new file mode 100644
index 0000000..8bc9ec6
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/d_starting_d_node3.log
@@ -0,0 +1,401 @@
+/usr/local/lib/python3.12/dist-packages/torchao/quantization/quant_api.py:1731: SyntaxWarning: invalid escape sequence '\.'
+  """Configuration class for applying different quantization configs to modules or parameters based on their fully qualified names (FQNs).
+/sgl-workspace/sglang/python/sglang/launch_server.py:56: UserWarning: 'python -m sglang.launch_server' is still supported, but 'sglang serve' is the recommended entrypoint.
+  Example: sglang serve --model-path <model> [options]
+  warnings.warn(
+[2026-08-27 04:36:00] KV cache is forced as chunk cache for decode server
+[2026-08-27 04:36:03] Breakable CUDA graph is incompatible with MLA attention; disabling prefill CUDA graph.
+[2026-08-27 04:36:03] Attention backend not specified. Use triton backend by default.
+[2026-08-27 04:36:05] server_args=ServerArgs(model_path='/data/hf_models/Kimi-K3', tokenizer_path='/data/hf_models/Kimi-K3', tokenizer_mode='auto', tokenizer_backend='huggingface', tokenizer_worker_num=1, detokenizer_worker_num=1, skip_tokenizer_init=False, load_format='auto', model_loader_extra_config='{}', trust_remote_code=True, context_length=None, is_embedding=False, enable_multimodal=None, revision=None, model_impl='auto', model_config_parser='auto', json_model_override_args='{}', dtype='auto', quantization=None, quantization_param_path=None, kv_cache_dtype='auto', enable_fp32_lm_head=False, modelopt_quant=None, modelopt_checkpoint_restore_path=None, modelopt_checkpoint_save_path=None, modelopt_export_path=None, quantize_and_serve=False, rl_quant_profile=None, enable_tf32_matmul=False, mem_fraction_static=0.88, max_running_requests=None, max_queued_requests=None, max_total_tokens=None, chunked_prefill_size=8192, enable_dynamic_chunking=False, max_prefill_tokens=16384, prefill_max_requests=None, schedule_policy='fcfs', enable_priority_scheduling=False, disable_priority_preemption=False, default_priority_value=None, abort_on_priority_when_disabled=False, schedule_low_priority_values_first=False, priority_scheduling_preemption_threshold=10, retraction_policy='length', schedule_conservativeness=1.0, page_size=64, swa_full_tokens_ratio=0.8, disable_hybrid_swa_memory=False, radix_eviction_policy='lru', prefill_only_disable_kv_cache=False, disable_radix_cache=True, enable_page_major_kv_layout=False, enable_unified_memory=False, disable_chunked_prefix_cache=False, disable_overlap_schedule=False, num_continuous_decode_steps=1, scheduler_recv_interval=1, enable_mixed_chunk=False, nccl_port=None, dist_timeout=3600, dist_init_addr='174.1.60.5:20000', nnodes=4, node_rank=3, tp_size=32, dcp_size=1, pp_size=1, pp_max_micro_batch_size=None, pp_async_batch_depth=0, dp_size=1, load_balance_method='round_robin', attn_cp_size=1, moe_dp_size=1, dcp_comm_backend='ag_rs', dcp_replicate_q_proj=None, dwdp_size=1, enable_prefill_cp=False, cp_strategy=None, enable_dsa_cache_layer_split=False, enable_dsa_prefill_context_parallel=False, dsa_prefill_cp_mode='round-robin-split', enable_prefill_context_parallel=False, prefill_cp_mode='in-seq-split', enable_cp_decode_attn_tp=False, enable_dp_attention=False, enable_dp_attention_local_control_broadcast=False, enable_dp_lm_head=False, enable_attn_tp_input_scattered=False, disable_attn_tp_gather=False, enable_p2p_check=False, device='cuda', base_gpu_id=0, gpu_id_step=1, random_seed=297832107, watchdog_timeout=300, soft_watchdog_timeout=None, sleep_on_idle=False, use_ray=False, custom_sigquit_handler=None, numa_node=None, gc_threshold=None, host='0.0.0.0', port=30000, fastapi_root_path='', smg_grpc_mode=False, grpc_mode=False, grpc_port=None, sidecar=None, sidecar_args=None, skip_server_warmup=False, warmups=None, enable_http2=False, ssl_keyfile=None, ssl_certfile=None, ssl_ca_certs=None, ssl_keyfile_password=None, enable_ssl_refresh=False, api_key=None, admin_api_key=None, served_model_name='kimi-k3', weight_version='default', chat_template=None, hf_chat_template_name=None, completion_template=None, file_storage_path='sglang_storage', enable_cache_report=False, reasoning_parser=None, default_chat_template_kwargs=None, strip_thinking_cache=False, enable_strict_thinking=False, tool_call_parser=None, tool_server=None, sampling_defaults='model', asr_max_buffer_seconds=60, asr_max_concurrent_sessions=32, preferred_sampling_params=None, allow_auto_truncate=False, stream_interval=1, batch_notify_size=16, stream_response_default_include_usage=False, incremental_streaming_output=False, enable_streaming_session=False, enable_session_radix_cache=False, log_level='info', log_level_http=None, log_requests=False, log_requests_level=2, log_requests_format='text', log_requests_target=None, uvicorn_access_log_exclude_prefixes=[], crash_dump_folder=None, show_time_cost=False, enable_metrics=False, smg_http_sidecar_port=None, enable_mfu_metrics=False, enable_metrics_for_all_schedulers=False, load_snapshot_publish_interval=15, tokenizer_metrics_custom_labels_header='x-custom-labels', tokenizer_metrics_allowed_custom_labels=None, extra_metric_labels=None, bucket_time_to_first_token=None, bucket_inter_token_latency=None, bucket_e2e_request_latency=None, prompt_tokens_buckets=None, generation_tokens_buckets=None, gc_warning_threshold_secs=0.0, decode_log_interval=40, enable_request_time_stats_logging=False, kv_events_config=None, enable_forward_pass_metrics=False, forward_pass_metrics_worker_id='', forward_pass_metrics_ipc_name=None, enable_trace=False, trace_modules='request', otlp_traces_endpoint='localhost:4317', export_metrics_to_file=False, export_metrics_to_file_dir=None, stat_loggers=None, constrained_json_whitespace_pattern=None, constrained_json_disable_any_whitespace=False, attention_backend='triton', decode_attention_backend=None, prefill_attention_backend=None, sampling_backend='flashinfer', grammar_backend='xgrammar', radix_cache_backend=None, mm_attention_backend=None, fp8_gemm_runner_backend='auto', fp4_gemm_runner_backend='auto', bf16_gemm_backend='auto', dsa_prefill_backend=None, dsa_decode_backend=None, dsa_paged_mqa_logits_backend='auto', dsa_topk_backend='sgl-kernel', disable_flashinfer_autotune=False, flashinfer_autotune_skip_ops=None, mamba_backend='triton', cuda_graph_config=CudaGraphConfig(decode=PhaseConfig(backend='full', max_bs=16, bs=[1, 2, 4, 8, 12, 16], tc_compiler='eager', full_prefill_max_req=None), prefill=PhaseConfig(backend='disabled', max_bs=2048, bs=[4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], tc_compiler='eager', full_prefill_max_req=None)), cuda_graph_backend_decode=None, cuda_graph_backend_prefill=None, cuda_graph_max_bs_decode=16, cuda_graph_max_bs_prefill=None, cuda_graph_bs_decode=None, cuda_graph_bs_prefill=None, cuda_graph_tc_compiler=None, disable_prefill_cuda_graph=False, disable_decode_cuda_graph=False, disable_cuda_graph=False, disable_cuda_graph_padding=False, enable_profile_cuda_graph=False, enable_cudagraph_gc=False, debug_cuda_graph=False, enable_layerwise_nvtx_marker=False, enable_nccl_nvls=False, enable_symm_mem=False, triton_attention_reduce_in_fp32=False, triton_attention_num_kv_splits=8, triton_attention_split_tile_size=None, flashinfer_mla_disable_ragged=False, enable_fused_qk_norm_rope=False, enable_precise_embedding_interpolation=False, enable_fused_moe_sum_all_reduce=False, enable_deepseek_v4_fp4_indexer=False, disable_custom_all_reduce=False, enable_mscclpp=False, enable_torch_symm_mem=False, enable_scattered_sconv=False, pre_warm_nccl=False, enable_quant_communications=False, enable_flashinfer_allreduce_fusion=False, enforce_disable_flashinfer_allreduce_fusion=False, flashinfer_allreduce_fusion_backend=None, enable_aiter_allreduce_fusion=False, enable_torch_compile=False, enable_torch_compile_debug_mode=False, torch_compile_max_bs=32, torchao_config='', speculative_algorithm=None, speculative_draft_model_path=None, speculative_draft_model_revision=None, speculative_draft_load_format=None, speculative_num_steps=None, speculative_eagle_topk=None, speculative_num_draft_tokens=None, speculative_dflash_block_size=None, speculative_dspark_block_size=None, speculative_dspark_sps_table_path=None, speculative_dspark_confidence_sts_path=None, speculative_dspark_align_verify_tokens_to_graph_tier=False, speculative_accept_threshold_single=1.0, speculative_accept_threshold_acc=1.0, speculative_use_rejection_sampling=False, speculative_token_map=None, speculative_attention_mode='prefill', speculative_draft_attention_backend=None, speculative_draft_window_size=None, speculative_moe_runner_backend='marlin', speculative_moe_a2a_backend=None, speculative_draft_model_quantization=None, speculative_skip_dp_mlp_sync=False, enable_multi_layer_eagle=False, speculative_adaptive=False, speculative_adaptive_config=None, decoupled_spec_bind_endpoint=None, decoupled_spec_connect_endpoints=None, decoupled_spec_rank=None, decoupled_spec_role='null', spec_trace_dir=None, speculative_ngram_min_bfs_breadth=1, speculative_ngram_max_bfs_breadth=10, speculative_ngram_match_type='BFS', speculative_ngram_max_trie_depth=18, speculative_ngram_capacity=10000000, speculative_ngram_external_corpus_path=None, speculative_ngram_external_sam_budget=0, speculative_ngram_external_corpus_max_tokens=10000000, ep_size=32, moe_a2a_backend='none', moe_runner_backend='marlin', flashinfer_mxfp4_moe_precision='default', deepep_mode='auto', fuseep_mode=2, deepep_dispatcher_output_dtype='auto', ep_num_redundant_experts=0, ep_dispatch_algorithm=None, init_expert_location='trivial', enable_eplb=False, eplb_algorithm='auto', eplb_rebalance_num_iterations=1000, eplb_rebalance_layers_per_chunk=None, eplb_min_rebalancing_utilization_threshold=1.0, expert_distribution_recorder_mode=None, expert_distribution_recorder_buffer_size=1000, enable_expert_distribution_metrics=False, deepep_config=None, moe_dense_tp_size=None, elastic_ep_backend=None, enable_elastic_expert_backup=False, mooncake_ib_device=None, enable_waterfill=False, ep_join_mode=None, ep_join_rank_offset=0, elastic_ep_initial_size=None, max_ep_size=None, elastic_ep_scale_timeout=600, elastic_ep_rejoin=False, disable_flashinfer_cutlass_moe_fp4_allgather=False, disable_shared_experts_fusion=False, enforce_shared_experts_fusion=False, max_mamba_cache_size=None, mamba_ssm_dtype=None, mamba_max_states_per_path=-1, enable_mamba_cache_stochastic_rounding=False, mamba_cache_philox_rounds=0, mamba_full_memory_ratio=0.21, mamba_radix_cache_strategy='extra_buffer_lazy', uses_mamba_radix_cache=False, mamba_track_interval=256, enable_int8_mamba_checkpoint=False, int8_mamba_ckpt_size=None, linear_attn_backend='triton', linear_attn_decode_backend=None, linear_attn_prefill_backend=None, linear_attn_verify_backend=None, enable_linear_replayssm=False, linear_replayssm_cache_len=16, enable_linear_replayssm_spec=False, enable_hierarchical_cache=False, hicache_ratio=2.0, hicache_size=0, hicache_write_policy='write_through', hicache_io_backend='kernel', hicache_mem_layout='page_first', hicache_storage_backend=None, hicache_storage_prefetch_policy='timeout', hicache_storage_backend_extra_config=None, enable_hisparse=False, hisparse_config=None, enable_broadcast_mm_inputs_process=False, enable_prefix_mm_cache=False, mm_enable_dp_encoder=False, mm_process_config={}, mm_processor_worker_num=0, mm_io_worker_num=0, limit_mm_data_per_request=None, enable_mm_global_cache=False, disable_fast_image_processor=False, mm_feature_transport='cpu', keep_mm_feature_on_device=False, enable_lora=None, enable_lora_overlap_loading=None, max_lora_rank=None, lora_target_modules=None, lora_paths=None, max_loaded_loras=None, max_loras_per_batch=8, lora_eviction_policy='lru', lora_backend='csgmv', max_lora_chunk_size=16, experts_shared_outer_loras=None, lora_use_virtual_experts=False, lora_strict_loading=False, lora_drain_wait_threshold=0.0, enable_two_batch_overlap=False, enable_single_batch_overlap=False, tbo_token_distribution_threshold=0.48, cpu_offload_gb=0, offload_group_size=-1, offload_num_in_group=1, offload_prefetch_step=1, offload_mode='cpu', enable_lmcache=False, lmcache_config_file=None, enable_flexkv=False, flexkv_config_file=None, kt_weight_path=None, kt_method='AMXINT4', kt_cpuinfer=None, kt_threadpool_count=2, kt_num_gpu_experts=None, kt_max_deferred_experts_per_token=None, dllm_algorithm=None, dllm_algorithm_config=None, dllm_fdfo=True, disaggregation_mode='decode', disaggregation_transfer_backend='mooncake', disaggregation_bootstrap_port=28800, disaggregation_ib_device='mlx5_0,mlx5_1,mlx5_2,mlx5_3', disaggregation_decode_enable_radix_cache=False, disaggregation_decode_enable_offload_kvcache=False, num_reserved_decode_tokens=512, disaggregation_decode_extra_slots=0, disaggregation_decode_polling_interval=1, optimistic_prefill_attempts=0, encoder_only=False, language_only=False, encoder_transfer_backend='zmq_to_tokenizer', encoder_urls=[], encoder_bootstrap_port=8997, encoder_register_urls=[], enable_adaptive_dispatch_to_encoder=False, enable_pdmux=False, pdmux_config_path=None, sm_group_num=8, custom_weight_loader=[], weight_loader_disable_mmap=False, weight_loader_prefetch_checkpoints=False, weight_loader_prefetch_num_threads=4, weight_loader_drop_cache_after_load=False, remote_instance_weight_loader_seed_instance_ip=None, remote_instance_weight_loader_seed_instance_service_port=None, remote_instance_weight_loader_send_weights_group_ports=None, remote_instance_weight_loader_backend='nccl', remote_instance_weight_loader_start_seed_via_transfer_engine=False, engine_info_bootstrap_port=6789, modelexpress_config=None, download_dir=None, model_checksum=None, delete_ckpt_after_loading=False, decrypted_config_file=None, decrypted_draft_config_file=None, checkpoint_engine_wait_weights_before_ready=False, enable_prefill_delayer=False, prefill_delayer_max_delay_passes=30, prefill_delayer_token_usage_low_watermark=None, prefill_delayer_forward_passes_buckets=None, prefill_delayer_wait_seconds_buckets=None, prefill_delayer_queue_min_ratio=None, prefill_delayer_max_delay_ms=None, min_free_slots_delay=None, enable_deterministic_inference=False, rl_on_policy_target=None, kv_canary='none', kv_canary_real_data='none', kv_canary_sweep_interval=0, enable_dynamic_batch_tokenizer=False, dynamic_batch_tokenizer_batch_size=32, dynamic_batch_tokenizer_batch_timeout=0.002, enable_tokenizer_batch_encode=False, disable_tokenizer_batch_decode=False, debug_tensor_dump_output_folder=None, debug_tensor_dump_layers=None, debug_tensor_dump_input_file=None, enable_memory_saver=False, enable_weights_cpu_backup=False, enable_draft_weights_cpu_backup=False, enable_custom_logit_processor=False, enable_return_hidden_states=False, enable_return_routed_experts=False, enable_return_indexer_topk=False, disable_outlines_disk_cache=False, enable_mis=False, weight_cache_mode='off', weight_cache_socket=None, weight_cache_timeout=1800, forward_hooks=None, msprobe_dump_config=None)
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:16 TP30 EP30] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:16 TP30 EP30] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:19 TP30 EP30] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:19.321749   241 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:19.321810   241 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:19.321820   241 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.8 port: 12001
+I0827 04:36:19.321856   241 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.8:16569
+I0827 04:36:19.321940   241 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:19.343153   241 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:19.347820   241 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:19.373265   241 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:19.380316   241 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:19.383023   241 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:19.383050   241 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:19.383101   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:19.390141   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:19.391583   241 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:08
+I0827 04:36:19.391615   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:19.417495   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:19.418977   241 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:08
+I0827 04:36:19.419003   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:19.445492   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:19.446861   241 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:08
+I0827 04:36:19.446884   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:19.473483   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:19.474817   241 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:08
+I0827 04:36:19.474922   241 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:19 TP30 EP30] Init torch distributed begin.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:20 TP27 EP27] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:20 TP27 EP27] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:21 TP25 EP25] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:21 TP25 EP25] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:21 TP26 EP26] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:21 TP26 EP26] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:21 TP28 EP28] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:21 TP28 EP28] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:21 TP24 EP24] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:21 TP24 EP24] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:21 TP29 EP29] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:21 TP29 EP29] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:21 TP31 EP31] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:21 TP31 EP31] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:22 TP27 EP27] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:22.734625   238 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:22.734687   238 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:22.734699   238 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.8 port: 12001
+I0827 04:36:22.734735   238 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.8:16566
+I0827 04:36:22.734848   238 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:22.760079   238 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:22.768187   238 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:22.777189   238 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:22.786154   238 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:22.789806   238 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:22.789830   238 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:22.789880   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:22.797986   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:22.799417   238 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:08
+I0827 04:36:22.799446   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:22.825555   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:22.826912   238 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:08
+I0827 04:36:22.826946   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:22.853536   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:22.854885   238 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:08
+I0827 04:36:22.854909   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:22.881553   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:22.882910   238 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:08
+I0827 04:36:22.883030   238 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:22 TP27 EP27] Init torch distributed begin.
+[2026-08-27 04:36:22 TP25 EP25] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:22 TP28 EP28] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:22 TP26 EP26] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:23 TP29 EP29] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:23 TP24 EP24] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:23.476856   236 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:23.476951   236 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:23.476967   236 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.8 port: 12001
+I0827 04:36:23.477013   236 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.8:15425
+I0827 04:36:23.477150   236 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:23.477173   239 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:23.477272   239 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:23.477288   239 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.8 port: 12001
+I0827 04:36:23.477344   239 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.8:15710
+I0827 04:36:23.477483   239 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:23.483197   236 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:23.483803   239 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:23.488940   236 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:23.489802   239 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:23.494866   236 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:23.495663   239 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:23.500973   236 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:23.501806   239 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:23.503927   236 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:23.503952   236 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:23.503998   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.504750   239 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:23.504770   239 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:23.504812   239 rdma_context.cpp:205] Using SIEVE endpoint store
+[2026-08-27 04:36:23 TP31 EP31] Multimodal data loading enabled with 16 worker threads (auto).
+I0827 04:36:23.510515   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:23.511322   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:23.511830   236 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:08
+I0827 04:36:23.511855   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.512616   239 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:08
+I0827 04:36:23.512640   239 rdma_context.cpp:205] Using SIEVE endpoint store
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:23.522899   237 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:23.523005   237 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:23.523020   237 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.8 port: 12001
+I0827 04:36:23.523065   237 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.8:16748
+I0827 04:36:23.523202   237 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:23.529469   237 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:23.535485   237 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:23.537460   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:23.537528   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:23.538903   236 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:08
+I0827 04:36:23.538913   239 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:08
+I0827 04:36:23.538928   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.538945   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.541347   237 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:23.546439   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:23.547672   237 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:23.547796   239 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:08
+I0827 04:36:23.547820   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.550041   237 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:23.550061   237 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:23.550102   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.555119   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:23.556490   239 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:08
+I0827 04:36:23.556598   239 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:23 TP28 EP28] Init torch distributed begin.
+I0827 04:36:23.557152   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:23.558562   237 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:08
+I0827 04:36:23.558589   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.565423   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:23.566766   236 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:08
+I0827 04:36:23.566788   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.589463   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:23.590811   237 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:08
+I0827 04:36:23.590835   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.593444   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:23.594800   236 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:08
+I0827 04:36:23.594898   236 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:23 TP25 EP25] Init torch distributed begin.
+I0827 04:36:23.598161   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:23.599563   237 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:08
+I0827 04:36:23.599589   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.625564   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:23.626922   237 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:08
+I0827 04:36:23.627038   237 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:23 TP26 EP26] Init torch distributed begin.
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:23.899124   240 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:23.899245   240 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:23.899261   240 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.8 port: 12001
+I0827 04:36:23.899313   240 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.8:15127
+I0827 04:36:23.899492   240 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:23.907753   235 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:23.907863   235 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:23.907878   235 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.8 port: 12001
+I0827 04:36:23.907927   235 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.8:15330
+I0827 04:36:23.908092   235 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:23.914011   235 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:23.919618   235 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:23.925108   235 topology.cpp:127] Device mlx5_2 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:23.925450   242 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:23.925558   242 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:23.925576   242 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.8 port: 12001
+I0827 04:36:23.925631   242 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.8:16235
+I0827 04:36:23.925693   240 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:23.925761   242 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:23.930419   242 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:23.933247   240 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:23.934639   242 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:23.939190   242 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:23.940605   240 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:23.943866   242 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:23.945531   242 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:23.945554   242 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:23.945595   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.947860   240 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:23.949263   235 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:23.951108   240 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:23.951148   240 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:23.951216   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.952417   235 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:23.952445   235 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:23.952502   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.958555   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:23.959836   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:23.960163   240 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:08
+I0827 04:36:23.960197   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.961452   235 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:08
+I0827 04:36:23.961483   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.969228   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:23.969322   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:23.970860   235 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:08
+I0827 04:36:23.970890   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.971050   242 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:08
+I0827 04:36:23.971084   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.979120   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:23.980706   242 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:08
+I0827 04:36:23.980736   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.989722   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:23.991313   240 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:08
+I0827 04:36:23.991346   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:24.001633   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:24.003211   235 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:08
+I0827 04:36:24.003237   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:24.005753   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:24.007426   242 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:08
+I0827 04:36:24.007457   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:24.021715   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:24.023293   240 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:08
+I0827 04:36:24.023322   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:24.034039   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:24.034071   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:24.035696   235 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:08
+I0827 04:36:24.035816   235 transfer_engine_impl.cpp:388] installTransport, type=rdma
+I0827 04:36:24.035794   242 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:08
+I0827 04:36:24.035987   242 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:24 TP24 EP24] Init torch distributed begin.
+[2026-08-27 04:36:24 TP31 EP31] Init torch distributed begin.
+I0827 04:36:24.049829   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:24.051388   240 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:08
+I0827 04:36:24.051532   240 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:24 TP29 EP29] Init torch distributed begin.
+[2026-08-27 04:36:48 TP31 EP31] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP30 EP30] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP28 EP28] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP29 EP29] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP27 EP27] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP26 EP26] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP25 EP25] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP24 EP24] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP31 EP31] Init torch distributed ends. elapsed=24.20 s, mem usage=0.40 GB
+[2026-08-27 04:36:48 TP30 EP30] Init torch distributed ends. elapsed=28.76 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP29 EP29] Init torch distributed ends. elapsed=24.18 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP28 EP28] Init torch distributed ends. elapsed=24.68 s, mem usage=0.40 GB
+[2026-08-27 04:36:48 TP27 EP27] Init torch distributed ends. elapsed=25.35 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP26 EP26] Init torch distributed ends. elapsed=24.61 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP25 EP25] Init torch distributed ends. elapsed=24.64 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP24 EP24] Init torch distributed ends. elapsed=24.20 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP28 EP28] Load weight begin. avail mem=82.19 GB
+[2026-08-27 04:36:48 TP27 EP27] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP30 EP30] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP24 EP24] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP29 EP29] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP26 EP26] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP25 EP25] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP28 EP28] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP30 EP30] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP27 EP27] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP29 EP29] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP24 EP24] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP26 EP26] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP25 EP25] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP31 EP31] Load weight begin. avail mem=82.19 GB
+[2026-08-27 04:36:48 TP31 EP31] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP27 EP27] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP28 EP28] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP24 EP24] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP27 EP27] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP30 EP30] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP29 EP29] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP26 EP26] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP25 EP25] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP28 EP28] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP24 EP24] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP29 EP29] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP31 EP31] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP30 EP30] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP26 EP26] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP25 EP25] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP31 EP31] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:50 TP29 EP29] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP28 EP28] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP30 EP30] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP26 EP26] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP25 EP25] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP31 EP31] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP27 EP27] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP24 EP24] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:38:54 TP26 EP26] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:54 TP27 EP27] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:54 TP24 EP24] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:54 TP26 EP26] Load weight end. elapsed=125.61 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:54 TP29 EP29] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:54 TP27 EP27] Load weight end. elapsed=125.61 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:54 TP24 EP24] Load weight end. elapsed=125.61 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:54 TP29 EP29] Load weight end. elapsed=125.61 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:54 TP25 EP25] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:54 TP25 EP25] Load weight end. elapsed=125.64 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:54 TP28 EP28] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:54 TP28 EP28] Load weight end. elapsed=125.69 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:54 TP30 EP30] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:54 TP30 EP30] Load weight end. elapsed=125.71 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:54 TP31 EP31] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:54 TP31 EP31] Load weight end. elapsed=125.77 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:39:01 TP26 EP26] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP26 EP26] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP30 EP30] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP30 EP30] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP28 EP28] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP28 EP28] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP27 EP27] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP27 EP27] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP25 EP25] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP25 EP25] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP31 EP31] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP31 EP31] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP24 EP24] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP24 EP24] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP29 EP29] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP29 EP29] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:02 TP29 EP29] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP28 EP28] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP26 EP26] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP27 EP27] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP30 EP30] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP25 EP25] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP31 EP31] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP24 EP24] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP29 EP29] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP30 EP30] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP28 EP28] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP26 EP26] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP27 EP27] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP25 EP25] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP24 EP24] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP31 EP31] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP29 EP29] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP29 EP29] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP28 EP28] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP28 EP28] Memory pool end. avail mem=9.18 GB
+[2026-08-27 04:39:02 TP27 EP27] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP24 EP24] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP26 EP26] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP27 EP27] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP31 EP31] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP24 EP24] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP26 EP26] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP31 EP31] Memory pool end. avail mem=9.18 GB
+[2026-08-27 04:39:02 TP30 EP30] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP25 EP25] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP30 EP30] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP25 EP25] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP29 EP29] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP29 EP29] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP24 EP24] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP24 EP24] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:39:02 TP28 EP28] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP28 EP28] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.17 GB
+[2026-08-27 04:39:02 TP25 EP25] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP31 EP31] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP25 EP25] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP31 EP31] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.17 GB
+[2026-08-27 04:39:02 TP26 EP26] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP26 EP26] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP27 EP27] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP27 EP27] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:39:02 TP30 EP30] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP30 EP30] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/d_starting_mc_master.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/d_starting_mc_master.log
new file mode 100644
index 0000000..8e503e7
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/d_starting_mc_master.log
@@ -0,0 +1,90 @@
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+I0827 04:28:16.835302    99 master.cpp:1366] Master service started on port 50051, max_threads=16, enable_metric_reporting=1, metrics_port=9003, default_kv_lease_ttl=10000, default_kv_soft_pin_ttl=1800000, allow_evict_soft_pinned_objects=1, eviction_ratio=0.05, eviction_high_watermark_ratio=0.9, enable_ha=0, enable_offload=0, enable_kv_events=0, kv_events_bind_endpoint=, kv_events_backend_id=, offload_on_evict=0, offload_force_evict=0, offloading_queue_limit=50000, offload_cap_ratio=0.5, ha_backend_type=etcd, ha_backend_connstring=, etcd_endpoints=, client_ttl=10, rpc_thread_num=16, rpc_port=50051, rpc_address=0.0.0.0, rpc_interface=, rpc_conn_timeout_seconds=0, rpc_enable_tcp_no_delay=1, rpc protocol=tcp, cluster_id=mooncake_cluster, root_fs_dir=, global_file_segment_size=9223372036854775807, memory_allocator=offset, enable_http_metadata_server=0, http_metadata_server_port=8080, http_metadata_server_host=0.0.0.0, enable_metadata_cleanup_on_timeout=0, put_start_discard_timeout_sec=30, put_start_release_timeout_sec=600, max_total_finished_tasks=10000, max_total_pending_tasks=10000, max_total_processing_tasks=10000, pending_task_timeout_sec=300, processing_task_timeout_sec=300, enable_snapshot=0, enable_snapshot_restore=0, snapshot_interval_seconds=600, snapshot_backup_dir=, snapshot_object_store_type=, snapshot_catalog_store_type=, snapshot_retention_count=2, max_retry_attempts=10, enable_cxl=0, cxl_path=/dev/dax0.0, cxl_size=8589934592
+I0827 04:28:16.866606   108 master_service.cpp:1520] Task cleanup thread started
+I0827 04:28:16.867491    99 master_admin_service.cpp:311] Master admin server started on port 9003
+I0827 04:28:16.869802   111 master_admin_service.cpp:302] Master Admin Metrics: role=standby, state=starting, service_ready=false, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:28:26.870203   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:28:36.870568   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:28:46.870873   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:28:56.871232   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:29:06.871615   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:29:16.871969   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:29:26.872316   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:29:36.872687   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:29:46.873068   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:29:56.873467   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:30:06.873914   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:30:16.874214   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:30:26.874567   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:30:36.874953   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:30:46.875332   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:30:56.875741   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:31:06.876119   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:31:16.876497   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:31:26.876866   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:31:36.877233   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:31:46.877605   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:31:56.877948   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:32:06.878437   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:32:16.879210   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:32:26.879711   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:32:36.880080   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:32:46.880565   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:32:56.881004   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:33:06.881398   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:33:16.881786   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:33:26.882170   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:33:36.882550   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:33:46.882939   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:33:56.883315   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:34:06.883746   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:34:16.884127   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:34:26.884568   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:34:36.884953   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:34:46.885334   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:34:56.885706   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:35:06.886085   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:35:16.886464   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:35:26.886835   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:35:36.887218   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:35:46.887596   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:35:56.887964   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:36:06.888356   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:36:16.888742   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:36:26.889113   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:36:36.889495   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:36:46.889882   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:36:56.890261   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:37:06.890656   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:37:16.891040   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:37:26.891430   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:37:36.891836   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:37:46.892215   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:37:56.892597   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:38:06.892982   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:38:16.893360   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:38:26.893747   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:38:36.894131   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:38:46.894515   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:38:56.894920   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:39:06.895303   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:39:16.895701   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:39:26.896085   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:39:36.896471   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:39:46.896852   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:39:56.897240   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:40:06.897629   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:40:16.898005   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:40:26.898384   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:40:36.898761   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:40:46.899144   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:40:56.899530   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:41:06.899937   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:41:16.900316   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:41:26.900695   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:41:36.901089   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:41:46.901479   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:41:56.901856   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:42:06.902242   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:42:16.902666   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:42:26.903064   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/d_starting_p_node0.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/d_starting_p_node0.log
new file mode 100644
index 0000000..85cfdec
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/d_starting_p_node0.log
@@ -0,0 +1,941 @@
+/usr/local/lib/python3.12/dist-packages/torchao/quantization/quant_api.py:1731: SyntaxWarning: invalid escape sequence '\.'
+  """Configuration class for applying different quantization configs to modules or parameters based on their fully qualified names (FQNs).
+/sgl-workspace/sglang/python/sglang/launch_server.py:56: UserWarning: 'python -m sglang.launch_server' is still supported, but 'sglang serve' is the recommended entrypoint.
+  Example: sglang serve --model-path <model> [options]
+  warnings.warn(
+[2026-08-27 04:28:54] Breakable CUDA graph is incompatible with MLA attention; disabling prefill CUDA graph.
+[2026-08-27 04:28:55] Attention backend not specified. Use triton backend by default.
+[2026-08-27 04:28:55] Pipeline parallelism is incompatible with overlap schedule.
+[2026-08-27 04:28:58] server_args=ServerArgs(model_path='/data/hf_models/Kimi-K3', tokenizer_path='/data/hf_models/Kimi-K3', tokenizer_mode='auto', tokenizer_backend='huggingface', tokenizer_worker_num=1, detokenizer_worker_num=1, skip_tokenizer_init=False, load_format='auto', model_loader_extra_config='{}', trust_remote_code=True, context_length=None, is_embedding=False, enable_multimodal=None, revision=None, model_impl='auto', model_config_parser='auto', json_model_override_args='{}', dtype='auto', quantization=None, quantization_param_path=None, kv_cache_dtype='auto', enable_fp32_lm_head=False, modelopt_quant=None, modelopt_checkpoint_restore_path=None, modelopt_checkpoint_save_path=None, modelopt_export_path=None, quantize_and_serve=False, rl_quant_profile=None, enable_tf32_matmul=False, mem_fraction_static=0.88, max_running_requests=None, max_queued_requests=None, max_total_tokens=None, chunked_prefill_size=8192, enable_dynamic_chunking=False, max_prefill_tokens=16384, prefill_max_requests=None, schedule_policy='fcfs', enable_priority_scheduling=False, disable_priority_preemption=False, default_priority_value=None, abort_on_priority_when_disabled=False, schedule_low_priority_values_first=False, priority_scheduling_preemption_threshold=10, retraction_policy='length', schedule_conservativeness=1.0, page_size=64, swa_full_tokens_ratio=0.8, disable_hybrid_swa_memory=False, radix_eviction_policy='lru', prefill_only_disable_kv_cache=False, disable_radix_cache=True, enable_page_major_kv_layout=False, enable_unified_memory=False, disable_chunked_prefix_cache=False, disable_overlap_schedule=True, num_continuous_decode_steps=1, scheduler_recv_interval=1, enable_mixed_chunk=False, nccl_port=None, dist_timeout=3600, dist_init_addr='174.1.60.1:20000', nnodes=4, node_rank=0, tp_size=4, dcp_size=1, pp_size=8, pp_max_micro_batch_size=None, pp_async_batch_depth=0, dp_size=1, load_balance_method='follow_bootstrap_room', attn_cp_size=1, moe_dp_size=1, dcp_comm_backend='ag_rs', dcp_replicate_q_proj=None, dwdp_size=1, enable_prefill_cp=False, cp_strategy=None, enable_dsa_cache_layer_split=False, enable_dsa_prefill_context_parallel=False, dsa_prefill_cp_mode='round-robin-split', enable_prefill_context_parallel=False, prefill_cp_mode='in-seq-split', enable_cp_decode_attn_tp=False, enable_dp_attention=False, enable_dp_attention_local_control_broadcast=False, enable_dp_lm_head=False, enable_attn_tp_input_scattered=False, disable_attn_tp_gather=False, enable_p2p_check=False, device='cuda', base_gpu_id=0, gpu_id_step=1, random_seed=984607766, watchdog_timeout=300, soft_watchdog_timeout=None, sleep_on_idle=False, use_ray=False, custom_sigquit_handler=None, numa_node=None, gc_threshold=None, host='0.0.0.0', port=30000, fastapi_root_path='', smg_grpc_mode=False, grpc_mode=False, grpc_port=None, sidecar=None, sidecar_args=None, skip_server_warmup=False, warmups=None, enable_http2=False, ssl_keyfile=None, ssl_certfile=None, ssl_ca_certs=None, ssl_keyfile_password=None, enable_ssl_refresh=False, api_key=None, admin_api_key=None, served_model_name='kimi-k3', weight_version='default', chat_template=None, hf_chat_template_name=None, completion_template=None, file_storage_path='sglang_storage', enable_cache_report=False, reasoning_parser=None, default_chat_template_kwargs=None, strip_thinking_cache=False, enable_strict_thinking=False, tool_call_parser=None, tool_server=None, sampling_defaults='model', asr_max_buffer_seconds=60, asr_max_concurrent_sessions=32, preferred_sampling_params=None, allow_auto_truncate=False, stream_interval=1, batch_notify_size=16, stream_response_default_include_usage=False, incremental_streaming_output=False, enable_streaming_session=False, enable_session_radix_cache=False, log_level='info', log_level_http=None, log_requests=False, log_requests_level=2, log_requests_format='text', log_requests_target=None, uvicorn_access_log_exclude_prefixes=[], crash_dump_folder=None, show_time_cost=False, enable_metrics=False, smg_http_sidecar_port=None, enable_mfu_metrics=False, enable_metrics_for_all_schedulers=False, load_snapshot_publish_interval=15, tokenizer_metrics_custom_labels_header='x-custom-labels', tokenizer_metrics_allowed_custom_labels=None, extra_metric_labels=None, bucket_time_to_first_token=None, bucket_inter_token_latency=None, bucket_e2e_request_latency=None, prompt_tokens_buckets=None, generation_tokens_buckets=None, gc_warning_threshold_secs=0.0, decode_log_interval=40, enable_request_time_stats_logging=False, kv_events_config=None, enable_forward_pass_metrics=False, forward_pass_metrics_worker_id='', forward_pass_metrics_ipc_name=None, enable_trace=False, trace_modules='request', otlp_traces_endpoint='localhost:4317', export_metrics_to_file=False, export_metrics_to_file_dir=None, stat_loggers=None, constrained_json_whitespace_pattern=None, constrained_json_disable_any_whitespace=False, attention_backend='triton', decode_attention_backend=None, prefill_attention_backend=None, sampling_backend='flashinfer', grammar_backend='xgrammar', radix_cache_backend=None, mm_attention_backend=None, fp8_gemm_runner_backend='auto', fp4_gemm_runner_backend='auto', bf16_gemm_backend='auto', dsa_prefill_backend=None, dsa_decode_backend=None, dsa_paged_mqa_logits_backend='auto', dsa_topk_backend='sgl-kernel', disable_flashinfer_autotune=False, flashinfer_autotune_skip_ops=None, mamba_backend='triton', cuda_graph_config=CudaGraphConfig(decode=PhaseConfig(backend='disabled', max_bs=16, bs=[1, 2, 4, 8, 12, 16], tc_compiler='eager', full_prefill_max_req=None), prefill=PhaseConfig(backend='disabled', max_bs=2048, bs=[4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], tc_compiler='eager', full_prefill_max_req=None)), cuda_graph_backend_decode=None, cuda_graph_backend_prefill=None, cuda_graph_max_bs_decode=16, cuda_graph_max_bs_prefill=None, cuda_graph_bs_decode=None, cuda_graph_bs_prefill=None, cuda_graph_tc_compiler=None, disable_prefill_cuda_graph=False, disable_decode_cuda_graph=False, disable_cuda_graph=False, disable_cuda_graph_padding=False, enable_profile_cuda_graph=False, enable_cudagraph_gc=False, debug_cuda_graph=False, enable_layerwise_nvtx_marker=False, enable_nccl_nvls=False, enable_symm_mem=False, triton_attention_reduce_in_fp32=False, triton_attention_num_kv_splits=8, triton_attention_split_tile_size=None, flashinfer_mla_disable_ragged=False, enable_fused_qk_norm_rope=False, enable_precise_embedding_interpolation=False, enable_fused_moe_sum_all_reduce=False, enable_deepseek_v4_fp4_indexer=False, disable_custom_all_reduce=False, enable_mscclpp=False, enable_torch_symm_mem=False, enable_scattered_sconv=False, pre_warm_nccl=False, enable_quant_communications=False, enable_flashinfer_allreduce_fusion=False, enforce_disable_flashinfer_allreduce_fusion=False, flashinfer_allreduce_fusion_backend=None, enable_aiter_allreduce_fusion=False, enable_torch_compile=False, enable_torch_compile_debug_mode=False, torch_compile_max_bs=32, torchao_config='', speculative_algorithm=None, speculative_draft_model_path=None, speculative_draft_model_revision=None, speculative_draft_load_format=None, speculative_num_steps=None, speculative_eagle_topk=None, speculative_num_draft_tokens=None, speculative_dflash_block_size=None, speculative_dspark_block_size=None, speculative_dspark_sps_table_path=None, speculative_dspark_confidence_sts_path=None, speculative_dspark_align_verify_tokens_to_graph_tier=False, speculative_accept_threshold_single=1.0, speculative_accept_threshold_acc=1.0, speculative_use_rejection_sampling=False, speculative_token_map=None, speculative_attention_mode='prefill', speculative_draft_attention_backend=None, speculative_draft_window_size=None, speculative_moe_runner_backend='flashinfer_mxfp4', speculative_moe_a2a_backend=None, speculative_draft_model_quantization=None, speculative_skip_dp_mlp_sync=False, enable_multi_layer_eagle=False, speculative_adaptive=False, speculative_adaptive_config=None, decoupled_spec_bind_endpoint=None, decoupled_spec_connect_endpoints=None, decoupled_spec_rank=None, decoupled_spec_role='null', spec_trace_dir=None, speculative_ngram_min_bfs_breadth=1, speculative_ngram_max_bfs_breadth=10, speculative_ngram_match_type='BFS', speculative_ngram_max_trie_depth=18, speculative_ngram_capacity=10000000, speculative_ngram_external_corpus_path=None, speculative_ngram_external_sam_budget=0, speculative_ngram_external_corpus_max_tokens=10000000, ep_size=4, moe_a2a_backend='none', moe_runner_backend='flashinfer_mxfp4', flashinfer_mxfp4_moe_precision='default', deepep_mode='auto', fuseep_mode=2, deepep_dispatcher_output_dtype='auto', ep_num_redundant_experts=0, ep_dispatch_algorithm=None, init_expert_location='trivial', enable_eplb=False, eplb_algorithm='auto', eplb_rebalance_num_iterations=1000, eplb_rebalance_layers_per_chunk=None, eplb_min_rebalancing_utilization_threshold=1.0, expert_distribution_recorder_mode=None, expert_distribution_recorder_buffer_size=1000, enable_expert_distribution_metrics=False, deepep_config=None, moe_dense_tp_size=None, elastic_ep_backend=None, enable_elastic_expert_backup=False, mooncake_ib_device=None, enable_waterfill=False, ep_join_mode=None, ep_join_rank_offset=0, elastic_ep_initial_size=None, max_ep_size=None, elastic_ep_scale_timeout=600, elastic_ep_rejoin=False, disable_flashinfer_cutlass_moe_fp4_allgather=False, disable_shared_experts_fusion=False, enforce_shared_experts_fusion=False, max_mamba_cache_size=None, mamba_ssm_dtype=None, mamba_max_states_per_path=-1, enable_mamba_cache_stochastic_rounding=False, mamba_cache_philox_rounds=0, mamba_full_memory_ratio=0.36, mamba_radix_cache_strategy='extra_buffer_lazy', uses_mamba_radix_cache=False, mamba_track_interval=256, enable_int8_mamba_checkpoint=False, int8_mamba_ckpt_size=None, linear_attn_backend='triton', linear_attn_decode_backend=None, linear_attn_prefill_backend=None, linear_attn_verify_backend=None, enable_linear_replayssm=False, linear_replayssm_cache_len=16, enable_linear_replayssm_spec=False, enable_hierarchical_cache=False, hicache_ratio=2.0, hicache_size=0, hicache_write_policy='write_through', hicache_io_backend='kernel', hicache_mem_layout='page_first', hicache_storage_backend=None, hicache_storage_prefetch_policy='timeout', hicache_storage_backend_extra_config=None, enable_hisparse=False, hisparse_config=None, enable_broadcast_mm_inputs_process=False, enable_prefix_mm_cache=False, mm_enable_dp_encoder=False, mm_process_config={}, mm_processor_worker_num=0, mm_io_worker_num=0, limit_mm_data_per_request=None, enable_mm_global_cache=False, disable_fast_image_processor=False, mm_feature_transport='cpu', keep_mm_feature_on_device=False, enable_lora=None, enable_lora_overlap_loading=None, max_lora_rank=None, lora_target_modules=None, lora_paths=None, max_loaded_loras=None, max_loras_per_batch=8, lora_eviction_policy='lru', lora_backend='csgmv', max_lora_chunk_size=16, experts_shared_outer_loras=None, lora_use_virtual_experts=False, lora_strict_loading=False, lora_drain_wait_threshold=0.0, enable_two_batch_overlap=False, enable_single_batch_overlap=False, tbo_token_distribution_threshold=0.48, cpu_offload_gb=0, offload_group_size=-1, offload_num_in_group=1, offload_prefetch_step=1, offload_mode='cpu', enable_lmcache=False, lmcache_config_file=None, enable_flexkv=False, flexkv_config_file=None, kt_weight_path=None, kt_method='AMXINT4', kt_cpuinfer=None, kt_threadpool_count=2, kt_num_gpu_experts=None, kt_max_deferred_experts_per_token=None, dllm_algorithm=None, dllm_algorithm_config=None, dllm_fdfo=True, disaggregation_mode='prefill', disaggregation_transfer_backend='mooncake', disaggregation_bootstrap_port=28800, disaggregation_ib_device='mlx5_0,mlx5_1,mlx5_2,mlx5_3', disaggregation_decode_enable_radix_cache=False, disaggregation_decode_enable_offload_kvcache=False, num_reserved_decode_tokens=512, disaggregation_decode_extra_slots=None, disaggregation_decode_polling_interval=1, optimistic_prefill_attempts=0, encoder_only=False, language_only=False, encoder_transfer_backend='zmq_to_tokenizer', encoder_urls=[], encoder_bootstrap_port=8997, encoder_register_urls=[], enable_adaptive_dispatch_to_encoder=False, enable_pdmux=False, pdmux_config_path=None, sm_group_num=8, custom_weight_loader=[], weight_loader_disable_mmap=False, weight_loader_prefetch_checkpoints=False, weight_loader_prefetch_num_threads=4, weight_loader_drop_cache_after_load=False, remote_instance_weight_loader_seed_instance_ip=None, remote_instance_weight_loader_seed_instance_service_port=None, remote_instance_weight_loader_send_weights_group_ports=None, remote_instance_weight_loader_backend='nccl', remote_instance_weight_loader_start_seed_via_transfer_engine=False, engine_info_bootstrap_port=6789, modelexpress_config=None, download_dir=None, model_checksum=None, delete_ckpt_after_loading=False, decrypted_config_file=None, decrypted_draft_config_file=None, checkpoint_engine_wait_weights_before_ready=False, enable_prefill_delayer=False, prefill_delayer_max_delay_passes=30, prefill_delayer_token_usage_low_watermark=None, prefill_delayer_forward_passes_buckets=None, prefill_delayer_wait_seconds_buckets=None, prefill_delayer_queue_min_ratio=None, prefill_delayer_max_delay_ms=None, min_free_slots_delay=None, enable_deterministic_inference=False, rl_on_policy_target=None, kv_canary='none', kv_canary_real_data='none', kv_canary_sweep_interval=0, enable_dynamic_batch_tokenizer=False, dynamic_batch_tokenizer_batch_size=32, dynamic_batch_tokenizer_batch_timeout=0.002, enable_tokenizer_batch_encode=False, disable_tokenizer_batch_decode=False, debug_tensor_dump_output_folder=None, debug_tensor_dump_layers=None, debug_tensor_dump_input_file=None, enable_memory_saver=False, enable_weights_cpu_backup=False, enable_draft_weights_cpu_backup=False, enable_custom_logit_processor=False, enable_return_hidden_states=False, enable_return_routed_experts=False, enable_return_indexer_topk=False, disable_outlines_disk_cache=False, enable_mis=False, weight_cache_mode='off', weight_cache_socket=None, weight_cache_timeout=1800, forward_hooks=None, msprobe_dump_config=None)
+[2026-08-27 04:28:58] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:58] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:58] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:58] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:58] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:58] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:58] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:58] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:01] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:01] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:01] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:03] Multimodal processor concurrency enabled with 2 isolated worker threads (auto).
+[2026-08-27 04:29:03] CommonKVBootstrapServer started successfully on 0.0.0.0:28800
+[2026-08-27 04:29:03] No HuggingFace chat template found
+[2026-08-27 04:29:03] No chat template found, defaulting to 'string' content format
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:15 PP0 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:15 PP0 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:16] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:16] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:16 PP1 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:16 PP1 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+/sgl-workspace/sglang/python/sglang/srt/utils/hf_transformers/tokenizer.py:423: UserWarning: Using a slow tokenizer. This might cause a significant slowdown. Consider using a fast tokenizer instead.
+  warnings.warn(
+[2026-08-27 04:29:16] Applying special tokens cache patch for Kimi tokenizer: <class 'transformers_modules.Kimi_hyphen_K3.caf40b2773f9029e.tokenization_kimi.TikTokenTokenizer'>
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:16 PP1 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:16 PP1 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:16 PP0 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:16 PP0 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:16 PP1 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:16 PP1 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:17 PP0 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:17 PP0 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:17 PP1 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:17 PP1 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:17 PP0 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:17 PP0 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:17 PP0 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:17.587481   237 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:17.587555   237 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:17.587566   237 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.1 port: 12001
+I0827 04:29:17.587602   237 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.1:16830
+I0827 04:29:17.587723   237 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:17.613691   237 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:17.620473   237 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:17.627439   237 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:17.653980   237 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:17.656663   237 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:17.656692   237 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:17.656746   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:17.663833   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:17.665359   237 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:01
+I0827 04:29:17.665387   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:17.690173   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:17.691654   237 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:01
+I0827 04:29:17.691689   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:17.718215   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:17.719650   237 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:01
+I0827 04:29:17.719686   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:17.746297   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:17.747779   237 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:01
+I0827 04:29:17.747891   237 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:17 PP0 TP1 EP1] Init torch distributed begin.
+[2026-08-27 04:29:17 PP1 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:18.219898   241 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:18.219964   241 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:18.219973   241 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.1 port: 12001
+I0827 04:29:18.220007   241 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.1:15425
+I0827 04:29:18.220109   241 transfer_engine_impl.cpp:258] Auto-discovering topology...
+[2026-08-27 04:29:18 PP1 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+I0827 04:29:18.244532   241 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:18.252967   241 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:18.261787   241 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:18.289944   241 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:18.293610   241 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:18.293633   241 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:18.293681   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.301677   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:18.303126   241 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:01
+I0827 04:29:18.303155   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.330181   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:18.331590   241 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:01
+I0827 04:29:18.331614   241 rdma_context.cpp:205] Using SIEVE endpoint store
+[2026-08-27 04:29:18 PP0 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+I0827 04:29:18.358213   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:18.359581   241 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:01
+I0827 04:29:18.359604   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.386204   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:18.387562   241 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:01
+I0827 04:29:18.387681   241 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:18 PP1 TP1 EP1] Init torch distributed begin.
+[2026-08-27 04:29:18 PP1 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:18.588883   242 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:18.588970   242 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:18.588984   242 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.1 port: 12001
+I0827 04:29:18.589022   242 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.1:15650
+I0827 04:29:18.589143   242 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:18.594419   242 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:18.599439   242 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:18.604553   242 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:18.609825   242 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:18.612036   242 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:18.612056   242 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:18.612097   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.618021   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:18.619308   242 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:01
+I0827 04:29:18.619333   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.646080   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:18.647465   242 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:01
+I0827 04:29:18.647495   242 rdma_context.cpp:205] Using SIEVE endpoint store
+[2026-08-27 04:29:18 PP0 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+I0827 04:29:18.674199   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:18.675598   242 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:01
+I0827 04:29:18.675621   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.682866   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:18.684276   242 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:01
+I0827 04:29:18.684391   242 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:18 PP1 TP2 EP2] Init torch distributed begin.
+[2026-08-27 04:29:18 PP1 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:18.695660   238 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:18.695755   238 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:18.695767   238 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.1 port: 12001
+I0827 04:29:18.695806   238 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.1:15123
+I0827 04:29:18.695909   238 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:18.701093   238 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:18.706161   238 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:18.711323   238 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:18.737910   238 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:18.740556   238 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:18.740602   238 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:18.740651   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.747717   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:18.749193   238 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:01
+I0827 04:29:18.749220   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.774145   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:18.775534   238 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:01
+I0827 04:29:18.775558   238 rdma_context.cpp:205] Using SIEVE endpoint store
+[2026-08-27 04:29:18 PP0 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+I0827 04:29:18.802196   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:18.803623   238 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:01
+I0827 04:29:18.803649   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.810976   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:18.812341   238 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:01
+I0827 04:29:18.812443   238 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:18 PP0 TP2 EP2] Init torch distributed begin.
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:19.084213   243 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:19.084332   243 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:19.084348   243 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.1 port: 12001
+I0827 04:29:19.084400   243 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.1:15411
+I0827 04:29:19.084602   243 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:19.114387   243 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:19.121773   243 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:19.146405   243 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:19.153759   243 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:19.156515   243 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:19.156548   243 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:19.156603   243 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.163982   243 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:19.165694   243 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:01
+I0827 04:29:19.165726   243 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.173307   243 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:19.174904   243 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:01
+I0827 04:29:19.174932   243 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.206454   243 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:19.208142   243 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:01
+I0827 04:29:19.208168   243 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.215909   243 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:19.217583   243 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:01
+I0827 04:29:19.217706   243 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:19 PP1 TP3 EP3] Init torch distributed begin.
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:19.239588   236 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:19.239699   236 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:19.239715   236 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.1 port: 12001
+I0827 04:29:19.239763   236 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.1:15605
+I0827 04:29:19.239933   236 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:19.240546   240 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:19.240644   240 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:19.240658   240 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.1 port: 12001
+I0827 04:29:19.240702   240 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.1:15487
+I0827 04:29:19.240862   240 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:19.242301   239 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:19.242408   239 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:19.242422   239 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.1 port: 12001
+I0827 04:29:19.242470   239 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.1:15552
+I0827 04:29:19.242636   239 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:19.246201   240 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:19.248451   239 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:19.251204   240 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:19.254226   239 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:19.256254   240 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:19.259852   239 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:19.261372   240 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:19.263411   240 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:19.263432   240 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:19.263473   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.265838   239 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:19.268059   239 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:19.268082   239 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:19.268126   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.269106   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:19.270298   236 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:19.270401   240 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:01
+I0827 04:29:19.270423   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.274480   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:19.275991   239 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:01
+I0827 04:29:19.276019   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.276877   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:19.278193   240 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:01
+I0827 04:29:19.278220   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.297820   236 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:19.302485   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:19.302623   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:19.303958   239 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:01
+I0827 04:29:19.303983   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.304121   240 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:01
+I0827 04:29:19.304147   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.311621   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:19.313341   240 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:01
+I0827 04:29:19.313478   240 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:19 PP1 TP0 EP0] Init torch distributed begin.
+I0827 04:29:19.322126   236 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:19.330281   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:19.331759   239 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:01
+I0827 04:29:19.331784   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.339274   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:19.340766   239 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:01
+I0827 04:29:19.340884   239 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:19 PP0 TP3 EP3] Init torch distributed begin.
+I0827 04:29:19.345958   236 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:19.348632   236 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:19.348659   236 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:19.348712   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.355962   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:19.417273   236 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:01
+I0827 04:29:19.417309   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.424894   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:19.426437   236 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:01
+I0827 04:29:19.426491   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.434231   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:19.435868   236 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:01
+I0827 04:29:19.435896   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.443578   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:19.445148   236 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:01
+I0827 04:29:19.445264   236 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:19 PP0 TP0 EP0] Init torch distributed begin.
+[2026-08-27 04:29:26 PP0 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:26 PP1 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:27 PP1 TP2 EP2] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP1 TP0 EP0] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP1 TP3 EP3] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP1 TP1 EP1] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP0 TP3 EP3] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP0 TP0 EP0] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP0 TP1 EP1] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP0 TP2 EP2] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP0 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:27 PP0 TP3 EP3] sglang is using nccl==2.28.9
+[2026-08-27 04:29:27 PP0 TP1 EP1] sglang is using nccl==2.28.9
+[2026-08-27 04:29:27 PP0 TP2 EP2] sglang is using nccl==2.28.9
+[2026-08-27 04:29:27 PP0 TP0 EP0] Init torch distributed ends. elapsed=8.22 s, mem usage=0.46 GB
+[2026-08-27 04:29:27 PP1 TP3 EP3] Init torch distributed ends. elapsed=8.45 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP1 TP1 EP1] Init torch distributed ends. elapsed=9.28 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP1 TP2 EP2] Init torch distributed ends. elapsed=8.98 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP1 TP0 EP0] Init torch distributed ends. elapsed=8.35 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP0 TP2 EP2] Init torch distributed ends. elapsed=8.85 s, mem usage=0.46 GB
+[2026-08-27 04:29:27 PP0 TP3 EP3] Init torch distributed ends. elapsed=8.32 s, mem usage=0.46 GB
+[2026-08-27 04:29:27 PP0 TP1 EP1] Init torch distributed ends. elapsed=9.92 s, mem usage=0.46 GB
+[2026-08-27 04:29:27 PP1 TP0 EP0] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP1 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP0 TP0 EP0] Load weight begin. avail mem=82.12 GB
+[2026-08-27 04:29:27 PP1 TP3 EP3] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP1 TP2 EP2] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP0 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP1 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP1 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP1 TP1 EP1] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP0 TP2 EP2] Load weight begin. avail mem=82.12 GB
+[2026-08-27 04:29:27 PP1 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP0 TP3 EP3] Load weight begin. avail mem=82.12 GB
+[2026-08-27 04:29:27 PP0 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP0 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP1 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:27 PP1 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:27 PP0 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:27 PP1 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:27 PP1 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:27 PP1 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:27 PP0 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:27 PP1 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:27 PP1 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:27 PP1 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:27 PP0 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:27 PP0 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:27 PP0 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:27 PP0 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:27 PP0 TP1 EP1] Load weight begin. avail mem=82.12 GB
+[2026-08-27 04:29:27 PP0 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP0 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP0 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+
Multi-thread loading shards:   0% Completed | 0/96 [00:00<?, ?it/s]
Multi-thread loading shards:   1% Completed | 1/96 [00:00<01:01,  1.55it/s]
Multi-thread loading shards:   2% Completed | 2/96 [00:09<08:49,  5.63s/it]
Multi-thread loading shards:   3% Completed | 3/96 [00:12<06:35,  4.25s/it]
Multi-thread loading shards:   4% Completed | 4/96 [00:14<05:26,  3.55s/it]
Multi-thread loading shards:   5% Completed | 5/96 [00:17<04:43,  3.11s/it]
Multi-thread loading shards:   6% Completed | 6/96 [00:19<04:17,  2.86s/it]
Multi-thread loading shards:   7% Completed | 7/96 [00:21<04:00,  2.71s/it]
Multi-thread loading shards:   8% Completed | 8/96 [00:24<03:49,  2.61s/it]
Multi-thread loading shards:   9% Completed | 9/96 [00:26<03:44,  2.58s/it]
Multi-thread loading shards:  10% Completed | 10/96 [00:29<03:40,  2.56s/it]
Multi-thread loading shards:  11% Completed | 11/96 [00:31<03:36,  2.54s/it]
Multi-thread loading shards:  12% Completed | 12/96 [00:32<02:43,  1.95s/it]
Multi-thread loading shards:  14% Completed | 13/96 [00:32<02:05,  1.52s/it]
Multi-thread loading shards:  15% Completed | 14/96 [00:33<01:33,  1.14s/it]
Multi-thread loading shards:  16% Completed | 15/96 [00:33<01:21,  1.01s/it]
Multi-thread loading shards:  17% Completed | 16/96 [00:34<01:19,  1.00it/s]
Multi-thread loading shards:  18% Completed | 17/96 [00:35<01:02,  1.26it/s]
Multi-thread loading shards:  19% Completed | 18/96 [00:35<00:51,  1.53it/s]
Multi-thread loading shards:  20% Completed | 19/96 [00:35<00:42,  1.80it/s]
Multi-thread loading shards:  21% Completed | 20/96 [00:36<00:37,  2.01it/s]
Multi-thread loading shards:  22% Completed | 21/96 [00:36<00:33,  2.23it/s]
Multi-thread loading shards:  23% Completed | 22/96 [00:36<00:30,  2.43it/s]
Multi-thread loading shards:  24% Completed | 23/96 [00:37<00:27,  2.68it/s]
Multi-thread loading shards:  25% Completed | 24/96 [00:37<00:25,  2.80it/s]
Multi-thread loading shards:  26% Completed | 25/96 [00:37<00:24,  2.89it/s]
Multi-thread loading shards:  27% Completed | 26/96 [00:38<00:23,  2.99it/s]
Multi-thread loading shards:  28% Completed | 27/96 [00:38<00:22,  3.01it/s]
Multi-thread loading shards:  29% Completed | 28/96 [00:38<00:22,  2.99it/s]
Multi-thread loading shards:  30% Completed | 29/96 [00:39<00:22,  2.98it/s]
Multi-thread loading shards:  31% Completed | 30/96 [00:39<00:22,  2.98it/s]
Multi-thread loading shards:  32% Completed | 31/96 [00:39<00:21,  3.01it/s]
Multi-thread loading shards:  33% Completed | 32/96 [00:40<00:21,  3.00it/s]
Multi-thread loading shards:  34% Completed | 33/96 [00:41<00:35,  1.79it/s]
Multi-thread loading shards:  35% Completed | 34/96 [00:41<00:30,  2.02it/s]
Multi-thread loading shards:  36% Completed | 35/96 [00:41<00:27,  2.23it/s]
Multi-thread loading shards:  38% Completed | 36/96 [00:42<00:24,  2.41it/s]
Multi-thread loading shards:  39% Completed | 37/96 [00:42<00:22,  2.57it/s]
Multi-thread loading shards:  40% Completed | 38/96 [00:42<00:21,  2.64it/s]
Multi-thread loading shards:  41% Completed | 39/96 [00:43<00:20,  2.73it/s]
Multi-thread loading shards:  42% Completed | 40/96 [00:43<00:20,  2.80it/s]
Multi-thread loading shards:  43% Completed | 41/96 [00:43<00:19,  2.79it/s]
Multi-thread loading shards:  44% Completed | 42/96 [00:44<00:19,  2.80it/s]
Multi-thread loading shards:  45% Completed | 43/96 [00:44<00:22,  2.39it/s]
Multi-thread loading shards:  46% Completed | 44/96 [00:45<00:20,  2.51it/s]
Multi-thread loading shards:  47% Completed | 45/96 [00:45<00:19,  2.64it/s]
Multi-thread loading shards:  48% Completed | 46/96 [00:45<00:18,  2.67it/s]
Multi-thread loading shards:  49% Completed | 47/96 [00:46<00:18,  2.61it/s]
Multi-thread loading shards:  50% Completed | 48/96 [00:46<00:17,  2.72it/s]
Multi-thread loading shards:  51% Completed | 49/96 [00:47<00:16,  2.84it/s]
Multi-thread loading shards:  52% Completed | 50/96 [00:47<00:16,  2.81it/s]
Multi-thread loading shards:  53% Completed | 51/96 [00:47<00:15,  2.93it/s]
Multi-thread loading shards:  54% Completed | 52/96 [00:48<00:21,  2.05it/s]
Multi-thread loading shards:  55% Completed | 53/96 [00:49<00:33,  1.30it/s]
Multi-thread loading shards:  56% Completed | 54/96 [00:50<00:27,  1.55it/s]
Multi-thread loading shards:  57% Completed | 55/96 [00:50<00:22,  1.82it/s]
Multi-thread loading shards:  58% Completed | 56/96 [00:50<00:19,  2.06it/s]
Multi-thread loading shards:  59% Completed | 57/96 [00:51<00:17,  2.25it/s]
Multi-thread loading shards:  60% Completed | 58/96 [00:51<00:15,  2.40it/s]
Multi-thread loading shards:  61% Completed | 59/96 [00:51<00:14,  2.56it/s]
Multi-thread loading shards:  62% Completed | 60/96 [00:52<00:13,  2.67it/s]
Multi-thread loading shards:  64% Completed | 61/96 [00:52<00:12,  2.77it/s]
Multi-thread loading shards:  65% Completed | 62/96 [00:52<00:11,  2.84it/s]
Multi-thread loading shards:  66% Completed | 63/96 [00:53<00:11,  2.89it/s]
Multi-thread loading shards:  67% Completed | 64/96 [00:53<00:11,  2.87it/s]
Multi-thread loading shards:  68% Completed | 65/96 [00:54<00:10,  2.89it/s]
Multi-thread loading shards:  69% Completed | 66/96 [00:54<00:10,  2.91it/s]
Multi-thread loading shards:  70% Completed | 67/96 [00:54<00:10,  2.90it/s]
Multi-thread loading shards:  71% Completed | 68/96 [00:55<00:09,  2.90it/s]
Multi-thread loading shards:  72% Completed | 69/96 [00:55<00:09,  2.89it/s]
Multi-thread loading shards:  73% Completed | 70/96 [00:56<00:14,  1.85it/s]
Multi-thread loading shards:  74% Completed | 71/96 [00:56<00:11,  2.10it/s]
Multi-thread loading shards:  75% Completed | 72/96 [00:57<00:10,  2.23it/s]
Multi-thread loading shards:  76% Completed | 73/96 [00:57<00:09,  2.39it/s]
Multi-thread loading shards:  77% Completed | 74/96 [00:57<00:08,  2.48it/s]
Multi-thread loading shards:  78% Completed | 75/96 [00:58<00:09,  2.15it/s]
Multi-thread loading shards:  79% Completed | 76/96 [00:59<00:10,  1.89it/s]
Multi-thread loading shards:  80% Completed | 77/96 [00:59<00:09,  2.05it/s]
Multi-thread loading shards:  81% Completed | 78/96 [00:59<00:08,  2.24it/s]
Multi-thread loading shards:  82% Completed | 79/96 [01:00<00:07,  2.39it/s]
Multi-thread loading shards:  83% Completed | 80/96 [01:00<00:06,  2.51it/s]
Multi-thread loading shards:  84% Completed | 81/96 [01:00<00:05,  2.65it/s]
Multi-thread loading shards:  85% Completed | 82/96 [01:01<00:04,  2.84it/s]
Multi-thread loading shards:  86% Completed | 83/96 [01:02<00:06,  1.99it/s]
Multi-thread loading shards:  88% Completed | 84/96 [01:02<00:06,  1.99it/s]
Multi-thread loading shards:  90% Completed | 86/96 [01:02<00:03,  3.27it/s]
Multi-thread loading shards:  92% Completed | 88/96 [01:02<00:01,  4.72it/s]
Multi-thread loading shards:  94% Completed | 90/96 [01:02<00:00,  6.49it/s]
Multi-thread loading shards:  96% Completed | 92/96 [01:03<00:00,  8.37it/s]
Multi-thread loading shards:  98% Completed | 94/96 [01:03<00:00,  8.91it/s]
Multi-thread loading shards: 100% Completed | 96/96 [01:03<00:00,  9.35it/s]
Multi-thread loading shards: 100% Completed | 96/96 [01:03<00:00,  1.51it/s]
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:33 PP0 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:33 PP0 TP2 EP2] Load weight end. elapsed=65.63 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=39.03 GB, mem usage=43.09 GB.
+[2026-08-27 04:30:33 PP0 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:33 PP0 TP0 EP0] Load weight end. elapsed=65.81 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=39.03 GB, mem usage=43.09 GB.
+[2026-08-27 04:30:35 PP0 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:35 PP0 TP3 EP3] Load weight end. elapsed=67.37 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=39.03 GB, mem usage=43.09 GB.
+[2026-08-27 04:30:35 PP0 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:35 PP0 TP1 EP1] Load weight end. elapsed=67.46 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=39.03 GB, mem usage=43.09 GB.
+[2026-08-27 04:30:35 PP0 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:35 PP0 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:35 PP0 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:35 PP0 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:35 PP0 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:35 PP0 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:35 PP0 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:35 PP0 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:36 PP1 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:36 PP1 TP3 EP3] Load weight end. elapsed=68.32 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=36.12 GB, mem usage=45.92 GB.
+[2026-08-27 04:30:36 PP1 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP1 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP1 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP1 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP1 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP1 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP1 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP1 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:37 PP1 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:37 PP1 TP1 EP1] Load weight end. elapsed=69.55 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=36.12 GB, mem usage=45.92 GB.
+[2026-08-27 04:30:37 PP1 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:37 PP1 TP0 EP0] Load weight end. elapsed=69.66 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=36.12 GB, mem usage=45.92 GB.
+[2026-08-27 04:30:39 PP1 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:39 PP1 TP2 EP2] Load weight end. elapsed=71.33 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=36.12 GB, mem usage=45.92 GB.
+[2026-08-27 04:30:39 PP1 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:39 PP1 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:39 PP1 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:39 PP1 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:39 PP1 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:39 PP1 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:39 PP1 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:39 PP1 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:53 PP0 TP0 EP0] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:53 PP1 TP3 EP3] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:53 PP1 TP2 EP2] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:53 PP1 TP1 EP1] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:53 PP1 TP0 EP0] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:53 PP0 TP2 EP2] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:53 PP0 TP3 EP3] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:53 PP0 TP1 EP1] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:53 PP0 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:53 PP1 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:53 PP1 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:53 PP1 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:53 PP0 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:53 PP1 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:53 PP0 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:53 PP0 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:53 PP0 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 8.16 GB
+[2026-08-27 04:30:53 PP0 TP0 EP0] Memory pool end. avail mem=30.14 GB
+[2026-08-27 04:30:53 PP0 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 8.16 GB
+[2026-08-27 04:30:53 PP0 TP2 EP2] Memory pool end. avail mem=30.14 GB
+[2026-08-27 04:30:53 PP1 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:53 PP1 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:53 PP1 TP0 EP0] Memory pool end. avail mem=23.24 GB
+[2026-08-27 04:30:53 PP1 TP2 EP2] Memory pool end. avail mem=23.24 GB
+[2026-08-27 04:30:53 PP0 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 8.16 GB
+[2026-08-27 04:30:53 PP1 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:53 PP0 TP3 EP3] Memory pool end. avail mem=30.14 GB
+[2026-08-27 04:30:53 PP1 TP1 EP1] Memory pool end. avail mem=23.24 GB
+[2026-08-27 04:30:53 PP1 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:53 PP0 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 8.16 GB
+[2026-08-27 04:30:53 PP1 TP3 EP3] Memory pool end. avail mem=23.24 GB
+[2026-08-27 04:30:53 PP0 TP1 EP1] Memory pool end. avail mem=30.14 GB
+[2026-08-27 04:30:54 PP0 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:30:54 PP0 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:30:54 PP0 TP0 EP0] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp0_pp0_dp0.json
+[2026-08-27 04:30:54 PP1 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:30:54 PP1 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:30:54 PP0 TP2 EP2] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp2_pp0_dp0.json
+[2026-08-27 04:30:54 PP1 TP2 EP2] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp2_pp1_dp0.json
+[2026-08-27 04:30:54 PP1 TP0 EP0] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp0_pp1_dp0.json
+[2026-08-27 04:30:54 PP1 TP1 EP1] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp1_pp1_dp0.json
+[2026-08-27 04:30:54 PP1 TP3 EP3] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp3_pp1_dp0.json
+[2026-08-27 04:30:54 PP0 TP1 EP1] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp1_pp0_dp0.json
+[2026-08-27 04:30:54 PP0 TP3 EP3] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp3_pp0_dp0.json
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.46profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.32profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.51profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  6.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.09profile/s]
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.07profile/s]
+[TensorRT-LLM][INFO] Set logger level to INFO
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.48profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.11profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.29profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.37profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.16profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.82profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.18profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.58profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:01<00:02,  1.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.66profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.03profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.49profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:01<00:01,  2.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.25profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:01,  2.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:02<00:00,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:02<00:00,  3.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.20profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.01profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  9.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  9.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  9.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 12.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 12.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 11.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 11.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.78profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.16profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  3.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.75profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.50profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.92profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.68profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  3.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.97profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.87profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 10.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.51profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.74profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 11.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.30profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.70profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.72profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.63profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.53profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.29profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.58profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  8.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 10.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.71profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.62profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.43profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.18profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.29profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.37profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.53profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.37profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.35profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.00profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.71profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.60profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.78profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.79profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  8.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.86profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  8.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.16profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.78profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  3.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.15profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.94profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.93profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.25profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.23profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.97profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.79profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.96profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.64profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.12profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.75profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.12profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.25profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.91profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.05profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.66profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.90profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  9.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.73profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.55profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.49profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.13profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.74profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.17profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  3.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.72profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.57profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.59profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.64profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.83profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 13.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.49profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 12.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  9.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  7.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.32profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.93profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.29profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.72profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.75profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.46profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.51profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  3.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.90profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.16profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.72profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.63profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.83profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.70profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.59profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.55profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  6.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.93profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.33profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.75profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.30profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  6.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  7.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.79profile/s]
+[2026-08-27 04:31:21 PP0 TP1 EP1] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.53profile/s][2026-08-27 04:31:21 PP0 TP0 EP0] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.97profile/s]
+[2026-08-27 04:31:21 PP0 TP3 EP3] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.43profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.06profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s][2026-08-27 04:31:21 PP0 TP2 EP2] FlashInfer autotune completed.
+[2026-08-27 04:31:21 PP0 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:21 PP0 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:21 PP0 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:21 PP0 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:21 PP0 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:21 PP0 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:21 PP0 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:21 PP0 TP0 EP0] max_total_num_tokens=3801408, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=54, context_len=1048576, available_gpu_mem=27.35 GB
+[2026-08-27 04:31:21 PP0 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.86profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.58profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:01<00:07,  1.18s/profile]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  2.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.27profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.55profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  3.96profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.86profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  6.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  7.01profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.23profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.05profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  6.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  7.06profile/s]
+[2026-08-27 04:31:26 PP1 TP1 EP1] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.40profile/s]
+[2026-08-27 04:31:26 PP1 TP2 EP2] FlashInfer autotune completed.
+[2026-08-27 04:31:26 PP1 TP0 EP0] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.81profile/s]
+[2026-08-27 04:31:26 PP1 TP3 EP3] FlashInfer autotune completed.
+[2026-08-27 04:31:26 PP1 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:26 PP1 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:26 PP1 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:26 PP1 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:26 PP1 TP0 EP0] max_total_num_tokens=3801408, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=54, context_len=1048576, available_gpu_mem=20.45 GB
+[2026-08-27 04:31:26 PP1 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:26 PP1 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:26 PP1 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:26 PP1 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+W0827 04:31:30.805795  2607 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fe48940, len: 6912: Operation not permitted [1]
+W0827 04:31:30.805797  2608 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4f684400, len: 6912: Operation not permitted [1]
+W0827 04:31:30.805945  2609 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4f962700, len: 6912: Operation not permitted [1]
+W0827 04:31:30.805979  2610 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4f9e8a00, len: 6912: Operation not permitted [1]
+W0827 04:31:30.806130  2612 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fe5f640, len: 55296: Operation not permitted [1]
+W0827 04:31:30.806133  2611 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4f83f000, len: 55296: Operation not permitted [1]
+W0827 04:31:30.806264  2613 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fe328c0, len: 6912: Operation not permitted [1]
+W0827 04:31:30.806334  2614 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fa75cc0, len: 13824: Operation not permitted [1]
+W0827 04:31:30.806449  2615 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fa87f80, len: 6912: Operation not permitted [1]
+W0827 04:31:30.806493  2616 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fa29a80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.203712  2635 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50c65e80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.203859  2636 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50c6bb40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.203975  2637 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50d1ffc0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.204025  2638 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50a0c300, len: 6912: Operation not permitted [1]
+W0827 04:31:31.204075  2640 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50de1940, len: 55296: Operation not permitted [1]
+W0827 04:31:31.204082  2639 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50dc93c0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.204438  2641 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50c87940, len: 6912: Operation not permitted [1]
+W0827 04:31:31.204561  2642 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50cffb00, len: 13824: Operation not permitted [1]
+W0827 04:31:31.204641  2644 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50dc7380, len: 6912: Operation not permitted [1]
+W0827 04:31:31.204722  2643 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50c243c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.323861  2661 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50b13c40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.323863  2662 memory_location.cpp:73] Failed to get NUMA node, addr: 0x511191c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.323912  2663 memory_location.cpp:73] Failed to get NUMA node, addr: 0x511b37c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.324069  2664 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50b2da80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.324110  2666 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5115c980, len: 55296: Operation not permitted [1]
+W0827 04:31:31.324155  2665 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5116ac40, len: 55296: Operation not permitted [1]
+W0827 04:31:31.324479  2667 memory_location.cpp:73] Failed to get NUMA node, addr: 0x511a7e80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.324587  2668 memory_location.cpp:73] Failed to get NUMA node, addr: 0x511b8cc0, len: 13824: Operation not permitted [1]
+W0827 04:31:31.324621  2669 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50fc6800, len: 6912: Operation not permitted [1]
+W0827 04:31:31.324682  2670 memory_location.cpp:73] Failed to get NUMA node, addr: 0x511bc780, len: 6912: Operation not permitted [1]
+W0827 04:31:31.543082  2689 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4902e440, len: 6912: Operation not permitted [1]
+W0827 04:31:31.543203  2690 memory_location.cpp:73] Failed to get NUMA node, addr: 0x49484200, len: 6912: Operation not permitted [1]
+W0827 04:31:31.543314  2691 memory_location.cpp:73] Failed to get NUMA node, addr: 0x49119600, len: 6912: Operation not permitted [1]
+W0827 04:31:31.543375  2692 memory_location.cpp:73] Failed to get NUMA node, addr: 0x493dc780, len: 6912: Operation not permitted [1]
+W0827 04:31:31.543467  2693 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51ac5fc0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.543601  2694 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51a9e780, len: 55296: Operation not permitted [1]
+W0827 04:31:31.543751  2695 memory_location.cpp:73] Failed to get NUMA node, addr: 0x49149a40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.543912  2696 memory_location.cpp:73] Failed to get NUMA node, addr: 0x518c0840, len: 13824: Operation not permitted [1]
+W0827 04:31:31.543947  2697 memory_location.cpp:73] Failed to get NUMA node, addr: 0x49303540, len: 6912: Operation not permitted [1]
+W0827 04:31:31.543990  2698 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51951a80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681525  2717 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50376e40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681525  2718 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5058ee40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681525  2716 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5008b980, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681548  2715 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fe0e840, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681727  2719 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5040eb80, len: 55296: Operation not permitted [1]
+W0827 04:31:31.681779  2724 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50209ac0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681811  2720 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5041c3c0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.681828  2722 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5048cc40, len: 13824: Operation not permitted [1]
+W0827 04:31:31.681844  2723 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fde7a00, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681851  2721 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50386580, len: 6912: Operation not permitted [1]
+W0827 04:31:31.730942  2745 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fce4240, len: 6912: Operation not permitted [1]
+W0827 04:31:31.730918  2744 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fed7d40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.730950  2743 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fdcaf80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.731007  2748 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5074c400, len: 55296: Operation not permitted [1]
+W0827 04:31:31.731073  2747 memory_location.cpp:73] Failed to get NUMA node, addr: 0x507bd9c0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.731086  2746 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4f931480, len: 6912: Operation not permitted [1]
+W0827 04:31:31.731294  2749 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd958c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.731361  2751 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd31d80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.731361  2750 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4ffabe00, len: 13824: Operation not permitted [1]
+W0827 04:31:31.731566  2752 memory_location.cpp:73] Failed to get NUMA node, addr: 0x508527c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.808167  2778 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd3ce40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.808190  2777 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd20740, len: 6912: Operation not permitted [1]
+W0827 04:31:31.808192  2776 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4ff86c80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.808637  2782 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fe88780, len: 6912: Operation not permitted [1]
+W0827 04:31:31.808724  2780 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fe48c80, len: 55296: Operation not permitted [1]
+W0827 04:31:31.808760  2781 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fe564c0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.809056  2785 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4ffa7e40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.809049  2779 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4feaa740, len: 6912: Operation not permitted [1]
+W0827 04:31:31.809075  2784 memory_location.cpp:73] Failed to get NUMA node, addr: 0x502fabc0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.809239  2783 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4ffa1f80, len: 13824: Operation not permitted [1]
+W0827 04:31:31.851475  2807 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4ff6fcc0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.851596  2808 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fc0a280, len: 6912: Operation not permitted [1]
+W0827 04:31:31.851615  2809 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd7d700, len: 6912: Operation not permitted [1]
+W0827 04:31:31.851732  2810 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd761c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.851951  2811 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fde6cc0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.851970  2812 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4feeb000, len: 55296: Operation not permitted [1]
+W0827 04:31:31.851962  2813 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fbe2680, len: 6912: Operation not permitted [1]
+W0827 04:31:31.851987  2816 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fbf07c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.852048  2815 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4ffa9940, len: 6912: Operation not permitted [1]
+W0827 04:31:31.852046  2814 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd235c0, len: 13824: Operation not permitted [1]
+[2026-08-27 04:31:32] INFO:     Started server process [1]
+[2026-08-27 04:31:32] INFO:     Waiting for application startup.
+[2026-08-27 04:31:32] Calling super().encode with {'add_special_tokens': False}
+[2026-08-27 04:31:32] Calling super().encode with {'add_special_tokens': False}
+[2026-08-27 04:31:32] INFO:     Application startup complete.
+[2026-08-27 04:31:32] INFO:     Uvicorn running on http://0.0.0.0:30000 (Press CTRL+C to quit)
+[2026-08-27 04:31:33] INFO:     127.0.0.1:60348 - "GET /model_info HTTP/1.1" 200 OK
+[2026-08-27 04:31:33] Start of pd disaggregation warmup ...
+[2026-08-27 04:31:37] INFO:     174.1.60.1:33088 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:38] INFO:     174.1.60.1:33092 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:39] INFO:     174.1.60.1:33106 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:40] INFO:     174.1.60.1:33110 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:41] INFO:     174.1.60.1:42028 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:42] INFO:     174.1.60.1:42030 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:43] INFO:     174.1.60.1:42046 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:44] INFO:     174.1.60.1:42050 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:45] INFO:     174.1.60.1:42060 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:46] INFO:     174.1.60.1:42072 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:47] INFO:     174.1.60.1:42086 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:48] INFO:     174.1.60.1:42092 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:49] INFO:     174.1.60.1:42098 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:50] INFO:     174.1.60.1:42100 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:51] INFO:     174.1.60.1:46852 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:52] INFO:     174.1.60.1:46856 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:53] INFO:     174.1.60.1:46872 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:54] INFO:     174.1.60.1:46876 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:55] INFO:     174.1.60.1:46886 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:56] INFO:     174.1.60.1:46890 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:57] INFO:     174.1.60.1:46898 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:58] INFO:     174.1.60.1:46912 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:59] INFO:     174.1.60.1:46928 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:00] INFO:     174.1.60.1:46936 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:01] INFO:     174.1.60.1:57672 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:02] INFO:     174.1.60.1:57676 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:03] INFO:     174.1.60.1:57686 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:04] INFO:     174.1.60.1:57688 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:05] INFO:     174.1.60.1:57696 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:06] INFO:     174.1.60.1:57708 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:12] INFO:     174.1.60.1:39714 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:13] INFO:     174.1.60.1:39728 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:14] INFO:     174.1.60.1:39736 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:15] INFO:     174.1.60.1:39746 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:16] INFO:     174.1.60.1:39760 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:17] INFO:     174.1.60.1:39772 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:18] INFO:     174.1.60.1:39782 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:19] INFO:     174.1.60.1:39786 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:20] INFO:     174.1.60.1:39800 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:21] INFO:     174.1.60.1:41036 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:22] INFO:     174.1.60.1:41052 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:23] INFO:     174.1.60.1:41058 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:24] INFO:     174.1.60.1:41068 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:25] INFO:     174.1.60.1:41074 - "GET /health HTTP/1.1" 503 Service Unavailable
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank1]:[W827 04:32:26.019157362 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank0]:[W827 04:32:26.019963648 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank3]:[W827 04:32:26.020028119 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank2]:[W827 04:32:26.020057305 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank6]:[W827 04:32:26.045976535 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank7]:[W827 04:32:26.046176399 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank5]:[W827 04:32:26.048033943 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank4]:[W827 04:32:26.048331224 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[2026-08-27 04:32:26] INFO:     174.1.60.1:41078 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:27] INFO:     174.1.60.1:41090 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:28] INFO:     174.1.60.1:41102 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:29] INFO:     174.1.60.1:41104 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:30] INFO:     174.1.60.1:41120 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:31] INFO:     174.1.60.1:57002 - "GET /health HTTP/1.1" 503 Service Unavailable
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank6]:[W827 04:32:31.709756552 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank7]:[W827 04:32:31.709839299 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank5]:[W827 04:32:31.709991082 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank4]:[W827 04:32:31.710123265 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[2026-08-27 04:32:32] INFO:     174.1.60.1:57016 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:33] INFO:     174.1.60.1:57030 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:34] INFO:     174.1.60.1:57046 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:35] INFO:     174.1.60.1:57056 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:36] INFO:     174.1.60.1:57066 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:37] INFO:     174.1.60.1:57070 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:38] INFO:     174.1.60.1:57072 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:39] INFO:     174.1.60.1:57082 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:40] INFO:     174.1.60.1:57098 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:41] INFO:     174.1.60.1:56606 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:50] INFO:     174.1.60.1:56620 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:51] INFO:     174.1.60.1:45658 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:52] INFO:     174.1.60.1:45672 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:53] INFO:     174.1.60.1:45674 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:54] INFO:     174.1.60.1:45690 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:55] INFO:     174.1.60.1:45704 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:56] INFO:     174.1.60.1:45708 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:57] INFO:     174.1.60.1:45714 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:58] INFO:     174.1.60.1:45722 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:00] INFO:     174.1.60.1:45736 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:01] INFO:     174.1.60.1:45746 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:02] INFO:     174.1.60.1:33310 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:03] INFO:     174.1.60.1:33320 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:04] INFO:     174.1.60.1:33324 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:05] INFO:     174.1.60.1:33328 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:06] INFO:     174.1.60.1:33336 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:07] INFO:     174.1.60.1:33350 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:08] INFO:     174.1.60.1:33354 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:09] INFO:     174.1.60.1:33356 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:10] INFO:     174.1.60.1:33368 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:11] INFO:     174.1.60.1:39318 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:12] INFO:     174.1.60.1:39328 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:13] INFO:     174.1.60.1:39336 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:14] INFO:     174.1.60.1:39344 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:15] INFO:     174.1.60.1:39346 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:16] INFO:     174.1.60.1:39352 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:17] INFO:     174.1.60.1:39364 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:18] INFO:     174.1.60.1:39378 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:19] INFO:     174.1.60.1:39390 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:20] INFO:     174.1.60.1:39402 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:25] INFO:     174.1.60.1:37792 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:26] INFO:     174.1.60.1:37804 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:27] INFO:     174.1.60.1:37812 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:28] INFO:     174.1.60.1:37826 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:29] INFO:     174.1.60.1:37830 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:30] INFO:     174.1.60.1:37844 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:31] INFO:     174.1.60.1:44208 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:32] INFO:     174.1.60.1:44220 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:33] INFO:     174.1.60.1:44226 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:34] INFO:     174.1.60.1:44228 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:35] INFO:     174.1.60.1:44238 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:37] INFO:     174.1.60.1:44248 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:38] INFO:     174.1.60.1:44258 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:39] INFO:     174.1.60.1:44272 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:40] INFO:     174.1.60.1:44276 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:41] INFO:     174.1.60.1:44194 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:42] INFO:     174.1.60.1:44198 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:43] INFO:     174.1.60.1:44212 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:44] INFO:     174.1.60.1:44216 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:45] INFO:     174.1.60.1:44222 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:46] INFO:     174.1.60.1:44236 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:47] INFO:     174.1.60.1:44246 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:48] INFO:     174.1.60.1:44252 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:49] INFO:     174.1.60.1:44264 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:50] INFO:     174.1.60.1:44278 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:51] INFO:     174.1.60.1:35728 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:52] INFO:     174.1.60.1:35736 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:53] INFO:     174.1.60.1:35752 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:54] INFO:     174.1.60.1:35754 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:55] INFO:     174.1.60.1:35758 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:04] INFO:     174.1.60.1:44514 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:05] INFO:     174.1.60.1:44518 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:06] INFO:     174.1.60.1:44520 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:07] INFO:     174.1.60.1:44526 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:08] INFO:     174.1.60.1:44530 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:09] INFO:     174.1.60.1:44546 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:10] INFO:     174.1.60.1:44550 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:11] INFO:     174.1.60.1:45098 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:12] INFO:     127.0.0.1:59722 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:12] INFO:     174.1.60.1:45106 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:13] INFO:     174.1.60.1:45122 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:14] INFO:     174.1.60.1:45124 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:15] INFO:     174.1.60.1:45136 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:16] INFO:     174.1.60.1:45146 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:17] INFO:     174.1.60.1:45152 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:19] INFO:     174.1.60.1:45158 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:20] INFO:     174.1.60.1:45170 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:21] INFO:     174.1.60.1:45182 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:22] INFO:     174.1.60.1:39946 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:23] INFO:     174.1.60.1:39958 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:24] INFO:     174.1.60.1:39962 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:25] INFO:     174.1.60.1:39978 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:26] INFO:     174.1.60.1:39990 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:27] INFO:     174.1.60.1:39994 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:28] INFO:     174.1.60.1:39996 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:29] INFO:     174.1.60.1:40002 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:30] INFO:     174.1.60.1:40012 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:31] INFO:     174.1.60.1:46978 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:32] INFO:     174.1.60.1:46988 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:33] INFO:     174.1.60.1:46996 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:34] INFO:     174.1.60.1:47004 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:39] INFO:     174.1.60.1:47020 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:40] INFO:     174.1.60.1:47022 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:41] INFO:     174.1.60.1:59158 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:42] INFO:     174.1.60.1:59170 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:43] INFO:     174.1.60.1:59172 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:44] INFO:     174.1.60.1:59180 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:45] INFO:     174.1.60.1:59184 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:46] INFO:     174.1.60.1:59186 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:47] INFO:     174.1.60.1:59200 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:48] INFO:     174.1.60.1:59202 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:49] INFO:     174.1.60.1:59214 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:50] INFO:     174.1.60.1:59220 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:52] INFO:     174.1.60.1:40566 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:53] INFO:     174.1.60.1:40576 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:54] INFO:     174.1.60.1:40588 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:55] INFO:     174.1.60.1:40596 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:56] INFO:     174.1.60.1:40598 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:57] INFO:     174.1.60.1:40604 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:58] INFO:     174.1.60.1:40608 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:59] INFO:     174.1.60.1:40610 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:00] INFO:     174.1.60.1:40622 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:01] INFO:     174.1.60.1:41102 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:02] INFO:     174.1.60.1:41108 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:03] INFO:     174.1.60.1:41122 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:04] INFO:     174.1.60.1:41124 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:05] INFO:     174.1.60.1:41134 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:06] INFO:     174.1.60.1:41140 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:07] INFO:     174.1.60.1:41142 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:08] INFO:     174.1.60.1:41158 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:09] INFO:     174.1.60.1:41160 - "GET /health HTTP/1.1" 503 Service Unavailable
+[rank3]:[W827 04:35:10.359210278 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank2]:[W827 04:35:10.359272236 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank0]:[W827 04:35:10.360047094 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank1]:[W827 04:35:10.370610409 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[2026-08-27 04:35:10 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.28
+[2026-08-27 04:35:10 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.29
+[2026-08-27 04:35:10] INFO:     127.0.0.1:60364 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 04:35:10] Disaggregation warmup requests completed for all 1 DP ranks
+[2026-08-27 04:35:10] End of disaggregation warmup
+[2026-08-27 04:35:10] The server is fired up and ready to roll!
+[2026-08-27 04:35:29 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.39
+[2026-08-27 04:35:29 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.39
+[2026-08-27 04:35:31 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25.83
+[2026-08-27 04:35:31 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25.84
+[2026-08-27 04:35:32] INFO:     174.1.60.1:49902 - "GET /health HTTP/1.1" 200 OK
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/d_starting_p_node1.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/d_starting_p_node1.log
new file mode 100644
index 0000000..d4818b3
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/d_starting_p_node1.log
@@ -0,0 +1,745 @@
+/usr/local/lib/python3.12/dist-packages/torchao/quantization/quant_api.py:1731: SyntaxWarning: invalid escape sequence '\.'
+  """Configuration class for applying different quantization configs to modules or parameters based on their fully qualified names (FQNs).
+/sgl-workspace/sglang/python/sglang/launch_server.py:56: UserWarning: 'python -m sglang.launch_server' is still supported, but 'sglang serve' is the recommended entrypoint.
+  Example: sglang serve --model-path <model> [options]
+  warnings.warn(
+[2026-08-27 04:28:43] Breakable CUDA graph is incompatible with MLA attention; disabling prefill CUDA graph.
+[2026-08-27 04:28:43] Attention backend not specified. Use triton backend by default.
+[2026-08-27 04:28:43] Pipeline parallelism is incompatible with overlap schedule.
+[2026-08-27 04:28:46] server_args=ServerArgs(model_path='/data/hf_models/Kimi-K3', tokenizer_path='/data/hf_models/Kimi-K3', tokenizer_mode='auto', tokenizer_backend='huggingface', tokenizer_worker_num=1, detokenizer_worker_num=1, skip_tokenizer_init=False, load_format='auto', model_loader_extra_config='{}', trust_remote_code=True, context_length=None, is_embedding=False, enable_multimodal=None, revision=None, model_impl='auto', model_config_parser='auto', json_model_override_args='{}', dtype='auto', quantization=None, quantization_param_path=None, kv_cache_dtype='auto', enable_fp32_lm_head=False, modelopt_quant=None, modelopt_checkpoint_restore_path=None, modelopt_checkpoint_save_path=None, modelopt_export_path=None, quantize_and_serve=False, rl_quant_profile=None, enable_tf32_matmul=False, mem_fraction_static=0.88, max_running_requests=None, max_queued_requests=None, max_total_tokens=None, chunked_prefill_size=8192, enable_dynamic_chunking=False, max_prefill_tokens=16384, prefill_max_requests=None, schedule_policy='fcfs', enable_priority_scheduling=False, disable_priority_preemption=False, default_priority_value=None, abort_on_priority_when_disabled=False, schedule_low_priority_values_first=False, priority_scheduling_preemption_threshold=10, retraction_policy='length', schedule_conservativeness=1.0, page_size=64, swa_full_tokens_ratio=0.8, disable_hybrid_swa_memory=False, radix_eviction_policy='lru', prefill_only_disable_kv_cache=False, disable_radix_cache=True, enable_page_major_kv_layout=False, enable_unified_memory=False, disable_chunked_prefix_cache=False, disable_overlap_schedule=True, num_continuous_decode_steps=1, scheduler_recv_interval=1, enable_mixed_chunk=False, nccl_port=None, dist_timeout=3600, dist_init_addr='174.1.60.1:20000', nnodes=4, node_rank=1, tp_size=4, dcp_size=1, pp_size=8, pp_max_micro_batch_size=None, pp_async_batch_depth=0, dp_size=1, load_balance_method='follow_bootstrap_room', attn_cp_size=1, moe_dp_size=1, dcp_comm_backend='ag_rs', dcp_replicate_q_proj=None, dwdp_size=1, enable_prefill_cp=False, cp_strategy=None, enable_dsa_cache_layer_split=False, enable_dsa_prefill_context_parallel=False, dsa_prefill_cp_mode='round-robin-split', enable_prefill_context_parallel=False, prefill_cp_mode='in-seq-split', enable_cp_decode_attn_tp=False, enable_dp_attention=False, enable_dp_attention_local_control_broadcast=False, enable_dp_lm_head=False, enable_attn_tp_input_scattered=False, disable_attn_tp_gather=False, enable_p2p_check=False, device='cuda', base_gpu_id=0, gpu_id_step=1, random_seed=577755302, watchdog_timeout=300, soft_watchdog_timeout=None, sleep_on_idle=False, use_ray=False, custom_sigquit_handler=None, numa_node=None, gc_threshold=None, host='0.0.0.0', port=30000, fastapi_root_path='', smg_grpc_mode=False, grpc_mode=False, grpc_port=None, sidecar=None, sidecar_args=None, skip_server_warmup=False, warmups=None, enable_http2=False, ssl_keyfile=None, ssl_certfile=None, ssl_ca_certs=None, ssl_keyfile_password=None, enable_ssl_refresh=False, api_key=None, admin_api_key=None, served_model_name='kimi-k3', weight_version='default', chat_template=None, hf_chat_template_name=None, completion_template=None, file_storage_path='sglang_storage', enable_cache_report=False, reasoning_parser=None, default_chat_template_kwargs=None, strip_thinking_cache=False, enable_strict_thinking=False, tool_call_parser=None, tool_server=None, sampling_defaults='model', asr_max_buffer_seconds=60, asr_max_concurrent_sessions=32, preferred_sampling_params=None, allow_auto_truncate=False, stream_interval=1, batch_notify_size=16, stream_response_default_include_usage=False, incremental_streaming_output=False, enable_streaming_session=False, enable_session_radix_cache=False, log_level='info', log_level_http=None, log_requests=False, log_requests_level=2, log_requests_format='text', log_requests_target=None, uvicorn_access_log_exclude_prefixes=[], crash_dump_folder=None, show_time_cost=False, enable_metrics=False, smg_http_sidecar_port=None, enable_mfu_metrics=False, enable_metrics_for_all_schedulers=False, load_snapshot_publish_interval=15, tokenizer_metrics_custom_labels_header='x-custom-labels', tokenizer_metrics_allowed_custom_labels=None, extra_metric_labels=None, bucket_time_to_first_token=None, bucket_inter_token_latency=None, bucket_e2e_request_latency=None, prompt_tokens_buckets=None, generation_tokens_buckets=None, gc_warning_threshold_secs=0.0, decode_log_interval=40, enable_request_time_stats_logging=False, kv_events_config=None, enable_forward_pass_metrics=False, forward_pass_metrics_worker_id='', forward_pass_metrics_ipc_name=None, enable_trace=False, trace_modules='request', otlp_traces_endpoint='localhost:4317', export_metrics_to_file=False, export_metrics_to_file_dir=None, stat_loggers=None, constrained_json_whitespace_pattern=None, constrained_json_disable_any_whitespace=False, attention_backend='triton', decode_attention_backend=None, prefill_attention_backend=None, sampling_backend='flashinfer', grammar_backend='xgrammar', radix_cache_backend=None, mm_attention_backend=None, fp8_gemm_runner_backend='auto', fp4_gemm_runner_backend='auto', bf16_gemm_backend='auto', dsa_prefill_backend=None, dsa_decode_backend=None, dsa_paged_mqa_logits_backend='auto', dsa_topk_backend='sgl-kernel', disable_flashinfer_autotune=False, flashinfer_autotune_skip_ops=None, mamba_backend='triton', cuda_graph_config=CudaGraphConfig(decode=PhaseConfig(backend='disabled', max_bs=16, bs=[1, 2, 4, 8, 12, 16], tc_compiler='eager', full_prefill_max_req=None), prefill=PhaseConfig(backend='disabled', max_bs=2048, bs=[4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], tc_compiler='eager', full_prefill_max_req=None)), cuda_graph_backend_decode=None, cuda_graph_backend_prefill=None, cuda_graph_max_bs_decode=16, cuda_graph_max_bs_prefill=None, cuda_graph_bs_decode=None, cuda_graph_bs_prefill=None, cuda_graph_tc_compiler=None, disable_prefill_cuda_graph=False, disable_decode_cuda_graph=False, disable_cuda_graph=False, disable_cuda_graph_padding=False, enable_profile_cuda_graph=False, enable_cudagraph_gc=False, debug_cuda_graph=False, enable_layerwise_nvtx_marker=False, enable_nccl_nvls=False, enable_symm_mem=False, triton_attention_reduce_in_fp32=False, triton_attention_num_kv_splits=8, triton_attention_split_tile_size=None, flashinfer_mla_disable_ragged=False, enable_fused_qk_norm_rope=False, enable_precise_embedding_interpolation=False, enable_fused_moe_sum_all_reduce=False, enable_deepseek_v4_fp4_indexer=False, disable_custom_all_reduce=False, enable_mscclpp=False, enable_torch_symm_mem=False, enable_scattered_sconv=False, pre_warm_nccl=False, enable_quant_communications=False, enable_flashinfer_allreduce_fusion=False, enforce_disable_flashinfer_allreduce_fusion=False, flashinfer_allreduce_fusion_backend=None, enable_aiter_allreduce_fusion=False, enable_torch_compile=False, enable_torch_compile_debug_mode=False, torch_compile_max_bs=32, torchao_config='', speculative_algorithm=None, speculative_draft_model_path=None, speculative_draft_model_revision=None, speculative_draft_load_format=None, speculative_num_steps=None, speculative_eagle_topk=None, speculative_num_draft_tokens=None, speculative_dflash_block_size=None, speculative_dspark_block_size=None, speculative_dspark_sps_table_path=None, speculative_dspark_confidence_sts_path=None, speculative_dspark_align_verify_tokens_to_graph_tier=False, speculative_accept_threshold_single=1.0, speculative_accept_threshold_acc=1.0, speculative_use_rejection_sampling=False, speculative_token_map=None, speculative_attention_mode='prefill', speculative_draft_attention_backend=None, speculative_draft_window_size=None, speculative_moe_runner_backend='flashinfer_mxfp4', speculative_moe_a2a_backend=None, speculative_draft_model_quantization=None, speculative_skip_dp_mlp_sync=False, enable_multi_layer_eagle=False, speculative_adaptive=False, speculative_adaptive_config=None, decoupled_spec_bind_endpoint=None, decoupled_spec_connect_endpoints=None, decoupled_spec_rank=None, decoupled_spec_role='null', spec_trace_dir=None, speculative_ngram_min_bfs_breadth=1, speculative_ngram_max_bfs_breadth=10, speculative_ngram_match_type='BFS', speculative_ngram_max_trie_depth=18, speculative_ngram_capacity=10000000, speculative_ngram_external_corpus_path=None, speculative_ngram_external_sam_budget=0, speculative_ngram_external_corpus_max_tokens=10000000, ep_size=4, moe_a2a_backend='none', moe_runner_backend='flashinfer_mxfp4', flashinfer_mxfp4_moe_precision='default', deepep_mode='auto', fuseep_mode=2, deepep_dispatcher_output_dtype='auto', ep_num_redundant_experts=0, ep_dispatch_algorithm=None, init_expert_location='trivial', enable_eplb=False, eplb_algorithm='auto', eplb_rebalance_num_iterations=1000, eplb_rebalance_layers_per_chunk=None, eplb_min_rebalancing_utilization_threshold=1.0, expert_distribution_recorder_mode=None, expert_distribution_recorder_buffer_size=1000, enable_expert_distribution_metrics=False, deepep_config=None, moe_dense_tp_size=None, elastic_ep_backend=None, enable_elastic_expert_backup=False, mooncake_ib_device=None, enable_waterfill=False, ep_join_mode=None, ep_join_rank_offset=0, elastic_ep_initial_size=None, max_ep_size=None, elastic_ep_scale_timeout=600, elastic_ep_rejoin=False, disable_flashinfer_cutlass_moe_fp4_allgather=False, disable_shared_experts_fusion=False, enforce_shared_experts_fusion=False, max_mamba_cache_size=None, mamba_ssm_dtype=None, mamba_max_states_per_path=-1, enable_mamba_cache_stochastic_rounding=False, mamba_cache_philox_rounds=0, mamba_full_memory_ratio=0.36, mamba_radix_cache_strategy='extra_buffer_lazy', uses_mamba_radix_cache=False, mamba_track_interval=256, enable_int8_mamba_checkpoint=False, int8_mamba_ckpt_size=None, linear_attn_backend='triton', linear_attn_decode_backend=None, linear_attn_prefill_backend=None, linear_attn_verify_backend=None, enable_linear_replayssm=False, linear_replayssm_cache_len=16, enable_linear_replayssm_spec=False, enable_hierarchical_cache=False, hicache_ratio=2.0, hicache_size=0, hicache_write_policy='write_through', hicache_io_backend='kernel', hicache_mem_layout='page_first', hicache_storage_backend=None, hicache_storage_prefetch_policy='timeout', hicache_storage_backend_extra_config=None, enable_hisparse=False, hisparse_config=None, enable_broadcast_mm_inputs_process=False, enable_prefix_mm_cache=False, mm_enable_dp_encoder=False, mm_process_config={}, mm_processor_worker_num=0, mm_io_worker_num=0, limit_mm_data_per_request=None, enable_mm_global_cache=False, disable_fast_image_processor=False, mm_feature_transport='cpu', keep_mm_feature_on_device=False, enable_lora=None, enable_lora_overlap_loading=None, max_lora_rank=None, lora_target_modules=None, lora_paths=None, max_loaded_loras=None, max_loras_per_batch=8, lora_eviction_policy='lru', lora_backend='csgmv', max_lora_chunk_size=16, experts_shared_outer_loras=None, lora_use_virtual_experts=False, lora_strict_loading=False, lora_drain_wait_threshold=0.0, enable_two_batch_overlap=False, enable_single_batch_overlap=False, tbo_token_distribution_threshold=0.48, cpu_offload_gb=0, offload_group_size=-1, offload_num_in_group=1, offload_prefetch_step=1, offload_mode='cpu', enable_lmcache=False, lmcache_config_file=None, enable_flexkv=False, flexkv_config_file=None, kt_weight_path=None, kt_method='AMXINT4', kt_cpuinfer=None, kt_threadpool_count=2, kt_num_gpu_experts=None, kt_max_deferred_experts_per_token=None, dllm_algorithm=None, dllm_algorithm_config=None, dllm_fdfo=True, disaggregation_mode='prefill', disaggregation_transfer_backend='mooncake', disaggregation_bootstrap_port=28800, disaggregation_ib_device='mlx5_0,mlx5_1,mlx5_2,mlx5_3', disaggregation_decode_enable_radix_cache=False, disaggregation_decode_enable_offload_kvcache=False, num_reserved_decode_tokens=512, disaggregation_decode_extra_slots=None, disaggregation_decode_polling_interval=1, optimistic_prefill_attempts=0, encoder_only=False, language_only=False, encoder_transfer_backend='zmq_to_tokenizer', encoder_urls=[], encoder_bootstrap_port=8997, encoder_register_urls=[], enable_adaptive_dispatch_to_encoder=False, enable_pdmux=False, pdmux_config_path=None, sm_group_num=8, custom_weight_loader=[], weight_loader_disable_mmap=False, weight_loader_prefetch_checkpoints=False, weight_loader_prefetch_num_threads=4, weight_loader_drop_cache_after_load=False, remote_instance_weight_loader_seed_instance_ip=None, remote_instance_weight_loader_seed_instance_service_port=None, remote_instance_weight_loader_send_weights_group_ports=None, remote_instance_weight_loader_backend='nccl', remote_instance_weight_loader_start_seed_via_transfer_engine=False, engine_info_bootstrap_port=6789, modelexpress_config=None, download_dir=None, model_checksum=None, delete_ckpt_after_loading=False, decrypted_config_file=None, decrypted_draft_config_file=None, checkpoint_engine_wait_weights_before_ready=False, enable_prefill_delayer=False, prefill_delayer_max_delay_passes=30, prefill_delayer_token_usage_low_watermark=None, prefill_delayer_forward_passes_buckets=None, prefill_delayer_wait_seconds_buckets=None, prefill_delayer_queue_min_ratio=None, prefill_delayer_max_delay_ms=None, min_free_slots_delay=None, enable_deterministic_inference=False, rl_on_policy_target=None, kv_canary='none', kv_canary_real_data='none', kv_canary_sweep_interval=0, enable_dynamic_batch_tokenizer=False, dynamic_batch_tokenizer_batch_size=32, dynamic_batch_tokenizer_batch_timeout=0.002, enable_tokenizer_batch_encode=False, disable_tokenizer_batch_decode=False, debug_tensor_dump_output_folder=None, debug_tensor_dump_layers=None, debug_tensor_dump_input_file=None, enable_memory_saver=False, enable_weights_cpu_backup=False, enable_draft_weights_cpu_backup=False, enable_custom_logit_processor=False, enable_return_hidden_states=False, enable_return_routed_experts=False, enable_return_indexer_topk=False, disable_outlines_disk_cache=False, enable_mis=False, weight_cache_mode='off', weight_cache_socket=None, weight_cache_timeout=1800, forward_hooks=None, msprobe_dump_config=None)
+[2026-08-27 04:28:46] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:46] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:46] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:46] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:46] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:46] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:46] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:46] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:02 PP3 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:02 PP3 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:03 PP3 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:03 PP3 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:03 PP3 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:03 PP3 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:03 PP2 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:03 PP2 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:03 PP2 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:03 PP2 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:04 PP2 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:04 PP2 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:04 PP2 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:04 PP2 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:04 PP3 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:04 PP3 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:05 PP3 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:05 PP2 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:05 PP3 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:05 PP3 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:05 PP2 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:06.044592   241 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:06.044694   241 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:06.044710   241 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.2 port: 12001
+I0827 04:29:06.044759   241 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.2:16343
+I0827 04:29:06.044883   241 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:06.053642   242 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:06.053738   242 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:06.053754   242 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.2 port: 12001
+I0827 04:29:06.053797   242 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.2:15170
+I0827 04:29:06.053925   242 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:06.058039   237 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:06.058135   237 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:06.058149   237 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.2 port: 12001
+I0827 04:29:06.058192   237 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.2:15486
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:06.058215   236 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:06.058310   236 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:06.058316   237 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:06.058326   236 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.2 port: 12001
+I0827 04:29:06.058367   236 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.2:16140
+I0827 04:29:06.058492   236 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:06.059734   239 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:06.059828   239 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:06.059844   239 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.2 port: 12001
+I0827 04:29:06.059890   239 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.2:15755
+I0827 04:29:06.060027   239 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:06.064433   237 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:06.064513   236 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:06.065588   239 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:06.069945   237 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:06.070852   236 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:06.071113   239 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:06.071978   241 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:06.079348   236 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:06.079967   242 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:06.080070   241 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:06.087044   241 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:06.087074   242 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:06.094024   242 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:06.095021   241 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:06.095822   237 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:06.097810   241 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:06.097839   241 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:06.097889   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.099939   239 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:06.100984   242 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:06.103683   242 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:06.103708   242 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:06.103756   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.103785   236 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:06.103833   237 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:06.106014   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:06.106982   236 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:06.107007   236 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:06.107053   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.107446   241 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:02
+I0827 04:29:06.107475   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.107936   239 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:06.107971   237 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:06.107997   237 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:06.108048   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.110760   239 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:06.110790   239 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:06.110847   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.111824   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:06.113214   242 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:02
+I0827 04:29:06.113242   242 rdma_context.cpp:205] Using SIEVE endpoint store
+[2026-08-27 04:29:06 PP2 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+I0827 04:29:06.114821   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:06.115196   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+[2026-08-27 04:29:06 PP2 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+I0827 04:29:06.116189   241 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:02
+I0827 04:29:06.116214   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.116621   237 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:02
+I0827 04:29:06.116650   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.118325   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:06.119874   239 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:02
+I0827 04:29:06.119906   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.121601   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:06.122952   242 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:02
+I0827 04:29:06.122977   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.127393   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:06.128841   239 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:02
+I0827 04:29:06.128867   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.132040   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:06.133428   236 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:02
+I0827 04:29:06.133458   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.136312   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:06.137817   239 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:02
+I0827 04:29:06.137846   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.144285   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:06.144373   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:06.145742   237 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:02
+I0827 04:29:06.145776   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.145833   241 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:02
+I0827 04:29:06.145866   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.152084   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:06.153820   242 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:02
+I0827 04:29:06.153860   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.154352   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:06.155728   241 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:02
+I0827 04:29:06.155835   241 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:06 PP3 TP2 EP2] Init torch distributed begin.
+I0827 04:29:06.163965   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:06.164443   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:06.164605   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:06.165319   236 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:02
+I0827 04:29:06.165345   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.165951   239 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:02
+I0827 04:29:06.166070   239 transfer_engine_impl.cpp:388] installTransport, type=rdma
+I0827 04:29:06.166186   242 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:02
+I0827 04:29:06.166311   242 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:06 PP3 TP0 EP0] Init torch distributed begin.
+[2026-08-27 04:29:06 PP3 TP3 EP3] Init torch distributed begin.
+I0827 04:29:06.172065   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:06.173699   237 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:02
+I0827 04:29:06.173743   237 rdma_context.cpp:205] Using SIEVE endpoint store
+[2026-08-27 04:29:06 PP3 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+I0827 04:29:06.192251   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:06.193763   236 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:02
+I0827 04:29:06.193797   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.200259   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:06.201810   237 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:02
+I0827 04:29:06.201937   237 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:06 PP2 TP2 EP2] Init torch distributed begin.
+I0827 04:29:06.220070   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:06.221505   236 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:02
+I0827 04:29:06.221653   236 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:06 PP2 TP1 EP1] Init torch distributed begin.
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:06.613179   235 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:06.613286   235 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:06.613301   235 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.2 port: 12001
+I0827 04:29:06.613344   235 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.2:16382
+I0827 04:29:06.613492   235 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:06.618471   235 topology.cpp:127] Device mlx5_0 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:06.618719   238 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:06.618824   238 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:06.618839   238 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.2 port: 12001
+I0827 04:29:06.618882   238 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.2:16170
+I0827 04:29:06.619020   238 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:06.624203   238 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:06.629066   238 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:06.633975   238 topology.cpp:127] Device mlx5_2 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:06.636366   240 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:06.636454   240 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:06.636466   240 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.2 port: 12001
+I0827 04:29:06.636509   240 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.2:15840
+I0827 04:29:06.636675   240 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:06.638876   238 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:06.640780   238 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:06.640802   238 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:06.640842   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.641396   240 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:06.643196   235 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:06.645701   240 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:06.649896   235 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:06.650310   240 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:06.654819   240 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:06.656375   240 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:06.656396   240 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:06.656432   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.657156   235 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:06.659863   235 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:06.659895   235 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:06.659950   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.661671   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:06.662956   240 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:02
+I0827 04:29:06.662979   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.664074   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:06.665762   238 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:02
+I0827 04:29:06.665796   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.667351   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:06.668808   235 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:02
+I0827 04:29:06.668838   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.676294   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:06.677721   235 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:02
+I0827 04:29:06.677748   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.692129   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:06.693621   240 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:02
+I0827 04:29:06.693648   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.696089   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:06.697444   238 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:02
+I0827 04:29:06.697470   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.707720   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:06.709362   235 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:02
+I0827 04:29:06.709394   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.719987   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:06.721463   240 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:02
+I0827 04:29:06.721489   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.728108   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:06.728855   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:06.729467   238 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:02
+I0827 04:29:06.729491   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.730392   240 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:02
+I0827 04:29:06.730533   240 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:06 PP3 TP1 EP1] Init torch distributed begin.
+I0827 04:29:06.736888   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:06.738317   238 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:02
+I0827 04:29:06.738422   238 transfer_engine_impl.cpp:388] installTransport, type=rdma
+I0827 04:29:06.738466   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+[2026-08-27 04:29:06 PP2 TP3 EP3] Init torch distributed begin.
+I0827 04:29:06.739869   235 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:02
+I0827 04:29:06.739982   235 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:06 PP2 TP0 EP0] Init torch distributed begin.
+[2026-08-27 04:29:26 PP3 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:26 PP2 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:27 PP2 TP3 EP3] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP2 TP1 EP1] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP2 TP2 EP2] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP2 TP0 EP0] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP3 TP3 EP3] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP3 TP0 EP0] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP3 TP2 EP2] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP3 TP1 EP1] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP3 TP3 EP3] Init torch distributed ends. elapsed=21.61 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP3 TP2 EP2] Init torch distributed ends. elapsed=21.62 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP3 TP1 EP1] Init torch distributed ends. elapsed=21.04 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP3 TP0 EP0] Init torch distributed ends. elapsed=21.61 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP2 TP2 EP2] Init torch distributed ends. elapsed=21.57 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP2 TP1 EP1] Init torch distributed ends. elapsed=21.55 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP2 TP0 EP0] Init torch distributed ends. elapsed=21.03 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP2 TP3 EP3] Init torch distributed ends. elapsed=21.04 s, mem usage=0.54 GB
+[2026-08-27 04:29:28 PP3 TP3 EP3] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP3 TP0 EP0] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP2 TP0 EP0] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP3 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP2 TP1 EP1] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP3 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP2 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP3 TP2 EP2] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP2 TP3 EP3] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP2 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP2 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP3 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP2 TP2 EP2] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP3 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP3 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP2 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP3 TP1 EP1] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP2 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP3 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP2 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP3 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP3 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP2 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP3 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP2 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP3 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP2 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP2 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP2 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP2 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP3 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP3 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:30:31 PP2 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:31 PP2 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:31 PP2 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:31 PP2 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:31 PP2 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:31 PP2 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:31 PP2 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:31 PP2 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP2 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:32 PP2 TP0 EP0] Load weight end. elapsed=64.92 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=36.14 GB, mem usage=45.91 GB.
+[2026-08-27 04:30:33 PP3 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:33 PP3 TP1 EP1] Load weight end. elapsed=65.31 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:34 PP2 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:34 PP2 TP1 EP1] Load weight end. elapsed=66.07 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=36.14 GB, mem usage=45.91 GB.
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO[TensorRT-LLM][INFO] Set logger level to INFO
+
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:35 PP3 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:35 PP3 TP0 EP0] Load weight end. elapsed=67.41 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:36 PP2 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:36 PP2 TP3 EP3] Load weight end. elapsed=68.04 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=36.14 GB, mem usage=45.91 GB.
+[2026-08-27 04:30:36 PP2 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:36 PP2 TP2 EP2] Load weight end. elapsed=68.23 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=36.14 GB, mem usage=45.91 GB.
+[2026-08-27 04:30:36 PP2 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:36 PP2 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:36 PP2 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:36 PP2 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:36 PP2 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:36 PP2 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:36 PP2 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:36 PP2 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:36 PP3 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:36 PP3 TP3 EP3] Load weight end. elapsed=68.86 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:37 PP3 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:37 PP3 TP2 EP2] Load weight end. elapsed=69.68 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:38 PP3 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:38 PP3 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:38 PP3 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:38 PP3 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:38 PP3 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:38 PP3 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:38 PP3 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:38 PP3 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:54 PP3 TP3 EP3] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP3 TP2 EP2] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP3 TP1 EP1] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP2 TP3 EP3] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP3 TP0 EP0] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP2 TP2 EP2] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP2 TP1 EP1] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP2 TP0 EP0] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP2 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:54 PP2 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:54 PP3 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP3 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP2 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:54 PP3 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP3 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP2 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:54 PP3 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP3 TP3 EP3] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP2 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP2 TP2 EP2] Memory pool end. avail mem=23.25 GB
+[2026-08-27 04:30:54 PP2 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP2 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP2 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP3 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP3 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP2 TP3 EP3] Memory pool end. avail mem=23.25 GB
+[2026-08-27 04:30:54 PP2 TP1 EP1] Memory pool end. avail mem=23.25 GB
+[2026-08-27 04:30:54 PP3 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP3 TP1 EP1] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP3 TP2 EP2] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP2 TP0 EP0] Memory pool end. avail mem=23.25 GB
+[2026-08-27 04:30:54 PP3 TP0 EP0] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP3 TP3 EP3] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp3_pp3_dp0.json
+[2026-08-27 04:30:54 PP2 TP1 EP1] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp1_pp2_dp0.json
+[2026-08-27 04:30:54 PP2 TP2 EP2] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp2_pp2_dp0.json
+[2026-08-27 04:30:54 PP3 TP1 EP1] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp1_pp3_dp0.json
+[2026-08-27 04:30:54 PP2 TP3 EP3] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp3_pp2_dp0.json
+[2026-08-27 04:30:54 PP3 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:30:54 PP3 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:30:54 PP3 TP2 EP2] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp2_pp3_dp0.json
+[2026-08-27 04:30:54 PP2 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:30:54 PP2 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:30:54 PP3 TP0 EP0] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp0_pp3_dp0.json
+[2026-08-27 04:30:54 PP2 TP0 EP0] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp0_pp2_dp0.json
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.65profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.60profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.79profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.51profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.86profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.27profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.62profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.98profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.86profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 12.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 13.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.93profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.92profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.09profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.63profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  6.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  7.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.33profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.58profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  8.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.56profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.23profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.63profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.36profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.09profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.97profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  8.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.30profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 13.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 13.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  8.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  8.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 11.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 11.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.23profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.48profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.23profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.53profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.27profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.59profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.62profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.97profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.60profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.46profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.12profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.68profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.77profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.62profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  6.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.87profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.30profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.40profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.83profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.94profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.59profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.66profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.29profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.23profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:01,  3.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.68profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.75profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  7.01profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.22profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.65profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.62profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.46profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.22profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  7.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:01<00:06,  1.03s/profile]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:01,  3.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.50profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.93profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.48profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.11profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.12profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.30profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.68profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  7.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 11.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  9.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 12.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 11.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 11.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 11.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.11profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.31profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.65profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.31profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.35profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.49profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:01<00:01,  2.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.65profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.68profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  7.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.09profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 11.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.88profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.98profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.50profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.62profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.51profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.91profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:01,  3.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.31profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.86profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.27profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.59profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.17profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.76profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.73profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  9.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.37profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.15profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.98profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.65profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.97profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.23profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.46profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  9.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.03profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.57profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.40profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.40profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.57profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.13profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.56profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  6.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.73profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.23profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  3.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.86profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.77profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.92profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.37profile/s]
+[2026-08-27 04:31:26 PP2 TP3 EP3] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.39profile/s]
+[2026-08-27 04:31:26 PP2 TP2 EP2] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.75profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.46profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.35profile/s]
+[2026-08-27 04:31:26 PP2 TP1 EP1] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.12profile/s]
+[2026-08-27 04:31:26 PP2 TP0 EP0] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.66profile/s][2026-08-27 04:31:26 PP2 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:26 PP2 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:26 PP2 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:26 PP2 TP0 EP0] max_total_num_tokens=3801408, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=54, context_len=1048576, available_gpu_mem=20.47 GB
+[2026-08-27 04:31:26 PP2 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:26 PP2 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:26 PP2 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:26 PP2 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:26 PP2 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.93profile/s]
+[2026-08-27 04:31:27 PP3 TP1 EP1] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.16profile/s]
+[2026-08-27 04:31:27 PP3 TP0 EP0] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.43profile/s]
+[2026-08-27 04:31:27 PP3 TP2 EP2] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.05profile/s]
+[2026-08-27 04:31:27 PP3 TP3 EP3] FlashInfer autotune completed.
+[2026-08-27 04:31:27 PP3 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:27 PP3 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:27 PP3 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:27 PP3 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:27 PP3 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:27 PP3 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:27 PP3 TP0 EP0] max_total_num_tokens=3801408, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=54, context_len=1048576, available_gpu_mem=16.54 GB
+[2026-08-27 04:31:27 PP3 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:27 PP3 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+W0827 04:31:30.590376  2521 memory_location.cpp:73] Failed to get NUMA node, addr: 0x513c5ec0, len: 6912: Operation not permitted [1]
+W0827 04:31:30.590432  2522 memory_location.cpp:73] Failed to get NUMA node, addr: 0x514a0900, len: 6912: Operation not permitted [1]
+W0827 04:31:30.590495  2525 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51226dc0, len: 55296: Operation not permitted [1]
+W0827 04:31:30.590492  2523 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5128db40, len: 6912: Operation not permitted [1]
+W0827 04:31:30.590533  2524 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5156ff80, len: 6912: Operation not permitted [1]
+W0827 04:31:30.590600  2526 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51320ac0, len: 55296: Operation not permitted [1]
+W0827 04:31:30.590636  2527 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51768540, len: 6912: Operation not permitted [1]
+W0827 04:31:30.590826  2528 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51483dc0, len: 13824: Operation not permitted [1]
+W0827 04:31:30.590914  2529 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51119380, len: 6912: Operation not permitted [1]
+W0827 04:31:30.590979  2530 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50fe1e00, len: 6912: Operation not permitted [1]
+W0827 04:31:30.846120  2547 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50ceebc0, len: 6912: Operation not permitted [1]
+W0827 04:31:30.846271  2548 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50c57640, len: 6912: Operation not permitted [1]
+W0827 04:31:30.846328  2549 memory_location.cpp:73] Failed to get NUMA node, addr: 0x502ed680, len: 6912: Operation not permitted [1]
+W0827 04:31:30.846354  2550 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50d1c800, len: 6912: Operation not permitted [1]
+W0827 04:31:30.846448  2551 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50510a40, len: 55296: Operation not permitted [1]
+W0827 04:31:30.846473  2552 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5051e280, len: 55296: Operation not permitted [1]
+W0827 04:31:30.846571  2553 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50d01340, len: 6912: Operation not permitted [1]
+W0827 04:31:30.846689  2554 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50ce3440, len: 13824: Operation not permitted [1]
+W0827 04:31:30.846858  2555 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5030f440, len: 6912: Operation not permitted [1]
+W0827 04:31:30.846926  2556 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50cae980, len: 6912: Operation not permitted [1]
+W0827 04:31:31.198441  2575 memory_location.cpp:73] Failed to get NUMA node, addr: 0x514f46c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.198501  2576 memory_location.cpp:73] Failed to get NUMA node, addr: 0x509c6400, len: 6912: Operation not permitted [1]
+W0827 04:31:31.198501  2577 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51513300, len: 6912: Operation not permitted [1]
+W0827 04:31:31.198679  2578 memory_location.cpp:73] Failed to get NUMA node, addr: 0x506eef40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.198787  2579 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50dc9180, len: 55296: Operation not permitted [1]
+W0827 04:31:31.198915  2580 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50904700, len: 55296: Operation not permitted [1]
+W0827 04:31:31.198982  2581 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50aa9440, len: 6912: Operation not permitted [1]
+W0827 04:31:31.199081  2582 memory_location.cpp:73] Failed to get NUMA node, addr: 0x515012c0, len: 13824: Operation not permitted [1]
+W0827 04:31:31.199148  2583 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5161ac80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.199185  2584 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50dd69c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.763885  2603 memory_location.cpp:73] Failed to get NUMA node, addr: 0x519f1500, len: 6912: Operation not permitted [1]
+W0827 04:31:31.763998  2605 memory_location.cpp:73] Failed to get NUMA node, addr: 0x517df940, len: 6912: Operation not permitted [1]
+W0827 04:31:31.763923  2604 memory_location.cpp:73] Failed to get NUMA node, addr: 0x514ab440, len: 6912: Operation not permitted [1]
+W0827 04:31:31.764139  2606 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51556180, len: 6912: Operation not permitted [1]
+W0827 04:31:31.764169  2608 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50ed8140, len: 55296: Operation not permitted [1]
+W0827 04:31:31.764184  2607 memory_location.cpp:73] Failed to get NUMA node, addr: 0x515df040, len: 55296: Operation not permitted [1]
+W0827 04:31:31.764356  2609 memory_location.cpp:73] Failed to get NUMA node, addr: 0x517e3580, len: 6912: Operation not permitted [1]
+W0827 04:31:31.764564  2610 memory_location.cpp:73] Failed to get NUMA node, addr: 0x519cdb80, len: 13824: Operation not permitted [1]
+W0827 04:31:31.764624  2611 memory_location.cpp:73] Failed to get NUMA node, addr: 0x519dbf80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.764714  2612 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50f08d80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.868036  2629 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50a7a680, len: 6912: Operation not permitted [1]
+W0827 04:31:31.868046  2630 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50b6e400, len: 6912: Operation not permitted [1]
+W0827 04:31:31.868139  2631 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50e5f580, len: 6912: Operation not permitted [1]
+W0827 04:31:31.868261  2632 memory_location.cpp:73] Failed to get NUMA node, addr: 0x503c7100, len: 6912: Operation not permitted [1]
+W0827 04:31:31.868330  2633 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50c94ac0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.868432  2635 memory_location.cpp:73] Failed to get NUMA node, addr: 0x509b2880, len: 6912: Operation not permitted [1]
+W0827 04:31:31.868502  2634 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50b14c80, len: 55296: Operation not permitted [1]
+W0827 04:31:31.868564  2636 memory_location.cpp:73] Failed to get NUMA node, addr: 0x503ee840, len: 13824: Operation not permitted [1]
+W0827 04:31:31.868708  2637 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51057d80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.868835  2638 memory_location.cpp:73] Failed to get NUMA node, addr: 0x509cd940, len: 6912: Operation not permitted [1]
+W0827 04:31:32.003983  2656 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50715b00, len: 6912: Operation not permitted [1]
+W0827 04:31:32.004096  2657 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51346e80, len: 6912: Operation not permitted [1]
+W0827 04:31:32.004005  2655 memory_location.cpp:73] Failed to get NUMA node, addr: 0x507176c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.004179  2658 memory_location.cpp:73] Failed to get NUMA node, addr: 0x507e3a40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.004302  2659 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5078aa80, len: 55296: Operation not permitted [1]
+W0827 04:31:32.004357  2660 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5135c980, len: 55296: Operation not permitted [1]
+W0827 04:31:32.004410  2661 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5099f180, len: 6912: Operation not permitted [1]
+W0827 04:31:32.004585  2662 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50b287c0, len: 13824: Operation not permitted [1]
+W0827 04:31:32.004739  2663 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5070b6c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.004812  2664 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50753540, len: 6912: Operation not permitted [1]
+W0827 04:31:32.144867  2688 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50aed700, len: 6912: Operation not permitted [1]
+W0827 04:31:32.144946  2689 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50792740, len: 6912: Operation not permitted [1]
+W0827 04:31:32.145030  2690 memory_location.cpp:73] Failed to get NUMA node, addr: 0x508fca40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.145135  2691 memory_location.cpp:73] Failed to get NUMA node, addr: 0x505d7c40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.145318  2693 memory_location.cpp:73] Failed to get NUMA node, addr: 0x505ec340, len: 55296: Operation not permitted [1]
+W0827 04:31:32.145387  2692 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50602cc0, len: 55296: Operation not permitted [1]
+W0827 04:31:32.145479  2694 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50748c00, len: 6912: Operation not permitted [1]
+W0827 04:31:32.145632  2695 memory_location.cpp:73] Failed to get NUMA node, addr: 0x507e3480, len: 13824: Operation not permitted [1]
+W0827 04:31:32.145767  2696 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50316400, len: 6912: Operation not permitted [1]
+W0827 04:31:32.145864  2697 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50788600, len: 6912: Operation not permitted [1]
+W0827 04:31:32.177534  2726 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fff3c80, len: 6912: Operation not permitted [1]
+W0827 04:31:32.177623  2728 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fad6780, len: 6912: Operation not permitted [1]
+W0827 04:31:32.177625  2727 memory_location.cpp:73] Failed to get NUMA node, addr: 0x501dc2c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.178279  2729 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50596480, len: 6912: Operation not permitted [1]
+W0827 04:31:32.178463  2730 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50222dc0, len: 55296: Operation not permitted [1]
+W0827 04:31:32.178498  2731 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5038f6c0, len: 55296: Operation not permitted [1]
+W0827 04:31:32.178581  2732 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4ffd7ec0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.178695  2733 memory_location.cpp:73] Failed to get NUMA node, addr: 0x506d84c0, len: 13824: Operation not permitted [1]
+W0827 04:31:32.178884  2734 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5078efc0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.178922  2735 memory_location.cpp:73] Failed to get NUMA node, addr: 0x502c7b00, len: 6912: Operation not permitted [1]
+[2026-08-27 04:31:32] Dummy health check server started in background thread at 0.0.0.0:30000
+[rank10]:[W827 04:32:32.605391940 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank11]:[W827 04:32:32.605815131 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank9]:[W827 04:32:32.606041691 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank8]:[W827 04:32:32.608313727 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank11]:[W827 04:33:19.698792619 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank8]:[W827 04:33:19.699392665 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank9]:[W827 04:33:19.699501937 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank10]:[W827 04:33:19.699589651 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank14]:[W827 04:33:19.728965278 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank12]:[W827 04:33:19.729026815 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank13]:[W827 04:33:19.729092243 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank15]:[W827 04:33:19.730463091 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank14]:[W827 04:33:25.763484084 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank13]:[W827 04:33:25.764518868 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank15]:[W827 04:33:25.764776298 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank12]:[W827 04:33:25.765271843 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[2026-08-27 04:35:10 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.29
+[2026-08-27 04:35:10 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.29
+[2026-08-27 04:35:29 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.39
+[2026-08-27 04:35:29 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.39
+[2026-08-27 04:35:31 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25.86
+[2026-08-27 04:35:31 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25.87
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/d_starting_p_node2.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/d_starting_p_node2.log
new file mode 100644
index 0000000..c8152e8
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/d_starting_p_node2.log
@@ -0,0 +1,754 @@
+/usr/local/lib/python3.12/dist-packages/torchao/quantization/quant_api.py:1731: SyntaxWarning: invalid escape sequence '\.'
+  """Configuration class for applying different quantization configs to modules or parameters based on their fully qualified names (FQNs).
+/sgl-workspace/sglang/python/sglang/launch_server.py:56: UserWarning: 'python -m sglang.launch_server' is still supported, but 'sglang serve' is the recommended entrypoint.
+  Example: sglang serve --model-path <model> [options]
+  warnings.warn(
+[2026-08-27 04:28:43] Breakable CUDA graph is incompatible with MLA attention; disabling prefill CUDA graph.
+[2026-08-27 04:28:43] Attention backend not specified. Use triton backend by default.
+[2026-08-27 04:28:43] Pipeline parallelism is incompatible with overlap schedule.
+[2026-08-27 04:28:45] server_args=ServerArgs(model_path='/data/hf_models/Kimi-K3', tokenizer_path='/data/hf_models/Kimi-K3', tokenizer_mode='auto', tokenizer_backend='huggingface', tokenizer_worker_num=1, detokenizer_worker_num=1, skip_tokenizer_init=False, load_format='auto', model_loader_extra_config='{}', trust_remote_code=True, context_length=None, is_embedding=False, enable_multimodal=None, revision=None, model_impl='auto', model_config_parser='auto', json_model_override_args='{}', dtype='auto', quantization=None, quantization_param_path=None, kv_cache_dtype='auto', enable_fp32_lm_head=False, modelopt_quant=None, modelopt_checkpoint_restore_path=None, modelopt_checkpoint_save_path=None, modelopt_export_path=None, quantize_and_serve=False, rl_quant_profile=None, enable_tf32_matmul=False, mem_fraction_static=0.88, max_running_requests=None, max_queued_requests=None, max_total_tokens=None, chunked_prefill_size=8192, enable_dynamic_chunking=False, max_prefill_tokens=16384, prefill_max_requests=None, schedule_policy='fcfs', enable_priority_scheduling=False, disable_priority_preemption=False, default_priority_value=None, abort_on_priority_when_disabled=False, schedule_low_priority_values_first=False, priority_scheduling_preemption_threshold=10, retraction_policy='length', schedule_conservativeness=1.0, page_size=64, swa_full_tokens_ratio=0.8, disable_hybrid_swa_memory=False, radix_eviction_policy='lru', prefill_only_disable_kv_cache=False, disable_radix_cache=True, enable_page_major_kv_layout=False, enable_unified_memory=False, disable_chunked_prefix_cache=False, disable_overlap_schedule=True, num_continuous_decode_steps=1, scheduler_recv_interval=1, enable_mixed_chunk=False, nccl_port=None, dist_timeout=3600, dist_init_addr='174.1.60.1:20000', nnodes=4, node_rank=2, tp_size=4, dcp_size=1, pp_size=8, pp_max_micro_batch_size=None, pp_async_batch_depth=0, dp_size=1, load_balance_method='follow_bootstrap_room', attn_cp_size=1, moe_dp_size=1, dcp_comm_backend='ag_rs', dcp_replicate_q_proj=None, dwdp_size=1, enable_prefill_cp=False, cp_strategy=None, enable_dsa_cache_layer_split=False, enable_dsa_prefill_context_parallel=False, dsa_prefill_cp_mode='round-robin-split', enable_prefill_context_parallel=False, prefill_cp_mode='in-seq-split', enable_cp_decode_attn_tp=False, enable_dp_attention=False, enable_dp_attention_local_control_broadcast=False, enable_dp_lm_head=False, enable_attn_tp_input_scattered=False, disable_attn_tp_gather=False, enable_p2p_check=False, device='cuda', base_gpu_id=0, gpu_id_step=1, random_seed=336932344, watchdog_timeout=300, soft_watchdog_timeout=None, sleep_on_idle=False, use_ray=False, custom_sigquit_handler=None, numa_node=None, gc_threshold=None, host='0.0.0.0', port=30000, fastapi_root_path='', smg_grpc_mode=False, grpc_mode=False, grpc_port=None, sidecar=None, sidecar_args=None, skip_server_warmup=False, warmups=None, enable_http2=False, ssl_keyfile=None, ssl_certfile=None, ssl_ca_certs=None, ssl_keyfile_password=None, enable_ssl_refresh=False, api_key=None, admin_api_key=None, served_model_name='kimi-k3', weight_version='default', chat_template=None, hf_chat_template_name=None, completion_template=None, file_storage_path='sglang_storage', enable_cache_report=False, reasoning_parser=None, default_chat_template_kwargs=None, strip_thinking_cache=False, enable_strict_thinking=False, tool_call_parser=None, tool_server=None, sampling_defaults='model', asr_max_buffer_seconds=60, asr_max_concurrent_sessions=32, preferred_sampling_params=None, allow_auto_truncate=False, stream_interval=1, batch_notify_size=16, stream_response_default_include_usage=False, incremental_streaming_output=False, enable_streaming_session=False, enable_session_radix_cache=False, log_level='info', log_level_http=None, log_requests=False, log_requests_level=2, log_requests_format='text', log_requests_target=None, uvicorn_access_log_exclude_prefixes=[], crash_dump_folder=None, show_time_cost=False, enable_metrics=False, smg_http_sidecar_port=None, enable_mfu_metrics=False, enable_metrics_for_all_schedulers=False, load_snapshot_publish_interval=15, tokenizer_metrics_custom_labels_header='x-custom-labels', tokenizer_metrics_allowed_custom_labels=None, extra_metric_labels=None, bucket_time_to_first_token=None, bucket_inter_token_latency=None, bucket_e2e_request_latency=None, prompt_tokens_buckets=None, generation_tokens_buckets=None, gc_warning_threshold_secs=0.0, decode_log_interval=40, enable_request_time_stats_logging=False, kv_events_config=None, enable_forward_pass_metrics=False, forward_pass_metrics_worker_id='', forward_pass_metrics_ipc_name=None, enable_trace=False, trace_modules='request', otlp_traces_endpoint='localhost:4317', export_metrics_to_file=False, export_metrics_to_file_dir=None, stat_loggers=None, constrained_json_whitespace_pattern=None, constrained_json_disable_any_whitespace=False, attention_backend='triton', decode_attention_backend=None, prefill_attention_backend=None, sampling_backend='flashinfer', grammar_backend='xgrammar', radix_cache_backend=None, mm_attention_backend=None, fp8_gemm_runner_backend='auto', fp4_gemm_runner_backend='auto', bf16_gemm_backend='auto', dsa_prefill_backend=None, dsa_decode_backend=None, dsa_paged_mqa_logits_backend='auto', dsa_topk_backend='sgl-kernel', disable_flashinfer_autotune=False, flashinfer_autotune_skip_ops=None, mamba_backend='triton', cuda_graph_config=CudaGraphConfig(decode=PhaseConfig(backend='disabled', max_bs=16, bs=[1, 2, 4, 8, 12, 16], tc_compiler='eager', full_prefill_max_req=None), prefill=PhaseConfig(backend='disabled', max_bs=2048, bs=[4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], tc_compiler='eager', full_prefill_max_req=None)), cuda_graph_backend_decode=None, cuda_graph_backend_prefill=None, cuda_graph_max_bs_decode=16, cuda_graph_max_bs_prefill=None, cuda_graph_bs_decode=None, cuda_graph_bs_prefill=None, cuda_graph_tc_compiler=None, disable_prefill_cuda_graph=False, disable_decode_cuda_graph=False, disable_cuda_graph=False, disable_cuda_graph_padding=False, enable_profile_cuda_graph=False, enable_cudagraph_gc=False, debug_cuda_graph=False, enable_layerwise_nvtx_marker=False, enable_nccl_nvls=False, enable_symm_mem=False, triton_attention_reduce_in_fp32=False, triton_attention_num_kv_splits=8, triton_attention_split_tile_size=None, flashinfer_mla_disable_ragged=False, enable_fused_qk_norm_rope=False, enable_precise_embedding_interpolation=False, enable_fused_moe_sum_all_reduce=False, enable_deepseek_v4_fp4_indexer=False, disable_custom_all_reduce=False, enable_mscclpp=False, enable_torch_symm_mem=False, enable_scattered_sconv=False, pre_warm_nccl=False, enable_quant_communications=False, enable_flashinfer_allreduce_fusion=False, enforce_disable_flashinfer_allreduce_fusion=False, flashinfer_allreduce_fusion_backend=None, enable_aiter_allreduce_fusion=False, enable_torch_compile=False, enable_torch_compile_debug_mode=False, torch_compile_max_bs=32, torchao_config='', speculative_algorithm=None, speculative_draft_model_path=None, speculative_draft_model_revision=None, speculative_draft_load_format=None, speculative_num_steps=None, speculative_eagle_topk=None, speculative_num_draft_tokens=None, speculative_dflash_block_size=None, speculative_dspark_block_size=None, speculative_dspark_sps_table_path=None, speculative_dspark_confidence_sts_path=None, speculative_dspark_align_verify_tokens_to_graph_tier=False, speculative_accept_threshold_single=1.0, speculative_accept_threshold_acc=1.0, speculative_use_rejection_sampling=False, speculative_token_map=None, speculative_attention_mode='prefill', speculative_draft_attention_backend=None, speculative_draft_window_size=None, speculative_moe_runner_backend='flashinfer_mxfp4', speculative_moe_a2a_backend=None, speculative_draft_model_quantization=None, speculative_skip_dp_mlp_sync=False, enable_multi_layer_eagle=False, speculative_adaptive=False, speculative_adaptive_config=None, decoupled_spec_bind_endpoint=None, decoupled_spec_connect_endpoints=None, decoupled_spec_rank=None, decoupled_spec_role='null', spec_trace_dir=None, speculative_ngram_min_bfs_breadth=1, speculative_ngram_max_bfs_breadth=10, speculative_ngram_match_type='BFS', speculative_ngram_max_trie_depth=18, speculative_ngram_capacity=10000000, speculative_ngram_external_corpus_path=None, speculative_ngram_external_sam_budget=0, speculative_ngram_external_corpus_max_tokens=10000000, ep_size=4, moe_a2a_backend='none', moe_runner_backend='flashinfer_mxfp4', flashinfer_mxfp4_moe_precision='default', deepep_mode='auto', fuseep_mode=2, deepep_dispatcher_output_dtype='auto', ep_num_redundant_experts=0, ep_dispatch_algorithm=None, init_expert_location='trivial', enable_eplb=False, eplb_algorithm='auto', eplb_rebalance_num_iterations=1000, eplb_rebalance_layers_per_chunk=None, eplb_min_rebalancing_utilization_threshold=1.0, expert_distribution_recorder_mode=None, expert_distribution_recorder_buffer_size=1000, enable_expert_distribution_metrics=False, deepep_config=None, moe_dense_tp_size=None, elastic_ep_backend=None, enable_elastic_expert_backup=False, mooncake_ib_device=None, enable_waterfill=False, ep_join_mode=None, ep_join_rank_offset=0, elastic_ep_initial_size=None, max_ep_size=None, elastic_ep_scale_timeout=600, elastic_ep_rejoin=False, disable_flashinfer_cutlass_moe_fp4_allgather=False, disable_shared_experts_fusion=False, enforce_shared_experts_fusion=False, max_mamba_cache_size=None, mamba_ssm_dtype=None, mamba_max_states_per_path=-1, enable_mamba_cache_stochastic_rounding=False, mamba_cache_philox_rounds=0, mamba_full_memory_ratio=0.36, mamba_radix_cache_strategy='extra_buffer_lazy', uses_mamba_radix_cache=False, mamba_track_interval=256, enable_int8_mamba_checkpoint=False, int8_mamba_ckpt_size=None, linear_attn_backend='triton', linear_attn_decode_backend=None, linear_attn_prefill_backend=None, linear_attn_verify_backend=None, enable_linear_replayssm=False, linear_replayssm_cache_len=16, enable_linear_replayssm_spec=False, enable_hierarchical_cache=False, hicache_ratio=2.0, hicache_size=0, hicache_write_policy='write_through', hicache_io_backend='kernel', hicache_mem_layout='page_first', hicache_storage_backend=None, hicache_storage_prefetch_policy='timeout', hicache_storage_backend_extra_config=None, enable_hisparse=False, hisparse_config=None, enable_broadcast_mm_inputs_process=False, enable_prefix_mm_cache=False, mm_enable_dp_encoder=False, mm_process_config={}, mm_processor_worker_num=0, mm_io_worker_num=0, limit_mm_data_per_request=None, enable_mm_global_cache=False, disable_fast_image_processor=False, mm_feature_transport='cpu', keep_mm_feature_on_device=False, enable_lora=None, enable_lora_overlap_loading=None, max_lora_rank=None, lora_target_modules=None, lora_paths=None, max_loaded_loras=None, max_loras_per_batch=8, lora_eviction_policy='lru', lora_backend='csgmv', max_lora_chunk_size=16, experts_shared_outer_loras=None, lora_use_virtual_experts=False, lora_strict_loading=False, lora_drain_wait_threshold=0.0, enable_two_batch_overlap=False, enable_single_batch_overlap=False, tbo_token_distribution_threshold=0.48, cpu_offload_gb=0, offload_group_size=-1, offload_num_in_group=1, offload_prefetch_step=1, offload_mode='cpu', enable_lmcache=False, lmcache_config_file=None, enable_flexkv=False, flexkv_config_file=None, kt_weight_path=None, kt_method='AMXINT4', kt_cpuinfer=None, kt_threadpool_count=2, kt_num_gpu_experts=None, kt_max_deferred_experts_per_token=None, dllm_algorithm=None, dllm_algorithm_config=None, dllm_fdfo=True, disaggregation_mode='prefill', disaggregation_transfer_backend='mooncake', disaggregation_bootstrap_port=28800, disaggregation_ib_device='mlx5_0,mlx5_1,mlx5_2,mlx5_3', disaggregation_decode_enable_radix_cache=False, disaggregation_decode_enable_offload_kvcache=False, num_reserved_decode_tokens=512, disaggregation_decode_extra_slots=None, disaggregation_decode_polling_interval=1, optimistic_prefill_attempts=0, encoder_only=False, language_only=False, encoder_transfer_backend='zmq_to_tokenizer', encoder_urls=[], encoder_bootstrap_port=8997, encoder_register_urls=[], enable_adaptive_dispatch_to_encoder=False, enable_pdmux=False, pdmux_config_path=None, sm_group_num=8, custom_weight_loader=[], weight_loader_disable_mmap=False, weight_loader_prefetch_checkpoints=False, weight_loader_prefetch_num_threads=4, weight_loader_drop_cache_after_load=False, remote_instance_weight_loader_seed_instance_ip=None, remote_instance_weight_loader_seed_instance_service_port=None, remote_instance_weight_loader_send_weights_group_ports=None, remote_instance_weight_loader_backend='nccl', remote_instance_weight_loader_start_seed_via_transfer_engine=False, engine_info_bootstrap_port=6789, modelexpress_config=None, download_dir=None, model_checksum=None, delete_ckpt_after_loading=False, decrypted_config_file=None, decrypted_draft_config_file=None, checkpoint_engine_wait_weights_before_ready=False, enable_prefill_delayer=False, prefill_delayer_max_delay_passes=30, prefill_delayer_token_usage_low_watermark=None, prefill_delayer_forward_passes_buckets=None, prefill_delayer_wait_seconds_buckets=None, prefill_delayer_queue_min_ratio=None, prefill_delayer_max_delay_ms=None, min_free_slots_delay=None, enable_deterministic_inference=False, rl_on_policy_target=None, kv_canary='none', kv_canary_real_data='none', kv_canary_sweep_interval=0, enable_dynamic_batch_tokenizer=False, dynamic_batch_tokenizer_batch_size=32, dynamic_batch_tokenizer_batch_timeout=0.002, enable_tokenizer_batch_encode=False, disable_tokenizer_batch_decode=False, debug_tensor_dump_output_folder=None, debug_tensor_dump_layers=None, debug_tensor_dump_input_file=None, enable_memory_saver=False, enable_weights_cpu_backup=False, enable_draft_weights_cpu_backup=False, enable_custom_logit_processor=False, enable_return_hidden_states=False, enable_return_routed_experts=False, enable_return_indexer_topk=False, disable_outlines_disk_cache=False, enable_mis=False, weight_cache_mode='off', weight_cache_socket=None, weight_cache_timeout=1800, forward_hooks=None, msprobe_dump_config=None)
+[2026-08-27 04:28:45] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:45] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:45] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:45] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:45] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:45] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:45] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:45] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:01 PP4 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:01 PP4 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:01 PP5 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:01 PP5 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:01 PP4 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:01 PP4 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:01 PP5 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:01 PP5 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:01 PP4 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:01 PP4 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:01 PP4 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:01 PP4 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:02 PP5 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:02 PP5 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:02 PP5 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:02 PP5 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:03 PP5 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:03 PP5 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:03 PP4 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:03 PP4 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:03 PP5 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:03 PP5 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:03 PP4 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:03 PP4 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:05.195727   240 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:05.195828   240 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:05.195843   240 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.3 port: 12001
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:05.195847   236 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:05.195899   240 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.3:15652
+I0827 04:29:05.195935   236 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:05.195948   236 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.3 port: 12001
+I0827 04:29:05.195989   236 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.3:15032
+I0827 04:29:05.196039   240 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:05.196110   236 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:05.206050   241 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:05.206164   241 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:05.206182   241 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.3 port: 12001
+I0827 04:29:05.206233   241 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.3:16487
+I0827 04:29:05.206414   241 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:05.206894   238 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:05.207010   238 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:05.207029   238 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.3 port: 12001
+I0827 04:29:05.207082   238 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.3:15213
+I0827 04:29:05.207248   238 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:05.213657   238 topology.cpp:127] Device mlx5_0 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:05.215921   237 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:05.216045   237 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:05.216063   237 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.3 port: 12001
+I0827 04:29:05.216125   237 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.3:16450
+I0827 04:29:05.216308   237 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:05.219630   235 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:05.219745   235 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:05.219763   235 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.3 port: 12001
+I0827 04:29:05.219815   235 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.3:15658
+I0827 04:29:05.219810   236 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:05.220000   235 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:05.222787   237 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:05.223301   240 topology.cpp:127] Device mlx5_0 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:05.223796   239 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+W0827 04:29:05.223798   242 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:05.223948   242 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:05.223948   239 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:05.223969   242 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.3 port: 12001
+I0827 04:29:05.223973   239 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.3 port: 12001
+I0827 04:29:05.224030   242 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.3:16182
+I0827 04:29:05.224030   239 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.3:15581
+I0827 04:29:05.224237   242 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:05.224242   239 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:05.226227   235 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:05.227162   236 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:05.228888   237 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:05.229261   240 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:05.232141   235 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:05.232787   239 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:05.232861   242 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:05.233434   241 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:05.233938   236 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:05.236400   237 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:05.236641   240 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:05.238986   235 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:05.241036   236 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:05.241355   238 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:05.241638   241 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:05.242061   239 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:05.242151   242 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:05.244179   236 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:05.244223   236 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:05.244287   237 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:05.244297   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.244899   240 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:05.245494   235 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:05.249099   237 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:05.249130   237 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:05.249187   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.251125   235 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:05.251137   240 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:05.251168   235 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:05.251168   240 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:05.251241   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.251241   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.251624   239 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:05.251729   242 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:05.251823   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:05.253651   236 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:03
+I0827 04:29:05.253687   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.256857   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:05.257697   239 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:05.258617   237 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:03
+I0827 04:29:05.258652   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.258757   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:05.258822   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:05.260108   239 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:05.260136   239 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:05.260191   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.260488   235 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:03
+I0827 04:29:05.260502   240 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:03
+I0827 04:29:05.260530   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.260533   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.262187   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:05.264094   236 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:03
+I0827 04:29:05.264137   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.266639   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:05.268234   237 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:03
+I0827 04:29:05.268261   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.268549   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:05.269187   241 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:05.269232   238 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:05.270207   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:05.270285   240 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:03
+I0827 04:29:05.270316   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.271827   235 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:03
+I0827 04:29:05.271884   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.275274   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:05.276836   238 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:05.276907   242 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:05.276996   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:05.277058   236 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:03
+I0827 04:29:05.277101   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.278084   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:05.278632   237 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:03
+I0827 04:29:05.278667   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.279789   240 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:03
+I0827 04:29:05.279817   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.282152   238 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:05.282193   238 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:05.282260   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.283128   242 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:05.283200   242 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:05.283263   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.286593   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:05.287469   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:05.288283   239 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:03
+I0827 04:29:05.288322   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.289064   240 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:03
+I0827 04:29:05.289194   240 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:05 PP5 TP1 EP1] Init torch distributed begin.
+I0827 04:29:05.290841   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:05.292457   242 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:03
+I0827 04:29:05.292488   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.292999   241 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:05.296203   241 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:05.296272   241 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:05.296341   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.300206   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:05.301280   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:05.301775   242 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:03
+I0827 04:29:05.301801   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.303004   235 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:03
+I0827 04:29:05.303042   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.303999   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:05.305142   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:05.305728   241 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:03
+I0827 04:29:05.305768   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.305974   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:05.306758   237 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:03
+I0827 04:29:05.306921   237 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:05 PP4 TP2 EP2] Init torch distributed begin.
+I0827 04:29:05.307497   236 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:03
+I0827 04:29:05.307627   236 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:05 PP4 TP1 EP1] Init torch distributed begin.
+I0827 04:29:05.309624   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:05.309773   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:05.311504   238 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:03
+I0827 04:29:05.311542   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.311590   242 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:03
+I0827 04:29:05.311633   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.313076   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:05.314762   239 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:03
+I0827 04:29:05.314813   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.319522   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:05.319566   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:05.321249   242 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:03
+I0827 04:29:05.321398   242 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:05 PP5 TP3 EP3] Init torch distributed begin.
+I0827 04:29:05.326727   238 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:03
+I0827 04:29:05.326759   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.329339   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:05.331007   235 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:03
+I0827 04:29:05.331142   235 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:05 PP4 TP0 EP0] Init torch distributed begin.
+I0827 04:29:05.334666   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:05.336285   238 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:03
+I0827 04:29:05.336313   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.337296   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:05.338828   241 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:03
+I0827 04:29:05.338877   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.344906   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:05.346731   239 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:03
+I0827 04:29:05.346776   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.365299   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:05.366928   238 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:03
+I0827 04:29:05.367062   238 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:05 PP4 TP3 EP3] Init torch distributed begin.
+I0827 04:29:05.368464   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:05.369644   241 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:03
+I0827 04:29:05.369663   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.377357   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:05.379019   239 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:03
+I0827 04:29:05.379146   239 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:05 PP5 TP0 EP0] Init torch distributed begin.
+I0827 04:29:05.396831   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:05.398531   241 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:03
+I0827 04:29:05.398662   241 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:05 PP5 TP2 EP2] Init torch distributed begin.
+[2026-08-27 04:29:26 PP5 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:26 PP4 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:27 PP4 TP1 EP1] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP4 TP3 EP3] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP4 TP2 EP2] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP4 TP0 EP0] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP5 TP2 EP2] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP5 TP3 EP3] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP5 TP0 EP0] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP5 TP1 EP1] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP5 TP0 EP0] Init torch distributed ends. elapsed=22.37 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP5 TP3 EP3] Init torch distributed ends. elapsed=22.42 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP5 TP2 EP2] Init torch distributed ends. elapsed=22.35 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP4 TP1 EP1] Init torch distributed ends. elapsed=22.44 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP5 TP1 EP1] Init torch distributed ends. elapsed=22.46 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP4 TP3 EP3] Init torch distributed ends. elapsed=22.38 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP4 TP2 EP2] Init torch distributed ends. elapsed=22.44 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP4 TP0 EP0] Init torch distributed ends. elapsed=22.41 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP5 TP0 EP0] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP4 TP1 EP1] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP4 TP3 EP3] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP4 TP2 EP2] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP5 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP4 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP4 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP5 TP2 EP2] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP4 TP0 EP0] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP4 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP5 TP3 EP3] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP5 TP1 EP1] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP4 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP5 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP5 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP5 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP4 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP5 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP4 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP5 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP5 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP4 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP4 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP5 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP5 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP4 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP4 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP4 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP5 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP5 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP4 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP5 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:30:33 PP4 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:33 PP4 TP2 EP2] Load weight end. elapsed=65.50 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:34 PP4 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:34 PP4 TP3 EP3] Load weight end. elapsed=66.61 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP4 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:36 PP4 TP1 EP1] Load weight end. elapsed=68.45 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:36 PP5 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:36 PP5 TP0 EP0] Load weight end. elapsed=69.00 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:37 PP5 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:37 PP5 TP3 EP3] Load weight end. elapsed=69.63 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:37 PP5 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:37 PP5 TP1 EP1] Load weight end. elapsed=69.72 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:38 PP5 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:38 PP5 TP2 EP2] Load weight end. elapsed=70.08 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:38 PP5 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:38 PP5 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:38 PP5 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:38 PP5 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:38 PP5 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:38 PP5 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:38 PP5 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:38 PP5 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:47 PP4 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:47 PP4 TP0 EP0] Load weight end. elapsed=79.31 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:47 PP4 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:47 PP4 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:47 PP4 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:47 PP4 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:47 PP4 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:47 PP4 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:47 PP4 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:47 PP4 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:54 PP5 TP3 EP3] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP5 TP1 EP1] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP5 TP2 EP2] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP5 TP0 EP0] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP4 TP3 EP3] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP4 TP2 EP2] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP4 TP1 EP1] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP4 TP0 EP0] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP5 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP5 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP4 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP4 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP4 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP5 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP4 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP5 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP5 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP5 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP4 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP5 TP3 EP3] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP4 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP4 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP4 TP1 EP1] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP5 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP5 TP2 EP2] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP4 TP3 EP3] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP4 TP0 EP0] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP5 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP5 TP1 EP1] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP5 TP0 EP0] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP4 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP4 TP2 EP2] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP4 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:30:54 PP4 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:30:54 PP4 TP0 EP0] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp0_pp4_dp0.json
+[2026-08-27 04:30:54 PP4 TP1 EP1] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp1_pp4_dp0.json
+[2026-08-27 04:30:54 PP5 TP2 EP2] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp2_pp5_dp0.json
+[2026-08-27 04:30:54 PP5 TP3 EP3] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp3_pp5_dp0.json
+[2026-08-27 04:30:54 PP5 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:30:54 PP5 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:30:54 PP5 TP0 EP0] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp0_pp5_dp0.json
+[2026-08-27 04:30:54 PP4 TP2 EP2] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp2_pp4_dp0.json
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO[TensorRT-LLM][INFO] Set logger level to INFO
+
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[2026-08-27 04:30:54 PP4 TP3 EP3] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp3_pp4_dp0.json
+[2026-08-27 04:30:54 PP5 TP1 EP1] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp1_pp5_dp0.json
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.37profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.12profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.87profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.32profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.91profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.77profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.94profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.88profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.20profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.87profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  7.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 10.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.46profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.67profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.91profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.29profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.49profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  7.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.30profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.94profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.47profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.12profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.06profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.35profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  9.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:01<00:02,  2.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.89profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.58profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.32profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.70profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.20profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.27profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  6.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  7.03profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.56profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.73profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.88profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.37profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:05,  1.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:01<00:02,  2.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.05profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.19profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.77profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.18profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.25profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  3.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.18profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.20profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.37profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  9.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.27profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  8.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.43profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.89profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.35profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.96profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.90profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.30profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.72profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.18profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.33profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.51profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  2.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.03profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.68profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.16profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.01profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.24profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 10.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:02,  2.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.06profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.83profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.35profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.87profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.19profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  3.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.06profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.86profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 13.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 13.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.13profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:01,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:01,  3.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.47profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  6.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.47profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.57profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.22profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.17profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.50profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.71profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.19profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.55profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.90profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:05,  1.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:01<00:06,  1.00s/profile]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.33profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.19profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.03profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.76profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.28profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.32profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.15profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.79profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.31profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.18profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 11.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:01,  3.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.06profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:01,  3.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  6.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  3.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.05profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.60profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.50profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.92profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.00profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.29profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 11.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.09profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.43profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.11profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.91profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  7.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:02,  2.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.18profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.78profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.75profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.89profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.64profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.05profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.47profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.47profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.65profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:05,  1.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.85profile/s]
+[2026-08-27 04:31:28 PP5 TP1 EP1] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.21profile/s]
+[2026-08-27 04:31:28 PP5 TP0 EP0] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.74profile/s]
+[2026-08-27 04:31:28 PP5 TP2 EP2] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.77profile/s]
+[2026-08-27 04:31:28 PP5 TP3 EP3] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.19profile/s][2026-08-27 04:31:28 PP5 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:28 PP5 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:28 PP5 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:28 PP5 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:28 PP5 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:28 PP5 TP0 EP0] max_total_num_tokens=3801408, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=54, context_len=1048576, available_gpu_mem=16.30 GB
+[2026-08-27 04:31:28 PP5 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:28 PP5 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:28 PP5 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.87profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.90profile/s]
+[2026-08-27 04:31:29 PP4 TP3 EP3] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.87profile/s]
+[2026-08-27 04:31:29 PP4 TP2 EP2] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.93profile/s]
+[2026-08-27 04:31:29 PP4 TP0 EP0] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.43profile/s]
+[2026-08-27 04:31:29 PP4 TP1 EP1] FlashInfer autotune completed.
+[2026-08-27 04:31:29 PP4 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:29 PP4 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:29 PP4 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:29 PP4 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:29 PP4 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:29 PP4 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:29 PP4 TP0 EP0] max_total_num_tokens=3801408, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=54, context_len=1048576, available_gpu_mem=16.30 GB
+[2026-08-27 04:31:29 PP4 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:29 PP4 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+W0827 04:31:30.513926  2527 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50a93b80, len: 6912: Operation not permitted [1]
+W0827 04:31:30.514082  2528 memory_location.cpp:73] Failed to get NUMA node, addr: 0x507df700, len: 6912: Operation not permitted [1]
+W0827 04:31:30.514106  2529 memory_location.cpp:73] Failed to get NUMA node, addr: 0x508aa940, len: 6912: Operation not permitted [1]
+W0827 04:31:30.514214  2530 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50f67bc0, len: 6912: Operation not permitted [1]
+W0827 04:31:30.514272  2531 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50a2f680, len: 55296: Operation not permitted [1]
+W0827 04:31:30.514393  2532 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50afecc0, len: 55296: Operation not permitted [1]
+W0827 04:31:30.514506  2533 memory_location.cpp:73] Failed to get NUMA node, addr: 0x502be580, len: 6912: Operation not permitted [1]
+W0827 04:31:30.514607  2534 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5054d3c0, len: 13824: Operation not permitted [1]
+W0827 04:31:30.514652  2535 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50ad4580, len: 6912: Operation not permitted [1]
+W0827 04:31:30.514743  2536 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50eb9880, len: 6912: Operation not permitted [1]
+W0827 04:31:31.680995  2556 memory_location.cpp:73] Failed to get NUMA node, addr: 0x501c3300, len: 6912: Operation not permitted [1]
+W0827 04:31:31.680985  2557 memory_location.cpp:73] Failed to get NUMA node, addr: 0x504ecec0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.680985  2555 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd4bac0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681169  2558 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50706900, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681272  2559 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5006d700, len: 55296: Operation not permitted [1]
+W0827 04:31:31.681360  2560 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50041cc0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.681468  2561 memory_location.cpp:73] Failed to get NUMA node, addr: 0x506f5a40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681562  2562 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fdfdf40, len: 13824: Operation not permitted [1]
+W0827 04:31:31.681607  2563 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd16080, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681685  2564 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fcfa300, len: 6912: Operation not permitted [1]
+W0827 04:31:31.733497  2583 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51554040, len: 6912: Operation not permitted [1]
+W0827 04:31:31.733552  2584 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5110a8c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.733618  2585 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51ca8e80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.733795  2586 memory_location.cpp:73] Failed to get NUMA node, addr: 0x52017680, len: 6912: Operation not permitted [1]
+W0827 04:31:31.733909  2587 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5174af80, len: 55296: Operation not permitted [1]
+W0827 04:31:31.734004  2588 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51770440, len: 55296: Operation not permitted [1]
+W0827 04:31:31.734061  2589 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51118000, len: 6912: Operation not permitted [1]
+W0827 04:31:31.734179  2590 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51564f40, len: 13824: Operation not permitted [1]
+W0827 04:31:31.734277  2591 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51f69480, len: 6912: Operation not permitted [1]
+W0827 04:31:31.734400  2592 memory_location.cpp:73] Failed to get NUMA node, addr: 0x517c2940, len: 6912: Operation not permitted [1]
+W0827 04:31:31.741586  2611 memory_location.cpp:73] Failed to get NUMA node, addr: 0x519520c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.741609  2612 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50ff8c40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.741679  2613 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51a2ad80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.741792  2614 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51b5b300, len: 6912: Operation not permitted [1]
+W0827 04:31:31.741938  2615 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50fe5b80, len: 55296: Operation not permitted [1]
+W0827 04:31:31.742043  2616 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51625140, len: 55296: Operation not permitted [1]
+W0827 04:31:31.742168  2617 memory_location.cpp:73] Failed to get NUMA node, addr: 0x516bf640, len: 6912: Operation not permitted [1]
+W0827 04:31:31.742278  2618 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51b73e80, len: 13824: Operation not permitted [1]
+W0827 04:31:31.742414  2619 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51ef60c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.742429  2620 memory_location.cpp:73] Failed to get NUMA node, addr: 0x513e95c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.900663  2641 memory_location.cpp:73] Failed to get NUMA node, addr: 0x517e8c80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.900679  2640 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5170b840, len: 6912: Operation not permitted [1]
+W0827 04:31:31.900738  2642 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5131d0c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.900669  2639 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51372400, len: 6912: Operation not permitted [1]
+W0827 04:31:31.900923  2643 memory_location.cpp:73] Failed to get NUMA node, addr: 0x512b7e00, len: 55296: Operation not permitted [1]
+W0827 04:31:31.900976  2644 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51380200, len: 55296: Operation not permitted [1]
+W0827 04:31:31.901000  2646 memory_location.cpp:73] Failed to get NUMA node, addr: 0x513151c0, len: 13824: Operation not permitted [1]
+W0827 04:31:31.901011  2645 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51194f40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.901038  2647 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51520740, len: 6912: Operation not permitted [1]
+W0827 04:31:31.901073  2648 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51724980, len: 6912: Operation not permitted [1]
+W0827 04:31:31.931138  2668 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51061000, len: 6912: Operation not permitted [1]
+W0827 04:31:31.931178  2669 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50d74c00, len: 6912: Operation not permitted [1]
+W0827 04:31:31.931098  2667 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50b2fb40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.931272  2670 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50de9000, len: 6912: Operation not permitted [1]
+W0827 04:31:31.931349  2671 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5097dc80, len: 55296: Operation not permitted [1]
+W0827 04:31:31.931519  2672 memory_location.cpp:73] Failed to get NUMA node, addr: 0x515aa000, len: 55296: Operation not permitted [1]
+W0827 04:31:31.931557  2675 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51269700, len: 6912: Operation not permitted [1]
+W0827 04:31:31.931556  2673 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5097bd00, len: 6912: Operation not permitted [1]
+W0827 04:31:31.931576  2674 memory_location.cpp:73] Failed to get NUMA node, addr: 0x514ee800, len: 13824: Operation not permitted [1]
+W0827 04:31:31.931759  2676 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50deac80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.983201  2696 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50819380, len: 6912: Operation not permitted [1]
+W0827 04:31:31.983201  2695 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50600a80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.983378  2697 memory_location.cpp:73] Failed to get NUMA node, addr: 0x506ddf40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.983438  2698 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50716800, len: 6912: Operation not permitted [1]
+W0827 04:31:31.983490  2699 memory_location.cpp:73] Failed to get NUMA node, addr: 0x501575c0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.983624  2700 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50863340, len: 55296: Operation not permitted [1]
+W0827 04:31:31.983737  2701 memory_location.cpp:73] Failed to get NUMA node, addr: 0x508d7a40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.983930  2702 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50668c40, len: 13824: Operation not permitted [1]
+W0827 04:31:31.984035  2703 memory_location.cpp:73] Failed to get NUMA node, addr: 0x506841c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.984107  2704 memory_location.cpp:73] Failed to get NUMA node, addr: 0x508f16c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.001853  2725 memory_location.cpp:73] Failed to get NUMA node, addr: 0x507c8d40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.001852  2724 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50b39240, len: 6912: Operation not permitted [1]
+W0827 04:31:32.002048  2727 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50a8ecc0, len: 55296: Operation not permitted [1]
+W0827 04:31:32.001888  2726 memory_location.cpp:73] Failed to get NUMA node, addr: 0x512a8140, len: 6912: Operation not permitted [1]
+W0827 04:31:32.002044  2728 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50a68880, len: 55296: Operation not permitted [1]
+W0827 04:31:32.002257  2729 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5087b880, len: 6912: Operation not permitted [1]
+W0827 04:31:32.002264  2731 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50fc6980, len: 6912: Operation not permitted [1]
+W0827 04:31:32.002266  2730 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51331100, len: 13824: Operation not permitted [1]
+W0827 04:31:32.002470  2732 memory_location.cpp:73] Failed to get NUMA node, addr: 0x513384c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.002780  2723 memory_location.cpp:73] Failed to get NUMA node, addr: 0x512c6940, len: 6912: Operation not permitted [1]
+[2026-08-27 04:31:32] Dummy health check server started in background thread at 0.0.0.0:30000
+[rank16]:[W827 04:33:25.972222579 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank17]:[W827 04:33:25.980800100 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank19]:[W827 04:33:25.981413562 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank18]:[W827 04:33:26.644294178 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank16]:[W827 04:34:10.467297472 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank18]:[W827 04:34:10.467837454 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank17]:[W827 04:34:10.468151853 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank19]:[W827 04:34:10.468427857 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank22]:[W827 04:34:10.495396864 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank23]:[W827 04:34:10.495456053 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank21]:[W827 04:34:10.495795969 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank20]:[W827 04:34:10.495814123 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank22]:[W827 04:34:16.703081720 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank20]:[W827 04:34:16.703915664 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank21]:[W827 04:34:16.704033167 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank23]:[W827 04:34:16.704314444 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[2026-08-27 04:35:10 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.29
+[2026-08-27 04:35:10 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.29
+[2026-08-27 04:35:29 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.39
+[2026-08-27 04:35:29 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.39
+[2026-08-27 04:35:31 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25.89
+[2026-08-27 04:35:31 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25.91
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/d_starting_p_node3.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/d_starting_p_node3.log
new file mode 100644
index 0000000..ee3007d
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/d_starting_p_node3.log
@@ -0,0 +1,757 @@
+/usr/local/lib/python3.12/dist-packages/torchao/quantization/quant_api.py:1731: SyntaxWarning: invalid escape sequence '\.'
+  """Configuration class for applying different quantization configs to modules or parameters based on their fully qualified names (FQNs).
+/sgl-workspace/sglang/python/sglang/launch_server.py:56: UserWarning: 'python -m sglang.launch_server' is still supported, but 'sglang serve' is the recommended entrypoint.
+  Example: sglang serve --model-path <model> [options]
+  warnings.warn(
+[2026-08-27 04:28:46] Breakable CUDA graph is incompatible with MLA attention; disabling prefill CUDA graph.
+[2026-08-27 04:28:47] Attention backend not specified. Use triton backend by default.
+[2026-08-27 04:28:47] Pipeline parallelism is incompatible with overlap schedule.
+[2026-08-27 04:28:49] server_args=ServerArgs(model_path='/data/hf_models/Kimi-K3', tokenizer_path='/data/hf_models/Kimi-K3', tokenizer_mode='auto', tokenizer_backend='huggingface', tokenizer_worker_num=1, detokenizer_worker_num=1, skip_tokenizer_init=False, load_format='auto', model_loader_extra_config='{}', trust_remote_code=True, context_length=None, is_embedding=False, enable_multimodal=None, revision=None, model_impl='auto', model_config_parser='auto', json_model_override_args='{}', dtype='auto', quantization=None, quantization_param_path=None, kv_cache_dtype='auto', enable_fp32_lm_head=False, modelopt_quant=None, modelopt_checkpoint_restore_path=None, modelopt_checkpoint_save_path=None, modelopt_export_path=None, quantize_and_serve=False, rl_quant_profile=None, enable_tf32_matmul=False, mem_fraction_static=0.88, max_running_requests=None, max_queued_requests=None, max_total_tokens=None, chunked_prefill_size=8192, enable_dynamic_chunking=False, max_prefill_tokens=16384, prefill_max_requests=None, schedule_policy='fcfs', enable_priority_scheduling=False, disable_priority_preemption=False, default_priority_value=None, abort_on_priority_when_disabled=False, schedule_low_priority_values_first=False, priority_scheduling_preemption_threshold=10, retraction_policy='length', schedule_conservativeness=1.0, page_size=64, swa_full_tokens_ratio=0.8, disable_hybrid_swa_memory=False, radix_eviction_policy='lru', prefill_only_disable_kv_cache=False, disable_radix_cache=True, enable_page_major_kv_layout=False, enable_unified_memory=False, disable_chunked_prefix_cache=False, disable_overlap_schedule=True, num_continuous_decode_steps=1, scheduler_recv_interval=1, enable_mixed_chunk=False, nccl_port=None, dist_timeout=3600, dist_init_addr='174.1.60.1:20000', nnodes=4, node_rank=3, tp_size=4, dcp_size=1, pp_size=8, pp_max_micro_batch_size=None, pp_async_batch_depth=0, dp_size=1, load_balance_method='follow_bootstrap_room', attn_cp_size=1, moe_dp_size=1, dcp_comm_backend='ag_rs', dcp_replicate_q_proj=None, dwdp_size=1, enable_prefill_cp=False, cp_strategy=None, enable_dsa_cache_layer_split=False, enable_dsa_prefill_context_parallel=False, dsa_prefill_cp_mode='round-robin-split', enable_prefill_context_parallel=False, prefill_cp_mode='in-seq-split', enable_cp_decode_attn_tp=False, enable_dp_attention=False, enable_dp_attention_local_control_broadcast=False, enable_dp_lm_head=False, enable_attn_tp_input_scattered=False, disable_attn_tp_gather=False, enable_p2p_check=False, device='cuda', base_gpu_id=0, gpu_id_step=1, random_seed=927276655, watchdog_timeout=300, soft_watchdog_timeout=None, sleep_on_idle=False, use_ray=False, custom_sigquit_handler=None, numa_node=None, gc_threshold=None, host='0.0.0.0', port=30000, fastapi_root_path='', smg_grpc_mode=False, grpc_mode=False, grpc_port=None, sidecar=None, sidecar_args=None, skip_server_warmup=False, warmups=None, enable_http2=False, ssl_keyfile=None, ssl_certfile=None, ssl_ca_certs=None, ssl_keyfile_password=None, enable_ssl_refresh=False, api_key=None, admin_api_key=None, served_model_name='kimi-k3', weight_version='default', chat_template=None, hf_chat_template_name=None, completion_template=None, file_storage_path='sglang_storage', enable_cache_report=False, reasoning_parser=None, default_chat_template_kwargs=None, strip_thinking_cache=False, enable_strict_thinking=False, tool_call_parser=None, tool_server=None, sampling_defaults='model', asr_max_buffer_seconds=60, asr_max_concurrent_sessions=32, preferred_sampling_params=None, allow_auto_truncate=False, stream_interval=1, batch_notify_size=16, stream_response_default_include_usage=False, incremental_streaming_output=False, enable_streaming_session=False, enable_session_radix_cache=False, log_level='info', log_level_http=None, log_requests=False, log_requests_level=2, log_requests_format='text', log_requests_target=None, uvicorn_access_log_exclude_prefixes=[], crash_dump_folder=None, show_time_cost=False, enable_metrics=False, smg_http_sidecar_port=None, enable_mfu_metrics=False, enable_metrics_for_all_schedulers=False, load_snapshot_publish_interval=15, tokenizer_metrics_custom_labels_header='x-custom-labels', tokenizer_metrics_allowed_custom_labels=None, extra_metric_labels=None, bucket_time_to_first_token=None, bucket_inter_token_latency=None, bucket_e2e_request_latency=None, prompt_tokens_buckets=None, generation_tokens_buckets=None, gc_warning_threshold_secs=0.0, decode_log_interval=40, enable_request_time_stats_logging=False, kv_events_config=None, enable_forward_pass_metrics=False, forward_pass_metrics_worker_id='', forward_pass_metrics_ipc_name=None, enable_trace=False, trace_modules='request', otlp_traces_endpoint='localhost:4317', export_metrics_to_file=False, export_metrics_to_file_dir=None, stat_loggers=None, constrained_json_whitespace_pattern=None, constrained_json_disable_any_whitespace=False, attention_backend='triton', decode_attention_backend=None, prefill_attention_backend=None, sampling_backend='flashinfer', grammar_backend='xgrammar', radix_cache_backend=None, mm_attention_backend=None, fp8_gemm_runner_backend='auto', fp4_gemm_runner_backend='auto', bf16_gemm_backend='auto', dsa_prefill_backend=None, dsa_decode_backend=None, dsa_paged_mqa_logits_backend='auto', dsa_topk_backend='sgl-kernel', disable_flashinfer_autotune=False, flashinfer_autotune_skip_ops=None, mamba_backend='triton', cuda_graph_config=CudaGraphConfig(decode=PhaseConfig(backend='disabled', max_bs=16, bs=[1, 2, 4, 8, 12, 16], tc_compiler='eager', full_prefill_max_req=None), prefill=PhaseConfig(backend='disabled', max_bs=2048, bs=[4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], tc_compiler='eager', full_prefill_max_req=None)), cuda_graph_backend_decode=None, cuda_graph_backend_prefill=None, cuda_graph_max_bs_decode=16, cuda_graph_max_bs_prefill=None, cuda_graph_bs_decode=None, cuda_graph_bs_prefill=None, cuda_graph_tc_compiler=None, disable_prefill_cuda_graph=False, disable_decode_cuda_graph=False, disable_cuda_graph=False, disable_cuda_graph_padding=False, enable_profile_cuda_graph=False, enable_cudagraph_gc=False, debug_cuda_graph=False, enable_layerwise_nvtx_marker=False, enable_nccl_nvls=False, enable_symm_mem=False, triton_attention_reduce_in_fp32=False, triton_attention_num_kv_splits=8, triton_attention_split_tile_size=None, flashinfer_mla_disable_ragged=False, enable_fused_qk_norm_rope=False, enable_precise_embedding_interpolation=False, enable_fused_moe_sum_all_reduce=False, enable_deepseek_v4_fp4_indexer=False, disable_custom_all_reduce=False, enable_mscclpp=False, enable_torch_symm_mem=False, enable_scattered_sconv=False, pre_warm_nccl=False, enable_quant_communications=False, enable_flashinfer_allreduce_fusion=False, enforce_disable_flashinfer_allreduce_fusion=False, flashinfer_allreduce_fusion_backend=None, enable_aiter_allreduce_fusion=False, enable_torch_compile=False, enable_torch_compile_debug_mode=False, torch_compile_max_bs=32, torchao_config='', speculative_algorithm=None, speculative_draft_model_path=None, speculative_draft_model_revision=None, speculative_draft_load_format=None, speculative_num_steps=None, speculative_eagle_topk=None, speculative_num_draft_tokens=None, speculative_dflash_block_size=None, speculative_dspark_block_size=None, speculative_dspark_sps_table_path=None, speculative_dspark_confidence_sts_path=None, speculative_dspark_align_verify_tokens_to_graph_tier=False, speculative_accept_threshold_single=1.0, speculative_accept_threshold_acc=1.0, speculative_use_rejection_sampling=False, speculative_token_map=None, speculative_attention_mode='prefill', speculative_draft_attention_backend=None, speculative_draft_window_size=None, speculative_moe_runner_backend='flashinfer_mxfp4', speculative_moe_a2a_backend=None, speculative_draft_model_quantization=None, speculative_skip_dp_mlp_sync=False, enable_multi_layer_eagle=False, speculative_adaptive=False, speculative_adaptive_config=None, decoupled_spec_bind_endpoint=None, decoupled_spec_connect_endpoints=None, decoupled_spec_rank=None, decoupled_spec_role='null', spec_trace_dir=None, speculative_ngram_min_bfs_breadth=1, speculative_ngram_max_bfs_breadth=10, speculative_ngram_match_type='BFS', speculative_ngram_max_trie_depth=18, speculative_ngram_capacity=10000000, speculative_ngram_external_corpus_path=None, speculative_ngram_external_sam_budget=0, speculative_ngram_external_corpus_max_tokens=10000000, ep_size=4, moe_a2a_backend='none', moe_runner_backend='flashinfer_mxfp4', flashinfer_mxfp4_moe_precision='default', deepep_mode='auto', fuseep_mode=2, deepep_dispatcher_output_dtype='auto', ep_num_redundant_experts=0, ep_dispatch_algorithm=None, init_expert_location='trivial', enable_eplb=False, eplb_algorithm='auto', eplb_rebalance_num_iterations=1000, eplb_rebalance_layers_per_chunk=None, eplb_min_rebalancing_utilization_threshold=1.0, expert_distribution_recorder_mode=None, expert_distribution_recorder_buffer_size=1000, enable_expert_distribution_metrics=False, deepep_config=None, moe_dense_tp_size=None, elastic_ep_backend=None, enable_elastic_expert_backup=False, mooncake_ib_device=None, enable_waterfill=False, ep_join_mode=None, ep_join_rank_offset=0, elastic_ep_initial_size=None, max_ep_size=None, elastic_ep_scale_timeout=600, elastic_ep_rejoin=False, disable_flashinfer_cutlass_moe_fp4_allgather=False, disable_shared_experts_fusion=False, enforce_shared_experts_fusion=False, max_mamba_cache_size=None, mamba_ssm_dtype=None, mamba_max_states_per_path=-1, enable_mamba_cache_stochastic_rounding=False, mamba_cache_philox_rounds=0, mamba_full_memory_ratio=0.36, mamba_radix_cache_strategy='extra_buffer_lazy', uses_mamba_radix_cache=False, mamba_track_interval=256, enable_int8_mamba_checkpoint=False, int8_mamba_ckpt_size=None, linear_attn_backend='triton', linear_attn_decode_backend=None, linear_attn_prefill_backend=None, linear_attn_verify_backend=None, enable_linear_replayssm=False, linear_replayssm_cache_len=16, enable_linear_replayssm_spec=False, enable_hierarchical_cache=False, hicache_ratio=2.0, hicache_size=0, hicache_write_policy='write_through', hicache_io_backend='kernel', hicache_mem_layout='page_first', hicache_storage_backend=None, hicache_storage_prefetch_policy='timeout', hicache_storage_backend_extra_config=None, enable_hisparse=False, hisparse_config=None, enable_broadcast_mm_inputs_process=False, enable_prefix_mm_cache=False, mm_enable_dp_encoder=False, mm_process_config={}, mm_processor_worker_num=0, mm_io_worker_num=0, limit_mm_data_per_request=None, enable_mm_global_cache=False, disable_fast_image_processor=False, mm_feature_transport='cpu', keep_mm_feature_on_device=False, enable_lora=None, enable_lora_overlap_loading=None, max_lora_rank=None, lora_target_modules=None, lora_paths=None, max_loaded_loras=None, max_loras_per_batch=8, lora_eviction_policy='lru', lora_backend='csgmv', max_lora_chunk_size=16, experts_shared_outer_loras=None, lora_use_virtual_experts=False, lora_strict_loading=False, lora_drain_wait_threshold=0.0, enable_two_batch_overlap=False, enable_single_batch_overlap=False, tbo_token_distribution_threshold=0.48, cpu_offload_gb=0, offload_group_size=-1, offload_num_in_group=1, offload_prefetch_step=1, offload_mode='cpu', enable_lmcache=False, lmcache_config_file=None, enable_flexkv=False, flexkv_config_file=None, kt_weight_path=None, kt_method='AMXINT4', kt_cpuinfer=None, kt_threadpool_count=2, kt_num_gpu_experts=None, kt_max_deferred_experts_per_token=None, dllm_algorithm=None, dllm_algorithm_config=None, dllm_fdfo=True, disaggregation_mode='prefill', disaggregation_transfer_backend='mooncake', disaggregation_bootstrap_port=28800, disaggregation_ib_device='mlx5_0,mlx5_1,mlx5_2,mlx5_3', disaggregation_decode_enable_radix_cache=False, disaggregation_decode_enable_offload_kvcache=False, num_reserved_decode_tokens=512, disaggregation_decode_extra_slots=None, disaggregation_decode_polling_interval=1, optimistic_prefill_attempts=0, encoder_only=False, language_only=False, encoder_transfer_backend='zmq_to_tokenizer', encoder_urls=[], encoder_bootstrap_port=8997, encoder_register_urls=[], enable_adaptive_dispatch_to_encoder=False, enable_pdmux=False, pdmux_config_path=None, sm_group_num=8, custom_weight_loader=[], weight_loader_disable_mmap=False, weight_loader_prefetch_checkpoints=False, weight_loader_prefetch_num_threads=4, weight_loader_drop_cache_after_load=False, remote_instance_weight_loader_seed_instance_ip=None, remote_instance_weight_loader_seed_instance_service_port=None, remote_instance_weight_loader_send_weights_group_ports=None, remote_instance_weight_loader_backend='nccl', remote_instance_weight_loader_start_seed_via_transfer_engine=False, engine_info_bootstrap_port=6789, modelexpress_config=None, download_dir=None, model_checksum=None, delete_ckpt_after_loading=False, decrypted_config_file=None, decrypted_draft_config_file=None, checkpoint_engine_wait_weights_before_ready=False, enable_prefill_delayer=False, prefill_delayer_max_delay_passes=30, prefill_delayer_token_usage_low_watermark=None, prefill_delayer_forward_passes_buckets=None, prefill_delayer_wait_seconds_buckets=None, prefill_delayer_queue_min_ratio=None, prefill_delayer_max_delay_ms=None, min_free_slots_delay=None, enable_deterministic_inference=False, rl_on_policy_target=None, kv_canary='none', kv_canary_real_data='none', kv_canary_sweep_interval=0, enable_dynamic_batch_tokenizer=False, dynamic_batch_tokenizer_batch_size=32, dynamic_batch_tokenizer_batch_timeout=0.002, enable_tokenizer_batch_encode=False, disable_tokenizer_batch_decode=False, debug_tensor_dump_output_folder=None, debug_tensor_dump_layers=None, debug_tensor_dump_input_file=None, enable_memory_saver=False, enable_weights_cpu_backup=False, enable_draft_weights_cpu_backup=False, enable_custom_logit_processor=False, enable_return_hidden_states=False, enable_return_routed_experts=False, enable_return_indexer_topk=False, disable_outlines_disk_cache=False, enable_mis=False, weight_cache_mode='off', weight_cache_socket=None, weight_cache_timeout=1800, forward_hooks=None, msprobe_dump_config=None)
+[2026-08-27 04:28:49] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:49] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:49] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:49] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:49] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:49] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:49] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:49] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:05 PP6 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:05 PP6 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:07 PP6 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:07 PP6 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:07 PP6 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:07 PP7 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:07 PP7 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:07 PP7 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:07 PP7 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:07 PP6 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:07 PP6 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:07.662256   238 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:07.662379   238 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:07.662392   238 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.4 port: 12001
+I0827 04:29:07.662429   238 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.4:15914
+I0827 04:29:07.662514   238 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:07.685189   238 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:07.689950   238 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:07.695539   238 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:07.701294   238 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:07.703691   238 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:07.703715   238 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:07.703760   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:07.710492   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:07.711953   238 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:04
+I0827 04:29:07.711989   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:07.735716   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:07.737195   238 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:04
+I0827 04:29:07.737223   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:07.763653   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:07.765146   238 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:04
+I0827 04:29:07.765172   238 rdma_context.cpp:205] Using SIEVE endpoint store
+[2026-08-27 04:29:07 PP7 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:07 PP7 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:07 PP7 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:07 PP7 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+I0827 04:29:07.791589   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:07.793025   238 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:04
+I0827 04:29:07.793139   238 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:07 PP6 TP3 EP3] Init torch distributed begin.
+[2026-08-27 04:29:07 PP6 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:07 PP6 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:08 PP6 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:08.916008   235 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:08.916072   235 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:08.916082   235 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.4 port: 12001
+I0827 04:29:08.916121   235 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.4:15737
+I0827 04:29:08.916215   235 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:08.941840   235 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:08.950320   235 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:08.979278   235 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:08.988204   235 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:08.991855   235 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:08.991879   235 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:08.991928   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:08.999985   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:09.001416   235 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:04
+I0827 04:29:09.001446   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:09.027477   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:09.028954   235 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:04
+I0827 04:29:09.028980   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:09.055524   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:09.056981   235 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:04
+I0827 04:29:09.057020   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:09.083544   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:09.084911   235 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:04
+I0827 04:29:09.085031   235 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:09 PP6 TP0 EP0] Init torch distributed begin.
+[2026-08-27 04:29:09 PP7 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:09 PP6 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:09 PP7 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:09 PP6 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:09 PP7 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:09 PP7 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:10.000486   242 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:10.000609   242 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:10.000628   242 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.4 port: 12001
+I0827 04:29:10.000689   242 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.4:16890
+I0827 04:29:10.000851   242 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:10.006592   236 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:10.006711   236 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:10.006728   236 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.4 port: 12001
+I0827 04:29:10.006786   236 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.4:15893
+I0827 04:29:10.006963   236 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:10.022238   241 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:10.022354   241 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:10.022370   241 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.4 port: 12001
+I0827 04:29:10.022423   241 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.4:16277
+I0827 04:29:10.022593   241 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:10.032052   242 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:10.032101   236 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:10.039772   236 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:10.039858   242 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:10.047292   236 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:10.047341   242 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:10.047785   241 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:10.054806   242 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:10.054838   236 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:10.055280   241 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:10.059140   236 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:10.059171   236 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:10.059175   242 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:10.059201   242 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:10.059227   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.059260   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.062610   241 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:10.066787   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:10.066826   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:10.068500   242 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:04
+I0827 04:29:10.068532   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.068558   236 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:04
+I0827 04:29:10.068590   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.069866   241 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:10.072566   241 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:10.072592   241 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:10.072649   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.080149   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:10.081809   241 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:04
+I0827 04:29:10.081843   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.095701   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:10.097368   236 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:04
+I0827 04:29:10.097400   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.099974   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:10.101672   242 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:04
+I0827 04:29:10.101717   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.107705   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:10.109316   241 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:04
+I0827 04:29:10.109347   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.123868   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:10.125504   236 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:04
+I0827 04:29:10.125536   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.127957   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:10.129673   242 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:04
+I0827 04:29:10.129703   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.139813   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:10.141404   241 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:04
+I0827 04:29:10.141431   241 rdma_context.cpp:205] Using SIEVE endpoint store
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:10.150050   237 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:10.150159   237 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:10.150173   237 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.4 port: 12001
+I0827 04:29:10.150219   237 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.4:15413
+I0827 04:29:10.150358   237 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:10.151880   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:10.153475   236 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:04
+I0827 04:29:10.153601   236 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:10 PP6 TP1 EP1] Init torch distributed begin.
+I0827 04:29:10.155849   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:10.155864   237 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:10.157462   242 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:04
+I0827 04:29:10.157585   242 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:10 PP7 TP3 EP3] Init torch distributed begin.
+I0827 04:29:10.171923   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:10.172516   240 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:10.172667   240 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:10.172688   240 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.4 port: 12001
+I0827 04:29:10.172762   240 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.4:15705
+I0827 04:29:10.172977   240 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:10.173573   241 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:04
+I0827 04:29:10.173692   241 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:10 PP7 TP2 EP2] Init torch distributed begin.
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:10.177717   239 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:10.177865   239 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:10.177886   239 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.4 port: 12001
+I0827 04:29:10.177956   239 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.4:16192
+I0827 04:29:10.178146   239 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:10.183638   237 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:10.187090   239 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:10.191001   237 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:10.195258   239 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:10.198271   237 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:10.200984   237 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:10.201025   237 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:10.201079   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.203014   239 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:10.203156   240 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:10.209589   240 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:10.210305   239 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:10.212793   239 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:10.212821   239 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:10.212878   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.216092   240 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:10.219946   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:10.221297   239 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:04
+I0827 04:29:10.221323   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.222187   240 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:10.224284   240 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:10.224304   240 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:10.224344   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.230640   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:10.231778   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:10.231937   240 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:04
+I0827 04:29:10.231962   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.233362   237 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:04
+I0827 04:29:10.233397   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.236478   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:10.237651   240 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:04
+I0827 04:29:10.237670   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.250197   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:10.251711   239 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:04
+I0827 04:29:10.251737   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.263743   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:10.265311   237 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:04
+I0827 04:29:10.265339   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.267139   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:10.268607   240 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:04
+I0827 04:29:10.268632   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.273075   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:10.274682   237 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:04
+I0827 04:29:10.274710   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.276602   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:10.278275   239 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:04
+I0827 04:29:10.278314   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.295709   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:10.297385   240 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:04
+I0827 04:29:10.297510   240 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:10 PP7 TP1 EP1] Init torch distributed begin.
+I0827 04:29:10.299893   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:10.301563   237 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:04
+I0827 04:29:10.301689   237 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:10 PP6 TP2 EP2] Init torch distributed begin.
+I0827 04:29:10.303737   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:10.305307   239 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:04
+I0827 04:29:10.305433   239 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:10 PP7 TP0 EP0] Init torch distributed begin.
+[2026-08-27 04:29:26 PP7 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:26 PP6 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:27 PP6 TP0 EP0] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP6 TP3 EP3] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP6 TP2 EP2] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP6 TP1 EP1] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP7 TP1 EP1] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP7 TP0 EP0] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP7 TP3 EP3] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP7 TP2 EP2] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP7 TP1 EP1] Init torch distributed ends. elapsed=17.45 s, mem usage=0.46 GB
+[2026-08-27 04:29:27 PP7 TP3 EP3] Init torch distributed ends. elapsed=17.59 s, mem usage=0.46 GB
+[2026-08-27 04:29:27 PP7 TP2 EP2] Init torch distributed ends. elapsed=17.57 s, mem usage=0.46 GB
+[2026-08-27 04:29:27 PP7 TP0 EP0] Init torch distributed ends. elapsed=17.44 s, mem usage=0.46 GB
+[2026-08-27 04:29:27 PP6 TP1 EP1] Init torch distributed ends. elapsed=17.59 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP6 TP3 EP3] Init torch distributed ends. elapsed=19.95 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP6 TP0 EP0] Init torch distributed ends. elapsed=18.66 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP6 TP2 EP2] Init torch distributed ends. elapsed=17.44 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP6 TP1 EP1] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP7 TP1 EP1] Load weight begin. avail mem=82.12 GB
+[2026-08-27 04:29:27 PP6 TP0 EP0] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP6 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP7 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP7 TP3 EP3] Load weight begin. avail mem=82.12 GB
+[2026-08-27 04:29:27 PP6 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP7 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP6 TP2 EP2] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP6 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP7 TP2 EP2] Load weight begin. avail mem=82.12 GB
+[2026-08-27 04:29:28 PP7 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP6 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP7 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP6 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP7 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP7 TP0 EP0] Load weight begin. avail mem=82.12 GB
+[2026-08-27 04:29:28 PP6 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP7 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP7 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP6 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP7 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP6 TP3 EP3] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP6 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP6 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP6 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP7 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP7 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP7 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP6 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP7 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP6 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:30:40 PP7 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:40 PP7 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:40 PP7 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:40 PP7 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:40 PP7 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:40 PP7 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:40 PP7 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:40 PP7 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:41 PP7 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:41 PP7 TP1 EP1] Load weight end. elapsed=73.17 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=31.91 GB, mem usage=50.21 GB.
+[2026-08-27 04:30:41 PP7 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:41 PP7 TP0 EP0] Load weight end. elapsed=73.80 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=31.91 GB, mem usage=50.21 GB.
+[2026-08-27 04:30:42 PP6 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:42 PP6 TP1 EP1] Load weight end. elapsed=74.47 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:45 PP6 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:45 PP6 TP2 EP2] Load weight end. elapsed=77.20 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:50 PP7 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:50 PP7 TP3 EP3] Load weight end. elapsed=82.79 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=31.91 GB, mem usage=50.21 GB.
+[2026-08-27 04:30:51 PP6 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:51 PP6 TP3 EP3] Load weight end. elapsed=83.19 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:51 PP7 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:51 PP7 TP2 EP2] Load weight end. elapsed=83.68 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=31.91 GB, mem usage=50.21 GB.
+[2026-08-27 04:30:51 PP7 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:51 PP7 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:52 PP7 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:52 PP7 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:52 PP7 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:52 PP7 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:52 PP7 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:52 PP7 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:53 PP6 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:53 PP6 TP0 EP0] Load weight end. elapsed=85.47 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:53 PP6 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:53 PP6 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:53 PP6 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:53 PP6 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:53 PP6 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:53 PP6 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:53 PP6 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:53 PP6 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:54 PP7 TP3 EP3] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP7 TP2 EP2] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP7 TP1 EP1] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP7 TP0 EP0] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP6 TP3 EP3] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP6 TP2 EP2] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP6 TP1 EP1] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP6 TP0 EP0] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP6 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP6 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP7 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:54 PP6 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP7 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:54 PP7 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:54 PP7 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:54 PP6 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP6 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP6 TP3 EP3] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP6 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP6 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP6 TP2 EP2] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP6 TP1 EP1] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP7 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 16.31 GB
+[2026-08-27 04:30:54 PP7 TP2 EP2] Memory pool end. avail mem=14.95 GB
+[2026-08-27 04:30:54 PP6 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP6 TP0 EP0] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP7 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 16.31 GB
+[2026-08-27 04:30:54 PP7 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 16.31 GB
+[2026-08-27 04:30:54 PP7 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 16.31 GB
+[2026-08-27 04:30:54 PP7 TP3 EP3] Memory pool end. avail mem=14.95 GB
+[2026-08-27 04:30:54 PP7 TP0 EP0] Memory pool end. avail mem=14.95 GB
+[2026-08-27 04:30:54 PP7 TP1 EP1] Memory pool end. avail mem=14.95 GB
+[2026-08-27 04:30:54 PP7 TP2 EP2] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp2_pp7_dp0.json
+[2026-08-27 04:30:54 PP6 TP3 EP3] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp3_pp6_dp0.json
+[2026-08-27 04:30:54 PP6 TP1 EP1] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp1_pp6_dp0.json
+[2026-08-27 04:30:54 PP7 TP3 EP3] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp3_pp7_dp0.json
+[2026-08-27 04:30:54 PP6 TP2 EP2] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp2_pp6_dp0.json
+[2026-08-27 04:30:54 PP6 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:30:54 PP6 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:30:54 PP6 TP0 EP0] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp0_pp6_dp0.json
+[2026-08-27 04:30:54 PP7 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:30:54 PP7 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:30:54 PP7 TP0 EP0] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp0_pp7_dp0.json
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[2026-08-27 04:30:54 PP7 TP1 EP1] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp1_pp7_dp0.json
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 11.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 13.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.76profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.62profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.66profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.72profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.68profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.55profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.12profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  6.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  7.16profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.92profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.15profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.59profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.56profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 19.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  8.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.67profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.40profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.88profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  8.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  6.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.83profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.50profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.32profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.92profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.77profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.94profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.57profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 12.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 12.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 13.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.36profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.66profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  6.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.72profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  6.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.22profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.65profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.13profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.03profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.46profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.59profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.16profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  6.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.96profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.76profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.67profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  8.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.27profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.88profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.01profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.57profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.12profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 11.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  2.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.68profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.90profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.43profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.92profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  8.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.66profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  6.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  7.13profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.32profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.71profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:05,  1.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.16profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.31profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.77profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.28profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.48profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.87profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.40profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 13.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.35profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.98profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.88profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.73profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  3.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.88profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.09profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.20profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.97profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.83profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.47profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.40profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.60profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  8.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  7.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.97profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.05profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.05profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  6.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.49profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.72profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.19profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 10.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.91profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.64profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.58profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  3.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.35profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  9.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 12.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.09profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:02,  2.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 12.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.15profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.76profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.49profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.03profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.79profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.96profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.47profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.65profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.55profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.03profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  9.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.63profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.98profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:05,  1.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:01,  3.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.50profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  6.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.94profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.13profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.78profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.53profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.35profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.56profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.36profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:01<00:06,  1.06s/profile]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  8.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.91profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.66profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  8.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  7.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:02,  2.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.83profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.01profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.66profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.13profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.59profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  8.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  8.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.71profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.29profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.97profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.88profile/s]
+[2026-08-27 04:31:27 PP6 TP3 EP3] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.86profile/s]
+[2026-08-27 04:31:27 PP6 TP1 EP1] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.24profile/s]
+[2026-08-27 04:31:27 PP6 TP2 EP2] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.33profile/s]
+[2026-08-27 04:31:27 PP6 TP0 EP0] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.56profile/s][2026-08-27 04:31:27 PP6 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:27 PP6 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:27 PP6 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:27 PP6 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:27 PP6 TP0 EP0] max_total_num_tokens=3801408, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=54, context_len=1048576, available_gpu_mem=16.52 GB
+[2026-08-27 04:31:27 PP6 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:27 PP6 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:27 PP6 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:27 PP6 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
+[2026-08-27 04:31:28 PP7 TP2 EP2] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.32profile/s]
+[2026-08-27 04:31:28 PP7 TP3 EP3] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.60profile/s]
+[2026-08-27 04:31:28 PP7 TP0 EP0] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.60profile/s]
+[2026-08-27 04:31:28 PP7 TP1 EP1] FlashInfer autotune completed.
+[2026-08-27 04:31:28 PP7 TP2 EP2] multimem all-gather disabled (no multicast for world_size=4)
+[2026-08-27 04:31:28 PP7 TP3 EP3] multimem all-gather disabled (no multicast for world_size=4)
+[2026-08-27 04:31:28 PP7 TP1 EP1] multimem all-gather disabled (no multicast for world_size=4)
+[2026-08-27 04:31:28 PP7 TP0 EP0] multimem all-gather disabled (no multicast for world_size=4)
+[2026-08-27 04:31:28 PP7 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:28 PP7 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:28 PP7 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:28 PP7 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:28 PP7 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:28 PP7 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:28 PP7 TP0 EP0] max_total_num_tokens=3801408, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=54, context_len=1048576, available_gpu_mem=12.14 GB
+[2026-08-27 04:31:28 PP7 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:28 PP7 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+W0827 04:31:31.959518  2596 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5efc8ec0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.959637  2598 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ee654c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.959640  2597 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5eda3c80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.959542  2595 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5efc6f80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.959741  2599 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5eeaf500, len: 55296: Operation not permitted [1]
+W0827 04:31:31.959794  2600 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ef4e380, len: 55296: Operation not permitted [1]
+W0827 04:31:31.959827  2601 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5f548500, len: 6912: Operation not permitted [1]
+W0827 04:31:31.959918  2602 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5eec3bc0, len: 13824: Operation not permitted [1]
+W0827 04:31:31.960085  2603 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5f01bcc0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.960222  2604 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5f4c8ac0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.980162  2623 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60ba35c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.980152  2624 memory_location.cpp:73] Failed to get NUMA node, addr: 0x609ccf40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.980242  2625 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60d38600, len: 6912: Operation not permitted [1]
+W0827 04:31:31.980361  2626 memory_location.cpp:73] Failed to get NUMA node, addr: 0x609546c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.980516  2628 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60c49a00, len: 55296: Operation not permitted [1]
+W0827 04:31:31.980537  2627 memory_location.cpp:73] Failed to get NUMA node, addr: 0x607c6800, len: 55296: Operation not permitted [1]
+W0827 04:31:31.980547  2629 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60d11800, len: 6912: Operation not permitted [1]
+W0827 04:31:31.980834  2630 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60a1b740, len: 13824: Operation not permitted [1]
+W0827 04:31:31.981053  2632 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60ce5000, len: 6912: Operation not permitted [1]
+W0827 04:31:31.981062  2631 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60aef300, len: 6912: Operation not permitted [1]
+W0827 04:31:32.117022  2651 memory_location.cpp:73] Failed to get NUMA node, addr: 0x74c79d40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.117132  2652 memory_location.cpp:73] Failed to get NUMA node, addr: 0x74f99700, len: 6912: Operation not permitted [1]
+W0827 04:31:32.117317  2653 memory_location.cpp:73] Failed to get NUMA node, addr: 0x7546c740, len: 6912: Operation not permitted [1]
+W0827 04:31:32.117331  2654 memory_location.cpp:73] Failed to get NUMA node, addr: 0x74f4f140, len: 6912: Operation not permitted [1]
+W0827 04:31:32.117414  2656 memory_location.cpp:73] Failed to get NUMA node, addr: 0x750037c0, len: 55296: Operation not permitted [1]
+W0827 04:31:32.117445  2655 memory_location.cpp:73] Failed to get NUMA node, addr: 0x74bf1200, len: 55296: Operation not permitted [1]
+W0827 04:31:32.117705  2657 memory_location.cpp:73] Failed to get NUMA node, addr: 0x74f80080, len: 6912: Operation not permitted [1]
+W0827 04:31:32.117738  2658 memory_location.cpp:73] Failed to get NUMA node, addr: 0x74fbd2c0, len: 13824: Operation not permitted [1]
+W0827 04:31:32.117790  2659 memory_location.cpp:73] Failed to get NUMA node, addr: 0x74fa5e40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.118012  2660 memory_location.cpp:73] Failed to get NUMA node, addr: 0x74c24880, len: 6912: Operation not permitted [1]
+W0827 04:31:32.137427  2677 memory_location.cpp:73] Failed to get NUMA node, addr: 0x685bea40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.137485  2678 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6865c240, len: 6912: Operation not permitted [1]
+W0827 04:31:32.137612  2680 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6832f280, len: 6912: Operation not permitted [1]
+W0827 04:31:32.137683  2679 memory_location.cpp:73] Failed to get NUMA node, addr: 0x68511ec0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.137791  2681 memory_location.cpp:73] Failed to get NUMA node, addr: 0x687ae280, len: 55296: Operation not permitted [1]
+W0827 04:31:32.137800  2682 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6863aec0, len: 55296: Operation not permitted [1]
+W0827 04:31:32.137928  2683 memory_location.cpp:73] Failed to get NUMA node, addr: 0x685f4540, len: 6912: Operation not permitted [1]
+W0827 04:31:32.138145  2685 memory_location.cpp:73] Failed to get NUMA node, addr: 0x68b70f80, len: 6912: Operation not permitted [1]
+W0827 04:31:32.138144  2684 memory_location.cpp:73] Failed to get NUMA node, addr: 0x686e87c0, len: 13824: Operation not permitted [1]
+W0827 04:31:32.138219  2686 memory_location.cpp:73] Failed to get NUMA node, addr: 0x685491c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.187127  2706 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8eaeff40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.187187  2707 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8eaa1740, len: 6912: Operation not permitted [1]
+W0827 04:31:32.187136  2705 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8e5f2600, len: 6912: Operation not permitted [1]
+W0827 04:31:32.187413  2709 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ec4eb00, len: 55296: Operation not permitted [1]
+W0827 04:31:32.187405  2708 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ea3b800, len: 6912: Operation not permitted [1]
+W0827 04:31:32.187651  2710 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ed92ec0, len: 55296: Operation not permitted [1]
+W0827 04:31:32.187799  2711 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8eb3fbc0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.188133  2712 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ed0e1c0, len: 13824: Operation not permitted [1]
+W0827 04:31:32.188231  2713 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8e9d20c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.188272  2714 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8f2c7a80, len: 6912: Operation not permitted [1]
+W0827 04:31:32.252096  2732 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72b6b180, len: 6912: Operation not permitted [1]
+W0827 04:31:32.252050  2731 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72be0600, len: 6912: Operation not permitted [1]
+W0827 04:31:32.252147  2733 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72716c00, len: 6912: Operation not permitted [1]
+W0827 04:31:32.252276  2734 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72739940, len: 6912: Operation not permitted [1]
+W0827 04:31:32.252379  2735 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72da8a80, len: 55296: Operation not permitted [1]
+W0827 04:31:32.252439  2736 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72d15800, len: 55296: Operation not permitted [1]
+W0827 04:31:32.252678  2737 memory_location.cpp:73] Failed to get NUMA node, addr: 0x7315db40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.252701  2738 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72e0e700, len: 13824: Operation not permitted [1]
+W0827 04:31:32.252745  2739 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72f5f480, len: 6912: Operation not permitted [1]
+W0827 04:31:32.252832  2740 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72b295c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.268508  2760 memory_location.cpp:73] Failed to get NUMA node, addr: 0x69eefc00, len: 6912: Operation not permitted [1]
+W0827 04:31:32.268532  2759 memory_location.cpp:73] Failed to get NUMA node, addr: 0x69ec8680, len: 6912: Operation not permitted [1]
+W0827 04:31:32.268577  2761 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6a02e280, len: 6912: Operation not permitted [1]
+W0827 04:31:32.268708  2762 memory_location.cpp:73] Failed to get NUMA node, addr: 0x69d75dc0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.268766  2763 memory_location.cpp:73] Failed to get NUMA node, addr: 0x69fc4880, len: 55296: Operation not permitted [1]
+W0827 04:31:32.268886  2764 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6a0c8540, len: 55296: Operation not permitted [1]
+W0827 04:31:32.268975  2765 memory_location.cpp:73] Failed to get NUMA node, addr: 0x69c75940, len: 6912: Operation not permitted [1]
+W0827 04:31:32.269125  2766 memory_location.cpp:73] Failed to get NUMA node, addr: 0x698ae1c0, len: 13824: Operation not permitted [1]
+W0827 04:31:32.269232  2767 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6a5457c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.269354  2768 memory_location.cpp:73] Failed to get NUMA node, addr: 0x69c8fe80, len: 6912: Operation not permitted [1]
+W0827 04:31:32.281724  2785 memory_location.cpp:73] Failed to get NUMA node, addr: 0x678e47c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.281751  2786 memory_location.cpp:73] Failed to get NUMA node, addr: 0x67a39b80, len: 6912: Operation not permitted [1]
+W0827 04:31:32.281852  2787 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6798d9c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.281994  2788 memory_location.cpp:73] Failed to get NUMA node, addr: 0x67981dc0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.282085  2789 memory_location.cpp:73] Failed to get NUMA node, addr: 0x67823300, len: 55296: Operation not permitted [1]
+W0827 04:31:32.282146  2790 memory_location.cpp:73] Failed to get NUMA node, addr: 0x67a3bb00, len: 55296: Operation not permitted [1]
+W0827 04:31:32.282214  2791 memory_location.cpp:73] Failed to get NUMA node, addr: 0x67cdad40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.282285  2792 memory_location.cpp:73] Failed to get NUMA node, addr: 0x68028e40, len: 13824: Operation not permitted [1]
+W0827 04:31:32.282544  2794 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6796cf40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.282545  2793 memory_location.cpp:73] Failed to get NUMA node, addr: 0x67fa6600, len: 6912: Operation not permitted [1]
+[2026-08-27 04:31:32] Dummy health check server started in background thread at 0.0.0.0:30000
+[rank27]:[W827 04:34:16.127713860 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank25]:[W827 04:34:16.128748517 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank24]:[W827 04:34:16.129831915 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank26]:[W827 04:34:16.136239904 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank26]:[W827 04:35:04.964905528 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank25]:[W827 04:35:04.966225881 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank27]:[W827 04:35:04.966239242 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank24]:[W827 04:35:04.966301664 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank31]:[W827 04:35:04.990701389 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank28]:[W827 04:35:04.991914270 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank29]:[W827 04:35:04.994546241 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank30]:[W827 04:35:04.999020079 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank31]:[W827 04:35:10.873647776 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank30]:[W827 04:35:10.873901815 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank28]:[W827 04:35:10.874227564 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank29]:[W827 04:35:10.885318244 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[2026-08-27 04:35:10 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.29
+[2026-08-27 04:35:10 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.29
+[2026-08-27 04:35:29 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.39
+[2026-08-27 04:35:29 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.39
+[2026-08-27 04:35:31 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25.93
+[2026-08-27 04:35:31 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25.95
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/d_starting_router.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/d_starting_router.log
new file mode 100644
index 0000000..1190be5
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/d_starting_router.log
@@ -0,0 +1 @@
+Error response from daemon: No such container: kimi3_pro6000_pd_pp8_standard_router
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/p_healthy_d_node0.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/p_healthy_d_node0.log
new file mode 100644
index 0000000..48f2248
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/p_healthy_d_node0.log
@@ -0,0 +1 @@
+Error response from daemon: No such container: kimi3_pro6000_pd_pp8_standard_decode_node0
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/p_healthy_d_node1.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/p_healthy_d_node1.log
new file mode 100644
index 0000000..140acb3
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/p_healthy_d_node1.log
@@ -0,0 +1 @@
+Error response from daemon: No such container: kimi3_pro6000_pd_pp8_standard_decode_node1
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/p_healthy_d_node2.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/p_healthy_d_node2.log
new file mode 100644
index 0000000..87ead82
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/p_healthy_d_node2.log
@@ -0,0 +1 @@
+Error response from daemon: No such container: kimi3_pro6000_pd_pp8_standard_decode_node2
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/p_healthy_d_node3.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/p_healthy_d_node3.log
new file mode 100644
index 0000000..32cb479
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/p_healthy_d_node3.log
@@ -0,0 +1 @@
+Error response from daemon: No such container: kimi3_pro6000_pd_pp8_standard_decode_node3
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/p_healthy_mc_master.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/p_healthy_mc_master.log
new file mode 100644
index 0000000..736c023
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/p_healthy_mc_master.log
@@ -0,0 +1,48 @@
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+I0827 04:28:16.835302    99 master.cpp:1366] Master service started on port 50051, max_threads=16, enable_metric_reporting=1, metrics_port=9003, default_kv_lease_ttl=10000, default_kv_soft_pin_ttl=1800000, allow_evict_soft_pinned_objects=1, eviction_ratio=0.05, eviction_high_watermark_ratio=0.9, enable_ha=0, enable_offload=0, enable_kv_events=0, kv_events_bind_endpoint=, kv_events_backend_id=, offload_on_evict=0, offload_force_evict=0, offloading_queue_limit=50000, offload_cap_ratio=0.5, ha_backend_type=etcd, ha_backend_connstring=, etcd_endpoints=, client_ttl=10, rpc_thread_num=16, rpc_port=50051, rpc_address=0.0.0.0, rpc_interface=, rpc_conn_timeout_seconds=0, rpc_enable_tcp_no_delay=1, rpc protocol=tcp, cluster_id=mooncake_cluster, root_fs_dir=, global_file_segment_size=9223372036854775807, memory_allocator=offset, enable_http_metadata_server=0, http_metadata_server_port=8080, http_metadata_server_host=0.0.0.0, enable_metadata_cleanup_on_timeout=0, put_start_discard_timeout_sec=30, put_start_release_timeout_sec=600, max_total_finished_tasks=10000, max_total_pending_tasks=10000, max_total_processing_tasks=10000, pending_task_timeout_sec=300, processing_task_timeout_sec=300, enable_snapshot=0, enable_snapshot_restore=0, snapshot_interval_seconds=600, snapshot_backup_dir=, snapshot_object_store_type=, snapshot_catalog_store_type=, snapshot_retention_count=2, max_retry_attempts=10, enable_cxl=0, cxl_path=/dev/dax0.0, cxl_size=8589934592
+I0827 04:28:16.866606   108 master_service.cpp:1520] Task cleanup thread started
+I0827 04:28:16.867491    99 master_admin_service.cpp:311] Master admin server started on port 9003
+I0827 04:28:16.869802   111 master_admin_service.cpp:302] Master Admin Metrics: role=standby, state=starting, service_ready=false, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:28:26.870203   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:28:36.870568   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:28:46.870873   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:28:56.871232   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:29:06.871615   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:29:16.871969   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:29:26.872316   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:29:36.872687   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:29:46.873068   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:29:56.873467   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:30:06.873914   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:30:16.874214   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:30:26.874567   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:30:36.874953   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:30:46.875332   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:30:56.875741   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:31:06.876119   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:31:16.876497   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:31:26.876866   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:31:36.877233   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:31:46.877605   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:31:56.877948   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:32:06.878437   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:32:16.879210   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:32:26.879711   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:32:36.880080   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:32:46.880565   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:32:56.881004   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:33:06.881398   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:33:16.881786   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:33:26.882170   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:33:36.882550   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:33:46.882939   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:33:56.883315   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:34:06.883746   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:34:16.884127   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:34:26.884568   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:34:36.884953   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:34:46.885334   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:34:56.885706   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:35:06.886085   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:35:16.886464   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:35:26.886835   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/p_healthy_p_node0.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/p_healthy_p_node0.log
new file mode 100644
index 0000000..85cfdec
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/p_healthy_p_node0.log
@@ -0,0 +1,941 @@
+/usr/local/lib/python3.12/dist-packages/torchao/quantization/quant_api.py:1731: SyntaxWarning: invalid escape sequence '\.'
+  """Configuration class for applying different quantization configs to modules or parameters based on their fully qualified names (FQNs).
+/sgl-workspace/sglang/python/sglang/launch_server.py:56: UserWarning: 'python -m sglang.launch_server' is still supported, but 'sglang serve' is the recommended entrypoint.
+  Example: sglang serve --model-path <model> [options]
+  warnings.warn(
+[2026-08-27 04:28:54] Breakable CUDA graph is incompatible with MLA attention; disabling prefill CUDA graph.
+[2026-08-27 04:28:55] Attention backend not specified. Use triton backend by default.
+[2026-08-27 04:28:55] Pipeline parallelism is incompatible with overlap schedule.
+[2026-08-27 04:28:58] server_args=ServerArgs(model_path='/data/hf_models/Kimi-K3', tokenizer_path='/data/hf_models/Kimi-K3', tokenizer_mode='auto', tokenizer_backend='huggingface', tokenizer_worker_num=1, detokenizer_worker_num=1, skip_tokenizer_init=False, load_format='auto', model_loader_extra_config='{}', trust_remote_code=True, context_length=None, is_embedding=False, enable_multimodal=None, revision=None, model_impl='auto', model_config_parser='auto', json_model_override_args='{}', dtype='auto', quantization=None, quantization_param_path=None, kv_cache_dtype='auto', enable_fp32_lm_head=False, modelopt_quant=None, modelopt_checkpoint_restore_path=None, modelopt_checkpoint_save_path=None, modelopt_export_path=None, quantize_and_serve=False, rl_quant_profile=None, enable_tf32_matmul=False, mem_fraction_static=0.88, max_running_requests=None, max_queued_requests=None, max_total_tokens=None, chunked_prefill_size=8192, enable_dynamic_chunking=False, max_prefill_tokens=16384, prefill_max_requests=None, schedule_policy='fcfs', enable_priority_scheduling=False, disable_priority_preemption=False, default_priority_value=None, abort_on_priority_when_disabled=False, schedule_low_priority_values_first=False, priority_scheduling_preemption_threshold=10, retraction_policy='length', schedule_conservativeness=1.0, page_size=64, swa_full_tokens_ratio=0.8, disable_hybrid_swa_memory=False, radix_eviction_policy='lru', prefill_only_disable_kv_cache=False, disable_radix_cache=True, enable_page_major_kv_layout=False, enable_unified_memory=False, disable_chunked_prefix_cache=False, disable_overlap_schedule=True, num_continuous_decode_steps=1, scheduler_recv_interval=1, enable_mixed_chunk=False, nccl_port=None, dist_timeout=3600, dist_init_addr='174.1.60.1:20000', nnodes=4, node_rank=0, tp_size=4, dcp_size=1, pp_size=8, pp_max_micro_batch_size=None, pp_async_batch_depth=0, dp_size=1, load_balance_method='follow_bootstrap_room', attn_cp_size=1, moe_dp_size=1, dcp_comm_backend='ag_rs', dcp_replicate_q_proj=None, dwdp_size=1, enable_prefill_cp=False, cp_strategy=None, enable_dsa_cache_layer_split=False, enable_dsa_prefill_context_parallel=False, dsa_prefill_cp_mode='round-robin-split', enable_prefill_context_parallel=False, prefill_cp_mode='in-seq-split', enable_cp_decode_attn_tp=False, enable_dp_attention=False, enable_dp_attention_local_control_broadcast=False, enable_dp_lm_head=False, enable_attn_tp_input_scattered=False, disable_attn_tp_gather=False, enable_p2p_check=False, device='cuda', base_gpu_id=0, gpu_id_step=1, random_seed=984607766, watchdog_timeout=300, soft_watchdog_timeout=None, sleep_on_idle=False, use_ray=False, custom_sigquit_handler=None, numa_node=None, gc_threshold=None, host='0.0.0.0', port=30000, fastapi_root_path='', smg_grpc_mode=False, grpc_mode=False, grpc_port=None, sidecar=None, sidecar_args=None, skip_server_warmup=False, warmups=None, enable_http2=False, ssl_keyfile=None, ssl_certfile=None, ssl_ca_certs=None, ssl_keyfile_password=None, enable_ssl_refresh=False, api_key=None, admin_api_key=None, served_model_name='kimi-k3', weight_version='default', chat_template=None, hf_chat_template_name=None, completion_template=None, file_storage_path='sglang_storage', enable_cache_report=False, reasoning_parser=None, default_chat_template_kwargs=None, strip_thinking_cache=False, enable_strict_thinking=False, tool_call_parser=None, tool_server=None, sampling_defaults='model', asr_max_buffer_seconds=60, asr_max_concurrent_sessions=32, preferred_sampling_params=None, allow_auto_truncate=False, stream_interval=1, batch_notify_size=16, stream_response_default_include_usage=False, incremental_streaming_output=False, enable_streaming_session=False, enable_session_radix_cache=False, log_level='info', log_level_http=None, log_requests=False, log_requests_level=2, log_requests_format='text', log_requests_target=None, uvicorn_access_log_exclude_prefixes=[], crash_dump_folder=None, show_time_cost=False, enable_metrics=False, smg_http_sidecar_port=None, enable_mfu_metrics=False, enable_metrics_for_all_schedulers=False, load_snapshot_publish_interval=15, tokenizer_metrics_custom_labels_header='x-custom-labels', tokenizer_metrics_allowed_custom_labels=None, extra_metric_labels=None, bucket_time_to_first_token=None, bucket_inter_token_latency=None, bucket_e2e_request_latency=None, prompt_tokens_buckets=None, generation_tokens_buckets=None, gc_warning_threshold_secs=0.0, decode_log_interval=40, enable_request_time_stats_logging=False, kv_events_config=None, enable_forward_pass_metrics=False, forward_pass_metrics_worker_id='', forward_pass_metrics_ipc_name=None, enable_trace=False, trace_modules='request', otlp_traces_endpoint='localhost:4317', export_metrics_to_file=False, export_metrics_to_file_dir=None, stat_loggers=None, constrained_json_whitespace_pattern=None, constrained_json_disable_any_whitespace=False, attention_backend='triton', decode_attention_backend=None, prefill_attention_backend=None, sampling_backend='flashinfer', grammar_backend='xgrammar', radix_cache_backend=None, mm_attention_backend=None, fp8_gemm_runner_backend='auto', fp4_gemm_runner_backend='auto', bf16_gemm_backend='auto', dsa_prefill_backend=None, dsa_decode_backend=None, dsa_paged_mqa_logits_backend='auto', dsa_topk_backend='sgl-kernel', disable_flashinfer_autotune=False, flashinfer_autotune_skip_ops=None, mamba_backend='triton', cuda_graph_config=CudaGraphConfig(decode=PhaseConfig(backend='disabled', max_bs=16, bs=[1, 2, 4, 8, 12, 16], tc_compiler='eager', full_prefill_max_req=None), prefill=PhaseConfig(backend='disabled', max_bs=2048, bs=[4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], tc_compiler='eager', full_prefill_max_req=None)), cuda_graph_backend_decode=None, cuda_graph_backend_prefill=None, cuda_graph_max_bs_decode=16, cuda_graph_max_bs_prefill=None, cuda_graph_bs_decode=None, cuda_graph_bs_prefill=None, cuda_graph_tc_compiler=None, disable_prefill_cuda_graph=False, disable_decode_cuda_graph=False, disable_cuda_graph=False, disable_cuda_graph_padding=False, enable_profile_cuda_graph=False, enable_cudagraph_gc=False, debug_cuda_graph=False, enable_layerwise_nvtx_marker=False, enable_nccl_nvls=False, enable_symm_mem=False, triton_attention_reduce_in_fp32=False, triton_attention_num_kv_splits=8, triton_attention_split_tile_size=None, flashinfer_mla_disable_ragged=False, enable_fused_qk_norm_rope=False, enable_precise_embedding_interpolation=False, enable_fused_moe_sum_all_reduce=False, enable_deepseek_v4_fp4_indexer=False, disable_custom_all_reduce=False, enable_mscclpp=False, enable_torch_symm_mem=False, enable_scattered_sconv=False, pre_warm_nccl=False, enable_quant_communications=False, enable_flashinfer_allreduce_fusion=False, enforce_disable_flashinfer_allreduce_fusion=False, flashinfer_allreduce_fusion_backend=None, enable_aiter_allreduce_fusion=False, enable_torch_compile=False, enable_torch_compile_debug_mode=False, torch_compile_max_bs=32, torchao_config='', speculative_algorithm=None, speculative_draft_model_path=None, speculative_draft_model_revision=None, speculative_draft_load_format=None, speculative_num_steps=None, speculative_eagle_topk=None, speculative_num_draft_tokens=None, speculative_dflash_block_size=None, speculative_dspark_block_size=None, speculative_dspark_sps_table_path=None, speculative_dspark_confidence_sts_path=None, speculative_dspark_align_verify_tokens_to_graph_tier=False, speculative_accept_threshold_single=1.0, speculative_accept_threshold_acc=1.0, speculative_use_rejection_sampling=False, speculative_token_map=None, speculative_attention_mode='prefill', speculative_draft_attention_backend=None, speculative_draft_window_size=None, speculative_moe_runner_backend='flashinfer_mxfp4', speculative_moe_a2a_backend=None, speculative_draft_model_quantization=None, speculative_skip_dp_mlp_sync=False, enable_multi_layer_eagle=False, speculative_adaptive=False, speculative_adaptive_config=None, decoupled_spec_bind_endpoint=None, decoupled_spec_connect_endpoints=None, decoupled_spec_rank=None, decoupled_spec_role='null', spec_trace_dir=None, speculative_ngram_min_bfs_breadth=1, speculative_ngram_max_bfs_breadth=10, speculative_ngram_match_type='BFS', speculative_ngram_max_trie_depth=18, speculative_ngram_capacity=10000000, speculative_ngram_external_corpus_path=None, speculative_ngram_external_sam_budget=0, speculative_ngram_external_corpus_max_tokens=10000000, ep_size=4, moe_a2a_backend='none', moe_runner_backend='flashinfer_mxfp4', flashinfer_mxfp4_moe_precision='default', deepep_mode='auto', fuseep_mode=2, deepep_dispatcher_output_dtype='auto', ep_num_redundant_experts=0, ep_dispatch_algorithm=None, init_expert_location='trivial', enable_eplb=False, eplb_algorithm='auto', eplb_rebalance_num_iterations=1000, eplb_rebalance_layers_per_chunk=None, eplb_min_rebalancing_utilization_threshold=1.0, expert_distribution_recorder_mode=None, expert_distribution_recorder_buffer_size=1000, enable_expert_distribution_metrics=False, deepep_config=None, moe_dense_tp_size=None, elastic_ep_backend=None, enable_elastic_expert_backup=False, mooncake_ib_device=None, enable_waterfill=False, ep_join_mode=None, ep_join_rank_offset=0, elastic_ep_initial_size=None, max_ep_size=None, elastic_ep_scale_timeout=600, elastic_ep_rejoin=False, disable_flashinfer_cutlass_moe_fp4_allgather=False, disable_shared_experts_fusion=False, enforce_shared_experts_fusion=False, max_mamba_cache_size=None, mamba_ssm_dtype=None, mamba_max_states_per_path=-1, enable_mamba_cache_stochastic_rounding=False, mamba_cache_philox_rounds=0, mamba_full_memory_ratio=0.36, mamba_radix_cache_strategy='extra_buffer_lazy', uses_mamba_radix_cache=False, mamba_track_interval=256, enable_int8_mamba_checkpoint=False, int8_mamba_ckpt_size=None, linear_attn_backend='triton', linear_attn_decode_backend=None, linear_attn_prefill_backend=None, linear_attn_verify_backend=None, enable_linear_replayssm=False, linear_replayssm_cache_len=16, enable_linear_replayssm_spec=False, enable_hierarchical_cache=False, hicache_ratio=2.0, hicache_size=0, hicache_write_policy='write_through', hicache_io_backend='kernel', hicache_mem_layout='page_first', hicache_storage_backend=None, hicache_storage_prefetch_policy='timeout', hicache_storage_backend_extra_config=None, enable_hisparse=False, hisparse_config=None, enable_broadcast_mm_inputs_process=False, enable_prefix_mm_cache=False, mm_enable_dp_encoder=False, mm_process_config={}, mm_processor_worker_num=0, mm_io_worker_num=0, limit_mm_data_per_request=None, enable_mm_global_cache=False, disable_fast_image_processor=False, mm_feature_transport='cpu', keep_mm_feature_on_device=False, enable_lora=None, enable_lora_overlap_loading=None, max_lora_rank=None, lora_target_modules=None, lora_paths=None, max_loaded_loras=None, max_loras_per_batch=8, lora_eviction_policy='lru', lora_backend='csgmv', max_lora_chunk_size=16, experts_shared_outer_loras=None, lora_use_virtual_experts=False, lora_strict_loading=False, lora_drain_wait_threshold=0.0, enable_two_batch_overlap=False, enable_single_batch_overlap=False, tbo_token_distribution_threshold=0.48, cpu_offload_gb=0, offload_group_size=-1, offload_num_in_group=1, offload_prefetch_step=1, offload_mode='cpu', enable_lmcache=False, lmcache_config_file=None, enable_flexkv=False, flexkv_config_file=None, kt_weight_path=None, kt_method='AMXINT4', kt_cpuinfer=None, kt_threadpool_count=2, kt_num_gpu_experts=None, kt_max_deferred_experts_per_token=None, dllm_algorithm=None, dllm_algorithm_config=None, dllm_fdfo=True, disaggregation_mode='prefill', disaggregation_transfer_backend='mooncake', disaggregation_bootstrap_port=28800, disaggregation_ib_device='mlx5_0,mlx5_1,mlx5_2,mlx5_3', disaggregation_decode_enable_radix_cache=False, disaggregation_decode_enable_offload_kvcache=False, num_reserved_decode_tokens=512, disaggregation_decode_extra_slots=None, disaggregation_decode_polling_interval=1, optimistic_prefill_attempts=0, encoder_only=False, language_only=False, encoder_transfer_backend='zmq_to_tokenizer', encoder_urls=[], encoder_bootstrap_port=8997, encoder_register_urls=[], enable_adaptive_dispatch_to_encoder=False, enable_pdmux=False, pdmux_config_path=None, sm_group_num=8, custom_weight_loader=[], weight_loader_disable_mmap=False, weight_loader_prefetch_checkpoints=False, weight_loader_prefetch_num_threads=4, weight_loader_drop_cache_after_load=False, remote_instance_weight_loader_seed_instance_ip=None, remote_instance_weight_loader_seed_instance_service_port=None, remote_instance_weight_loader_send_weights_group_ports=None, remote_instance_weight_loader_backend='nccl', remote_instance_weight_loader_start_seed_via_transfer_engine=False, engine_info_bootstrap_port=6789, modelexpress_config=None, download_dir=None, model_checksum=None, delete_ckpt_after_loading=False, decrypted_config_file=None, decrypted_draft_config_file=None, checkpoint_engine_wait_weights_before_ready=False, enable_prefill_delayer=False, prefill_delayer_max_delay_passes=30, prefill_delayer_token_usage_low_watermark=None, prefill_delayer_forward_passes_buckets=None, prefill_delayer_wait_seconds_buckets=None, prefill_delayer_queue_min_ratio=None, prefill_delayer_max_delay_ms=None, min_free_slots_delay=None, enable_deterministic_inference=False, rl_on_policy_target=None, kv_canary='none', kv_canary_real_data='none', kv_canary_sweep_interval=0, enable_dynamic_batch_tokenizer=False, dynamic_batch_tokenizer_batch_size=32, dynamic_batch_tokenizer_batch_timeout=0.002, enable_tokenizer_batch_encode=False, disable_tokenizer_batch_decode=False, debug_tensor_dump_output_folder=None, debug_tensor_dump_layers=None, debug_tensor_dump_input_file=None, enable_memory_saver=False, enable_weights_cpu_backup=False, enable_draft_weights_cpu_backup=False, enable_custom_logit_processor=False, enable_return_hidden_states=False, enable_return_routed_experts=False, enable_return_indexer_topk=False, disable_outlines_disk_cache=False, enable_mis=False, weight_cache_mode='off', weight_cache_socket=None, weight_cache_timeout=1800, forward_hooks=None, msprobe_dump_config=None)
+[2026-08-27 04:28:58] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:58] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:58] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:58] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:58] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:58] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:58] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:58] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:01] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:01] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:01] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:03] Multimodal processor concurrency enabled with 2 isolated worker threads (auto).
+[2026-08-27 04:29:03] CommonKVBootstrapServer started successfully on 0.0.0.0:28800
+[2026-08-27 04:29:03] No HuggingFace chat template found
+[2026-08-27 04:29:03] No chat template found, defaulting to 'string' content format
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:15 PP0 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:15 PP0 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:16] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:16] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:16 PP1 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:16 PP1 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+/sgl-workspace/sglang/python/sglang/srt/utils/hf_transformers/tokenizer.py:423: UserWarning: Using a slow tokenizer. This might cause a significant slowdown. Consider using a fast tokenizer instead.
+  warnings.warn(
+[2026-08-27 04:29:16] Applying special tokens cache patch for Kimi tokenizer: <class 'transformers_modules.Kimi_hyphen_K3.caf40b2773f9029e.tokenization_kimi.TikTokenTokenizer'>
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:16 PP1 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:16 PP1 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:16 PP0 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:16 PP0 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:16 PP1 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:16 PP1 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:17 PP0 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:17 PP0 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:17 PP1 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:17 PP1 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:17 PP0 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:17 PP0 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:17 PP0 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:17.587481   237 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:17.587555   237 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:17.587566   237 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.1 port: 12001
+I0827 04:29:17.587602   237 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.1:16830
+I0827 04:29:17.587723   237 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:17.613691   237 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:17.620473   237 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:17.627439   237 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:17.653980   237 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:17.656663   237 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:17.656692   237 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:17.656746   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:17.663833   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:17.665359   237 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:01
+I0827 04:29:17.665387   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:17.690173   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:17.691654   237 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:01
+I0827 04:29:17.691689   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:17.718215   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:17.719650   237 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:01
+I0827 04:29:17.719686   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:17.746297   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:17.747779   237 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:01
+I0827 04:29:17.747891   237 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:17 PP0 TP1 EP1] Init torch distributed begin.
+[2026-08-27 04:29:17 PP1 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:18.219898   241 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:18.219964   241 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:18.219973   241 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.1 port: 12001
+I0827 04:29:18.220007   241 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.1:15425
+I0827 04:29:18.220109   241 transfer_engine_impl.cpp:258] Auto-discovering topology...
+[2026-08-27 04:29:18 PP1 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+I0827 04:29:18.244532   241 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:18.252967   241 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:18.261787   241 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:18.289944   241 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:18.293610   241 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:18.293633   241 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:18.293681   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.301677   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:18.303126   241 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:01
+I0827 04:29:18.303155   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.330181   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:18.331590   241 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:01
+I0827 04:29:18.331614   241 rdma_context.cpp:205] Using SIEVE endpoint store
+[2026-08-27 04:29:18 PP0 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+I0827 04:29:18.358213   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:18.359581   241 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:01
+I0827 04:29:18.359604   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.386204   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:18.387562   241 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:01
+I0827 04:29:18.387681   241 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:18 PP1 TP1 EP1] Init torch distributed begin.
+[2026-08-27 04:29:18 PP1 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:18.588883   242 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:18.588970   242 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:18.588984   242 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.1 port: 12001
+I0827 04:29:18.589022   242 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.1:15650
+I0827 04:29:18.589143   242 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:18.594419   242 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:18.599439   242 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:18.604553   242 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:18.609825   242 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:18.612036   242 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:18.612056   242 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:18.612097   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.618021   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:18.619308   242 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:01
+I0827 04:29:18.619333   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.646080   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:18.647465   242 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:01
+I0827 04:29:18.647495   242 rdma_context.cpp:205] Using SIEVE endpoint store
+[2026-08-27 04:29:18 PP0 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+I0827 04:29:18.674199   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:18.675598   242 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:01
+I0827 04:29:18.675621   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.682866   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:18.684276   242 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:01
+I0827 04:29:18.684391   242 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:18 PP1 TP2 EP2] Init torch distributed begin.
+[2026-08-27 04:29:18 PP1 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:18.695660   238 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:18.695755   238 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:18.695767   238 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.1 port: 12001
+I0827 04:29:18.695806   238 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.1:15123
+I0827 04:29:18.695909   238 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:18.701093   238 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:18.706161   238 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:18.711323   238 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:18.737910   238 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:18.740556   238 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:18.740602   238 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:18.740651   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.747717   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:18.749193   238 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:01
+I0827 04:29:18.749220   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.774145   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:18.775534   238 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:01
+I0827 04:29:18.775558   238 rdma_context.cpp:205] Using SIEVE endpoint store
+[2026-08-27 04:29:18 PP0 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+I0827 04:29:18.802196   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:18.803623   238 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:01
+I0827 04:29:18.803649   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.810976   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:18.812341   238 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:01
+I0827 04:29:18.812443   238 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:18 PP0 TP2 EP2] Init torch distributed begin.
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:19.084213   243 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:19.084332   243 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:19.084348   243 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.1 port: 12001
+I0827 04:29:19.084400   243 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.1:15411
+I0827 04:29:19.084602   243 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:19.114387   243 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:19.121773   243 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:19.146405   243 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:19.153759   243 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:19.156515   243 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:19.156548   243 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:19.156603   243 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.163982   243 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:19.165694   243 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:01
+I0827 04:29:19.165726   243 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.173307   243 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:19.174904   243 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:01
+I0827 04:29:19.174932   243 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.206454   243 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:19.208142   243 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:01
+I0827 04:29:19.208168   243 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.215909   243 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:19.217583   243 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:01
+I0827 04:29:19.217706   243 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:19 PP1 TP3 EP3] Init torch distributed begin.
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:19.239588   236 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:19.239699   236 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:19.239715   236 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.1 port: 12001
+I0827 04:29:19.239763   236 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.1:15605
+I0827 04:29:19.239933   236 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:19.240546   240 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:19.240644   240 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:19.240658   240 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.1 port: 12001
+I0827 04:29:19.240702   240 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.1:15487
+I0827 04:29:19.240862   240 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:19.242301   239 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:19.242408   239 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:19.242422   239 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.1 port: 12001
+I0827 04:29:19.242470   239 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.1:15552
+I0827 04:29:19.242636   239 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:19.246201   240 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:19.248451   239 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:19.251204   240 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:19.254226   239 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:19.256254   240 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:19.259852   239 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:19.261372   240 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:19.263411   240 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:19.263432   240 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:19.263473   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.265838   239 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:19.268059   239 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:19.268082   239 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:19.268126   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.269106   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:19.270298   236 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:19.270401   240 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:01
+I0827 04:29:19.270423   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.274480   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:19.275991   239 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:01
+I0827 04:29:19.276019   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.276877   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:19.278193   240 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:01
+I0827 04:29:19.278220   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.297820   236 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:19.302485   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:19.302623   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:19.303958   239 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:01
+I0827 04:29:19.303983   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.304121   240 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:01
+I0827 04:29:19.304147   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.311621   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:19.313341   240 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:01
+I0827 04:29:19.313478   240 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:19 PP1 TP0 EP0] Init torch distributed begin.
+I0827 04:29:19.322126   236 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:19.330281   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:19.331759   239 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:01
+I0827 04:29:19.331784   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.339274   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:19.340766   239 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:01
+I0827 04:29:19.340884   239 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:19 PP0 TP3 EP3] Init torch distributed begin.
+I0827 04:29:19.345958   236 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:19.348632   236 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:19.348659   236 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:19.348712   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.355962   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:19.417273   236 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:01
+I0827 04:29:19.417309   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.424894   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:19.426437   236 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:01
+I0827 04:29:19.426491   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.434231   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:19.435868   236 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:01
+I0827 04:29:19.435896   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.443578   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:19.445148   236 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:01
+I0827 04:29:19.445264   236 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:19 PP0 TP0 EP0] Init torch distributed begin.
+[2026-08-27 04:29:26 PP0 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:26 PP1 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:27 PP1 TP2 EP2] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP1 TP0 EP0] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP1 TP3 EP3] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP1 TP1 EP1] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP0 TP3 EP3] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP0 TP0 EP0] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP0 TP1 EP1] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP0 TP2 EP2] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP0 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:27 PP0 TP3 EP3] sglang is using nccl==2.28.9
+[2026-08-27 04:29:27 PP0 TP1 EP1] sglang is using nccl==2.28.9
+[2026-08-27 04:29:27 PP0 TP2 EP2] sglang is using nccl==2.28.9
+[2026-08-27 04:29:27 PP0 TP0 EP0] Init torch distributed ends. elapsed=8.22 s, mem usage=0.46 GB
+[2026-08-27 04:29:27 PP1 TP3 EP3] Init torch distributed ends. elapsed=8.45 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP1 TP1 EP1] Init torch distributed ends. elapsed=9.28 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP1 TP2 EP2] Init torch distributed ends. elapsed=8.98 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP1 TP0 EP0] Init torch distributed ends. elapsed=8.35 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP0 TP2 EP2] Init torch distributed ends. elapsed=8.85 s, mem usage=0.46 GB
+[2026-08-27 04:29:27 PP0 TP3 EP3] Init torch distributed ends. elapsed=8.32 s, mem usage=0.46 GB
+[2026-08-27 04:29:27 PP0 TP1 EP1] Init torch distributed ends. elapsed=9.92 s, mem usage=0.46 GB
+[2026-08-27 04:29:27 PP1 TP0 EP0] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP1 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP0 TP0 EP0] Load weight begin. avail mem=82.12 GB
+[2026-08-27 04:29:27 PP1 TP3 EP3] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP1 TP2 EP2] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP0 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP1 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP1 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP1 TP1 EP1] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP0 TP2 EP2] Load weight begin. avail mem=82.12 GB
+[2026-08-27 04:29:27 PP1 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP0 TP3 EP3] Load weight begin. avail mem=82.12 GB
+[2026-08-27 04:29:27 PP0 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP0 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP1 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:27 PP1 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:27 PP0 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:27 PP1 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:27 PP1 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:27 PP1 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:27 PP0 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:27 PP1 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:27 PP1 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:27 PP1 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:27 PP0 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:27 PP0 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:27 PP0 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:27 PP0 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:27 PP0 TP1 EP1] Load weight begin. avail mem=82.12 GB
+[2026-08-27 04:29:27 PP0 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP0 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP0 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+
Multi-thread loading shards:   0% Completed | 0/96 [00:00<?, ?it/s]
Multi-thread loading shards:   1% Completed | 1/96 [00:00<01:01,  1.55it/s]
Multi-thread loading shards:   2% Completed | 2/96 [00:09<08:49,  5.63s/it]
Multi-thread loading shards:   3% Completed | 3/96 [00:12<06:35,  4.25s/it]
Multi-thread loading shards:   4% Completed | 4/96 [00:14<05:26,  3.55s/it]
Multi-thread loading shards:   5% Completed | 5/96 [00:17<04:43,  3.11s/it]
Multi-thread loading shards:   6% Completed | 6/96 [00:19<04:17,  2.86s/it]
Multi-thread loading shards:   7% Completed | 7/96 [00:21<04:00,  2.71s/it]
Multi-thread loading shards:   8% Completed | 8/96 [00:24<03:49,  2.61s/it]
Multi-thread loading shards:   9% Completed | 9/96 [00:26<03:44,  2.58s/it]
Multi-thread loading shards:  10% Completed | 10/96 [00:29<03:40,  2.56s/it]
Multi-thread loading shards:  11% Completed | 11/96 [00:31<03:36,  2.54s/it]
Multi-thread loading shards:  12% Completed | 12/96 [00:32<02:43,  1.95s/it]
Multi-thread loading shards:  14% Completed | 13/96 [00:32<02:05,  1.52s/it]
Multi-thread loading shards:  15% Completed | 14/96 [00:33<01:33,  1.14s/it]
Multi-thread loading shards:  16% Completed | 15/96 [00:33<01:21,  1.01s/it]
Multi-thread loading shards:  17% Completed | 16/96 [00:34<01:19,  1.00it/s]
Multi-thread loading shards:  18% Completed | 17/96 [00:35<01:02,  1.26it/s]
Multi-thread loading shards:  19% Completed | 18/96 [00:35<00:51,  1.53it/s]
Multi-thread loading shards:  20% Completed | 19/96 [00:35<00:42,  1.80it/s]
Multi-thread loading shards:  21% Completed | 20/96 [00:36<00:37,  2.01it/s]
Multi-thread loading shards:  22% Completed | 21/96 [00:36<00:33,  2.23it/s]
Multi-thread loading shards:  23% Completed | 22/96 [00:36<00:30,  2.43it/s]
Multi-thread loading shards:  24% Completed | 23/96 [00:37<00:27,  2.68it/s]
Multi-thread loading shards:  25% Completed | 24/96 [00:37<00:25,  2.80it/s]
Multi-thread loading shards:  26% Completed | 25/96 [00:37<00:24,  2.89it/s]
Multi-thread loading shards:  27% Completed | 26/96 [00:38<00:23,  2.99it/s]
Multi-thread loading shards:  28% Completed | 27/96 [00:38<00:22,  3.01it/s]
Multi-thread loading shards:  29% Completed | 28/96 [00:38<00:22,  2.99it/s]
Multi-thread loading shards:  30% Completed | 29/96 [00:39<00:22,  2.98it/s]
Multi-thread loading shards:  31% Completed | 30/96 [00:39<00:22,  2.98it/s]
Multi-thread loading shards:  32% Completed | 31/96 [00:39<00:21,  3.01it/s]
Multi-thread loading shards:  33% Completed | 32/96 [00:40<00:21,  3.00it/s]
Multi-thread loading shards:  34% Completed | 33/96 [00:41<00:35,  1.79it/s]
Multi-thread loading shards:  35% Completed | 34/96 [00:41<00:30,  2.02it/s]
Multi-thread loading shards:  36% Completed | 35/96 [00:41<00:27,  2.23it/s]
Multi-thread loading shards:  38% Completed | 36/96 [00:42<00:24,  2.41it/s]
Multi-thread loading shards:  39% Completed | 37/96 [00:42<00:22,  2.57it/s]
Multi-thread loading shards:  40% Completed | 38/96 [00:42<00:21,  2.64it/s]
Multi-thread loading shards:  41% Completed | 39/96 [00:43<00:20,  2.73it/s]
Multi-thread loading shards:  42% Completed | 40/96 [00:43<00:20,  2.80it/s]
Multi-thread loading shards:  43% Completed | 41/96 [00:43<00:19,  2.79it/s]
Multi-thread loading shards:  44% Completed | 42/96 [00:44<00:19,  2.80it/s]
Multi-thread loading shards:  45% Completed | 43/96 [00:44<00:22,  2.39it/s]
Multi-thread loading shards:  46% Completed | 44/96 [00:45<00:20,  2.51it/s]
Multi-thread loading shards:  47% Completed | 45/96 [00:45<00:19,  2.64it/s]
Multi-thread loading shards:  48% Completed | 46/96 [00:45<00:18,  2.67it/s]
Multi-thread loading shards:  49% Completed | 47/96 [00:46<00:18,  2.61it/s]
Multi-thread loading shards:  50% Completed | 48/96 [00:46<00:17,  2.72it/s]
Multi-thread loading shards:  51% Completed | 49/96 [00:47<00:16,  2.84it/s]
Multi-thread loading shards:  52% Completed | 50/96 [00:47<00:16,  2.81it/s]
Multi-thread loading shards:  53% Completed | 51/96 [00:47<00:15,  2.93it/s]
Multi-thread loading shards:  54% Completed | 52/96 [00:48<00:21,  2.05it/s]
Multi-thread loading shards:  55% Completed | 53/96 [00:49<00:33,  1.30it/s]
Multi-thread loading shards:  56% Completed | 54/96 [00:50<00:27,  1.55it/s]
Multi-thread loading shards:  57% Completed | 55/96 [00:50<00:22,  1.82it/s]
Multi-thread loading shards:  58% Completed | 56/96 [00:50<00:19,  2.06it/s]
Multi-thread loading shards:  59% Completed | 57/96 [00:51<00:17,  2.25it/s]
Multi-thread loading shards:  60% Completed | 58/96 [00:51<00:15,  2.40it/s]
Multi-thread loading shards:  61% Completed | 59/96 [00:51<00:14,  2.56it/s]
Multi-thread loading shards:  62% Completed | 60/96 [00:52<00:13,  2.67it/s]
Multi-thread loading shards:  64% Completed | 61/96 [00:52<00:12,  2.77it/s]
Multi-thread loading shards:  65% Completed | 62/96 [00:52<00:11,  2.84it/s]
Multi-thread loading shards:  66% Completed | 63/96 [00:53<00:11,  2.89it/s]
Multi-thread loading shards:  67% Completed | 64/96 [00:53<00:11,  2.87it/s]
Multi-thread loading shards:  68% Completed | 65/96 [00:54<00:10,  2.89it/s]
Multi-thread loading shards:  69% Completed | 66/96 [00:54<00:10,  2.91it/s]
Multi-thread loading shards:  70% Completed | 67/96 [00:54<00:10,  2.90it/s]
Multi-thread loading shards:  71% Completed | 68/96 [00:55<00:09,  2.90it/s]
Multi-thread loading shards:  72% Completed | 69/96 [00:55<00:09,  2.89it/s]
Multi-thread loading shards:  73% Completed | 70/96 [00:56<00:14,  1.85it/s]
Multi-thread loading shards:  74% Completed | 71/96 [00:56<00:11,  2.10it/s]
Multi-thread loading shards:  75% Completed | 72/96 [00:57<00:10,  2.23it/s]
Multi-thread loading shards:  76% Completed | 73/96 [00:57<00:09,  2.39it/s]
Multi-thread loading shards:  77% Completed | 74/96 [00:57<00:08,  2.48it/s]
Multi-thread loading shards:  78% Completed | 75/96 [00:58<00:09,  2.15it/s]
Multi-thread loading shards:  79% Completed | 76/96 [00:59<00:10,  1.89it/s]
Multi-thread loading shards:  80% Completed | 77/96 [00:59<00:09,  2.05it/s]
Multi-thread loading shards:  81% Completed | 78/96 [00:59<00:08,  2.24it/s]
Multi-thread loading shards:  82% Completed | 79/96 [01:00<00:07,  2.39it/s]
Multi-thread loading shards:  83% Completed | 80/96 [01:00<00:06,  2.51it/s]
Multi-thread loading shards:  84% Completed | 81/96 [01:00<00:05,  2.65it/s]
Multi-thread loading shards:  85% Completed | 82/96 [01:01<00:04,  2.84it/s]
Multi-thread loading shards:  86% Completed | 83/96 [01:02<00:06,  1.99it/s]
Multi-thread loading shards:  88% Completed | 84/96 [01:02<00:06,  1.99it/s]
Multi-thread loading shards:  90% Completed | 86/96 [01:02<00:03,  3.27it/s]
Multi-thread loading shards:  92% Completed | 88/96 [01:02<00:01,  4.72it/s]
Multi-thread loading shards:  94% Completed | 90/96 [01:02<00:00,  6.49it/s]
Multi-thread loading shards:  96% Completed | 92/96 [01:03<00:00,  8.37it/s]
Multi-thread loading shards:  98% Completed | 94/96 [01:03<00:00,  8.91it/s]
Multi-thread loading shards: 100% Completed | 96/96 [01:03<00:00,  9.35it/s]
Multi-thread loading shards: 100% Completed | 96/96 [01:03<00:00,  1.51it/s]
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:33 PP0 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:33 PP0 TP2 EP2] Load weight end. elapsed=65.63 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=39.03 GB, mem usage=43.09 GB.
+[2026-08-27 04:30:33 PP0 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:33 PP0 TP0 EP0] Load weight end. elapsed=65.81 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=39.03 GB, mem usage=43.09 GB.
+[2026-08-27 04:30:35 PP0 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:35 PP0 TP3 EP3] Load weight end. elapsed=67.37 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=39.03 GB, mem usage=43.09 GB.
+[2026-08-27 04:30:35 PP0 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:35 PP0 TP1 EP1] Load weight end. elapsed=67.46 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=39.03 GB, mem usage=43.09 GB.
+[2026-08-27 04:30:35 PP0 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:35 PP0 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:35 PP0 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:35 PP0 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:35 PP0 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:35 PP0 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:35 PP0 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:35 PP0 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:36 PP1 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:36 PP1 TP3 EP3] Load weight end. elapsed=68.32 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=36.12 GB, mem usage=45.92 GB.
+[2026-08-27 04:30:36 PP1 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP1 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP1 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP1 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP1 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP1 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP1 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP1 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:37 PP1 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:37 PP1 TP1 EP1] Load weight end. elapsed=69.55 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=36.12 GB, mem usage=45.92 GB.
+[2026-08-27 04:30:37 PP1 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:37 PP1 TP0 EP0] Load weight end. elapsed=69.66 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=36.12 GB, mem usage=45.92 GB.
+[2026-08-27 04:30:39 PP1 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:39 PP1 TP2 EP2] Load weight end. elapsed=71.33 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=36.12 GB, mem usage=45.92 GB.
+[2026-08-27 04:30:39 PP1 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:39 PP1 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:39 PP1 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:39 PP1 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:39 PP1 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:39 PP1 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:39 PP1 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:39 PP1 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:53 PP0 TP0 EP0] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:53 PP1 TP3 EP3] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:53 PP1 TP2 EP2] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:53 PP1 TP1 EP1] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:53 PP1 TP0 EP0] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:53 PP0 TP2 EP2] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:53 PP0 TP3 EP3] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:53 PP0 TP1 EP1] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:53 PP0 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:53 PP1 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:53 PP1 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:53 PP1 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:53 PP0 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:53 PP1 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:53 PP0 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:53 PP0 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:53 PP0 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 8.16 GB
+[2026-08-27 04:30:53 PP0 TP0 EP0] Memory pool end. avail mem=30.14 GB
+[2026-08-27 04:30:53 PP0 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 8.16 GB
+[2026-08-27 04:30:53 PP0 TP2 EP2] Memory pool end. avail mem=30.14 GB
+[2026-08-27 04:30:53 PP1 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:53 PP1 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:53 PP1 TP0 EP0] Memory pool end. avail mem=23.24 GB
+[2026-08-27 04:30:53 PP1 TP2 EP2] Memory pool end. avail mem=23.24 GB
+[2026-08-27 04:30:53 PP0 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 8.16 GB
+[2026-08-27 04:30:53 PP1 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:53 PP0 TP3 EP3] Memory pool end. avail mem=30.14 GB
+[2026-08-27 04:30:53 PP1 TP1 EP1] Memory pool end. avail mem=23.24 GB
+[2026-08-27 04:30:53 PP1 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:53 PP0 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 8.16 GB
+[2026-08-27 04:30:53 PP1 TP3 EP3] Memory pool end. avail mem=23.24 GB
+[2026-08-27 04:30:53 PP0 TP1 EP1] Memory pool end. avail mem=30.14 GB
+[2026-08-27 04:30:54 PP0 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:30:54 PP0 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:30:54 PP0 TP0 EP0] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp0_pp0_dp0.json
+[2026-08-27 04:30:54 PP1 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:30:54 PP1 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:30:54 PP0 TP2 EP2] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp2_pp0_dp0.json
+[2026-08-27 04:30:54 PP1 TP2 EP2] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp2_pp1_dp0.json
+[2026-08-27 04:30:54 PP1 TP0 EP0] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp0_pp1_dp0.json
+[2026-08-27 04:30:54 PP1 TP1 EP1] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp1_pp1_dp0.json
+[2026-08-27 04:30:54 PP1 TP3 EP3] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp3_pp1_dp0.json
+[2026-08-27 04:30:54 PP0 TP1 EP1] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp1_pp0_dp0.json
+[2026-08-27 04:30:54 PP0 TP3 EP3] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp3_pp0_dp0.json
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.46profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.32profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.51profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  6.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.09profile/s]
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.07profile/s]
+[TensorRT-LLM][INFO] Set logger level to INFO
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.48profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.11profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.29profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.37profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.16profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.82profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.18profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.58profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:01<00:02,  1.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.66profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.03profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.49profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:01<00:01,  2.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.25profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:01,  2.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:02<00:00,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:02<00:00,  3.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.20profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.01profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  9.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  9.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  9.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 12.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 12.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 11.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 11.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.78profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.16profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  3.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.75profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.50profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.92profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.68profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  3.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.97profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.87profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 10.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.51profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.74profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 11.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.30profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.70profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.72profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.63profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.53profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.29profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.58profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  8.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 10.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.71profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.62profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.43profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.18profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.29profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.37profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.53profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.37profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.35profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.00profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.71profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.60profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.78profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.79profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  8.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.86profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  8.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.16profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.78profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  3.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.15profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.94profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.93profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.25profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.23profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.97profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.79profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.96profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.64profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.12profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.75profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.12profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.25profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.91profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.05profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.66profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.90profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  9.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.73profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.55profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.49profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.13profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.74profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.17profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  3.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.72profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.57profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.59profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.64profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.83profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 13.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.49profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 12.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  9.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  7.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.32profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.93profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.29profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.72profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.75profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.46profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.51profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  3.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.90profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.16profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.72profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.63profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.83profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.70profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.59profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.55profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  6.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.93profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.33profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.75profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.30profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  6.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  7.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.79profile/s]
+[2026-08-27 04:31:21 PP0 TP1 EP1] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.53profile/s][2026-08-27 04:31:21 PP0 TP0 EP0] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.97profile/s]
+[2026-08-27 04:31:21 PP0 TP3 EP3] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.43profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.06profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s][2026-08-27 04:31:21 PP0 TP2 EP2] FlashInfer autotune completed.
+[2026-08-27 04:31:21 PP0 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:21 PP0 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:21 PP0 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:21 PP0 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:21 PP0 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:21 PP0 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:21 PP0 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:21 PP0 TP0 EP0] max_total_num_tokens=3801408, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=54, context_len=1048576, available_gpu_mem=27.35 GB
+[2026-08-27 04:31:21 PP0 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.86profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.58profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:01<00:07,  1.18s/profile]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  2.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.27profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.55profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  3.96profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.86profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  6.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  7.01profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.23profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.05profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  6.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  7.06profile/s]
+[2026-08-27 04:31:26 PP1 TP1 EP1] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.40profile/s]
+[2026-08-27 04:31:26 PP1 TP2 EP2] FlashInfer autotune completed.
+[2026-08-27 04:31:26 PP1 TP0 EP0] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.81profile/s]
+[2026-08-27 04:31:26 PP1 TP3 EP3] FlashInfer autotune completed.
+[2026-08-27 04:31:26 PP1 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:26 PP1 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:26 PP1 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:26 PP1 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:26 PP1 TP0 EP0] max_total_num_tokens=3801408, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=54, context_len=1048576, available_gpu_mem=20.45 GB
+[2026-08-27 04:31:26 PP1 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:26 PP1 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:26 PP1 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:26 PP1 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+W0827 04:31:30.805795  2607 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fe48940, len: 6912: Operation not permitted [1]
+W0827 04:31:30.805797  2608 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4f684400, len: 6912: Operation not permitted [1]
+W0827 04:31:30.805945  2609 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4f962700, len: 6912: Operation not permitted [1]
+W0827 04:31:30.805979  2610 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4f9e8a00, len: 6912: Operation not permitted [1]
+W0827 04:31:30.806130  2612 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fe5f640, len: 55296: Operation not permitted [1]
+W0827 04:31:30.806133  2611 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4f83f000, len: 55296: Operation not permitted [1]
+W0827 04:31:30.806264  2613 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fe328c0, len: 6912: Operation not permitted [1]
+W0827 04:31:30.806334  2614 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fa75cc0, len: 13824: Operation not permitted [1]
+W0827 04:31:30.806449  2615 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fa87f80, len: 6912: Operation not permitted [1]
+W0827 04:31:30.806493  2616 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fa29a80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.203712  2635 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50c65e80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.203859  2636 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50c6bb40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.203975  2637 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50d1ffc0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.204025  2638 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50a0c300, len: 6912: Operation not permitted [1]
+W0827 04:31:31.204075  2640 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50de1940, len: 55296: Operation not permitted [1]
+W0827 04:31:31.204082  2639 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50dc93c0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.204438  2641 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50c87940, len: 6912: Operation not permitted [1]
+W0827 04:31:31.204561  2642 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50cffb00, len: 13824: Operation not permitted [1]
+W0827 04:31:31.204641  2644 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50dc7380, len: 6912: Operation not permitted [1]
+W0827 04:31:31.204722  2643 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50c243c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.323861  2661 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50b13c40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.323863  2662 memory_location.cpp:73] Failed to get NUMA node, addr: 0x511191c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.323912  2663 memory_location.cpp:73] Failed to get NUMA node, addr: 0x511b37c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.324069  2664 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50b2da80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.324110  2666 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5115c980, len: 55296: Operation not permitted [1]
+W0827 04:31:31.324155  2665 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5116ac40, len: 55296: Operation not permitted [1]
+W0827 04:31:31.324479  2667 memory_location.cpp:73] Failed to get NUMA node, addr: 0x511a7e80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.324587  2668 memory_location.cpp:73] Failed to get NUMA node, addr: 0x511b8cc0, len: 13824: Operation not permitted [1]
+W0827 04:31:31.324621  2669 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50fc6800, len: 6912: Operation not permitted [1]
+W0827 04:31:31.324682  2670 memory_location.cpp:73] Failed to get NUMA node, addr: 0x511bc780, len: 6912: Operation not permitted [1]
+W0827 04:31:31.543082  2689 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4902e440, len: 6912: Operation not permitted [1]
+W0827 04:31:31.543203  2690 memory_location.cpp:73] Failed to get NUMA node, addr: 0x49484200, len: 6912: Operation not permitted [1]
+W0827 04:31:31.543314  2691 memory_location.cpp:73] Failed to get NUMA node, addr: 0x49119600, len: 6912: Operation not permitted [1]
+W0827 04:31:31.543375  2692 memory_location.cpp:73] Failed to get NUMA node, addr: 0x493dc780, len: 6912: Operation not permitted [1]
+W0827 04:31:31.543467  2693 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51ac5fc0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.543601  2694 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51a9e780, len: 55296: Operation not permitted [1]
+W0827 04:31:31.543751  2695 memory_location.cpp:73] Failed to get NUMA node, addr: 0x49149a40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.543912  2696 memory_location.cpp:73] Failed to get NUMA node, addr: 0x518c0840, len: 13824: Operation not permitted [1]
+W0827 04:31:31.543947  2697 memory_location.cpp:73] Failed to get NUMA node, addr: 0x49303540, len: 6912: Operation not permitted [1]
+W0827 04:31:31.543990  2698 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51951a80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681525  2717 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50376e40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681525  2718 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5058ee40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681525  2716 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5008b980, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681548  2715 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fe0e840, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681727  2719 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5040eb80, len: 55296: Operation not permitted [1]
+W0827 04:31:31.681779  2724 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50209ac0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681811  2720 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5041c3c0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.681828  2722 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5048cc40, len: 13824: Operation not permitted [1]
+W0827 04:31:31.681844  2723 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fde7a00, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681851  2721 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50386580, len: 6912: Operation not permitted [1]
+W0827 04:31:31.730942  2745 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fce4240, len: 6912: Operation not permitted [1]
+W0827 04:31:31.730918  2744 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fed7d40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.730950  2743 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fdcaf80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.731007  2748 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5074c400, len: 55296: Operation not permitted [1]
+W0827 04:31:31.731073  2747 memory_location.cpp:73] Failed to get NUMA node, addr: 0x507bd9c0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.731086  2746 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4f931480, len: 6912: Operation not permitted [1]
+W0827 04:31:31.731294  2749 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd958c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.731361  2751 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd31d80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.731361  2750 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4ffabe00, len: 13824: Operation not permitted [1]
+W0827 04:31:31.731566  2752 memory_location.cpp:73] Failed to get NUMA node, addr: 0x508527c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.808167  2778 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd3ce40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.808190  2777 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd20740, len: 6912: Operation not permitted [1]
+W0827 04:31:31.808192  2776 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4ff86c80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.808637  2782 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fe88780, len: 6912: Operation not permitted [1]
+W0827 04:31:31.808724  2780 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fe48c80, len: 55296: Operation not permitted [1]
+W0827 04:31:31.808760  2781 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fe564c0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.809056  2785 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4ffa7e40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.809049  2779 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4feaa740, len: 6912: Operation not permitted [1]
+W0827 04:31:31.809075  2784 memory_location.cpp:73] Failed to get NUMA node, addr: 0x502fabc0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.809239  2783 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4ffa1f80, len: 13824: Operation not permitted [1]
+W0827 04:31:31.851475  2807 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4ff6fcc0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.851596  2808 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fc0a280, len: 6912: Operation not permitted [1]
+W0827 04:31:31.851615  2809 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd7d700, len: 6912: Operation not permitted [1]
+W0827 04:31:31.851732  2810 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd761c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.851951  2811 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fde6cc0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.851970  2812 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4feeb000, len: 55296: Operation not permitted [1]
+W0827 04:31:31.851962  2813 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fbe2680, len: 6912: Operation not permitted [1]
+W0827 04:31:31.851987  2816 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fbf07c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.852048  2815 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4ffa9940, len: 6912: Operation not permitted [1]
+W0827 04:31:31.852046  2814 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd235c0, len: 13824: Operation not permitted [1]
+[2026-08-27 04:31:32] INFO:     Started server process [1]
+[2026-08-27 04:31:32] INFO:     Waiting for application startup.
+[2026-08-27 04:31:32] Calling super().encode with {'add_special_tokens': False}
+[2026-08-27 04:31:32] Calling super().encode with {'add_special_tokens': False}
+[2026-08-27 04:31:32] INFO:     Application startup complete.
+[2026-08-27 04:31:32] INFO:     Uvicorn running on http://0.0.0.0:30000 (Press CTRL+C to quit)
+[2026-08-27 04:31:33] INFO:     127.0.0.1:60348 - "GET /model_info HTTP/1.1" 200 OK
+[2026-08-27 04:31:33] Start of pd disaggregation warmup ...
+[2026-08-27 04:31:37] INFO:     174.1.60.1:33088 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:38] INFO:     174.1.60.1:33092 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:39] INFO:     174.1.60.1:33106 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:40] INFO:     174.1.60.1:33110 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:41] INFO:     174.1.60.1:42028 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:42] INFO:     174.1.60.1:42030 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:43] INFO:     174.1.60.1:42046 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:44] INFO:     174.1.60.1:42050 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:45] INFO:     174.1.60.1:42060 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:46] INFO:     174.1.60.1:42072 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:47] INFO:     174.1.60.1:42086 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:48] INFO:     174.1.60.1:42092 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:49] INFO:     174.1.60.1:42098 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:50] INFO:     174.1.60.1:42100 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:51] INFO:     174.1.60.1:46852 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:52] INFO:     174.1.60.1:46856 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:53] INFO:     174.1.60.1:46872 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:54] INFO:     174.1.60.1:46876 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:55] INFO:     174.1.60.1:46886 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:56] INFO:     174.1.60.1:46890 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:57] INFO:     174.1.60.1:46898 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:58] INFO:     174.1.60.1:46912 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:59] INFO:     174.1.60.1:46928 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:00] INFO:     174.1.60.1:46936 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:01] INFO:     174.1.60.1:57672 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:02] INFO:     174.1.60.1:57676 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:03] INFO:     174.1.60.1:57686 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:04] INFO:     174.1.60.1:57688 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:05] INFO:     174.1.60.1:57696 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:06] INFO:     174.1.60.1:57708 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:12] INFO:     174.1.60.1:39714 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:13] INFO:     174.1.60.1:39728 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:14] INFO:     174.1.60.1:39736 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:15] INFO:     174.1.60.1:39746 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:16] INFO:     174.1.60.1:39760 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:17] INFO:     174.1.60.1:39772 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:18] INFO:     174.1.60.1:39782 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:19] INFO:     174.1.60.1:39786 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:20] INFO:     174.1.60.1:39800 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:21] INFO:     174.1.60.1:41036 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:22] INFO:     174.1.60.1:41052 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:23] INFO:     174.1.60.1:41058 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:24] INFO:     174.1.60.1:41068 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:25] INFO:     174.1.60.1:41074 - "GET /health HTTP/1.1" 503 Service Unavailable
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank1]:[W827 04:32:26.019157362 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank0]:[W827 04:32:26.019963648 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank3]:[W827 04:32:26.020028119 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank2]:[W827 04:32:26.020057305 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank6]:[W827 04:32:26.045976535 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank7]:[W827 04:32:26.046176399 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank5]:[W827 04:32:26.048033943 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank4]:[W827 04:32:26.048331224 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[2026-08-27 04:32:26] INFO:     174.1.60.1:41078 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:27] INFO:     174.1.60.1:41090 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:28] INFO:     174.1.60.1:41102 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:29] INFO:     174.1.60.1:41104 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:30] INFO:     174.1.60.1:41120 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:31] INFO:     174.1.60.1:57002 - "GET /health HTTP/1.1" 503 Service Unavailable
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank6]:[W827 04:32:31.709756552 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank7]:[W827 04:32:31.709839299 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank5]:[W827 04:32:31.709991082 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank4]:[W827 04:32:31.710123265 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[2026-08-27 04:32:32] INFO:     174.1.60.1:57016 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:33] INFO:     174.1.60.1:57030 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:34] INFO:     174.1.60.1:57046 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:35] INFO:     174.1.60.1:57056 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:36] INFO:     174.1.60.1:57066 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:37] INFO:     174.1.60.1:57070 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:38] INFO:     174.1.60.1:57072 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:39] INFO:     174.1.60.1:57082 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:40] INFO:     174.1.60.1:57098 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:41] INFO:     174.1.60.1:56606 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:50] INFO:     174.1.60.1:56620 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:51] INFO:     174.1.60.1:45658 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:52] INFO:     174.1.60.1:45672 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:53] INFO:     174.1.60.1:45674 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:54] INFO:     174.1.60.1:45690 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:55] INFO:     174.1.60.1:45704 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:56] INFO:     174.1.60.1:45708 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:57] INFO:     174.1.60.1:45714 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:58] INFO:     174.1.60.1:45722 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:00] INFO:     174.1.60.1:45736 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:01] INFO:     174.1.60.1:45746 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:02] INFO:     174.1.60.1:33310 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:03] INFO:     174.1.60.1:33320 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:04] INFO:     174.1.60.1:33324 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:05] INFO:     174.1.60.1:33328 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:06] INFO:     174.1.60.1:33336 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:07] INFO:     174.1.60.1:33350 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:08] INFO:     174.1.60.1:33354 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:09] INFO:     174.1.60.1:33356 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:10] INFO:     174.1.60.1:33368 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:11] INFO:     174.1.60.1:39318 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:12] INFO:     174.1.60.1:39328 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:13] INFO:     174.1.60.1:39336 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:14] INFO:     174.1.60.1:39344 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:15] INFO:     174.1.60.1:39346 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:16] INFO:     174.1.60.1:39352 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:17] INFO:     174.1.60.1:39364 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:18] INFO:     174.1.60.1:39378 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:19] INFO:     174.1.60.1:39390 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:20] INFO:     174.1.60.1:39402 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:25] INFO:     174.1.60.1:37792 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:26] INFO:     174.1.60.1:37804 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:27] INFO:     174.1.60.1:37812 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:28] INFO:     174.1.60.1:37826 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:29] INFO:     174.1.60.1:37830 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:30] INFO:     174.1.60.1:37844 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:31] INFO:     174.1.60.1:44208 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:32] INFO:     174.1.60.1:44220 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:33] INFO:     174.1.60.1:44226 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:34] INFO:     174.1.60.1:44228 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:35] INFO:     174.1.60.1:44238 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:37] INFO:     174.1.60.1:44248 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:38] INFO:     174.1.60.1:44258 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:39] INFO:     174.1.60.1:44272 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:40] INFO:     174.1.60.1:44276 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:41] INFO:     174.1.60.1:44194 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:42] INFO:     174.1.60.1:44198 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:43] INFO:     174.1.60.1:44212 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:44] INFO:     174.1.60.1:44216 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:45] INFO:     174.1.60.1:44222 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:46] INFO:     174.1.60.1:44236 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:47] INFO:     174.1.60.1:44246 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:48] INFO:     174.1.60.1:44252 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:49] INFO:     174.1.60.1:44264 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:50] INFO:     174.1.60.1:44278 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:51] INFO:     174.1.60.1:35728 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:52] INFO:     174.1.60.1:35736 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:53] INFO:     174.1.60.1:35752 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:54] INFO:     174.1.60.1:35754 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:55] INFO:     174.1.60.1:35758 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:04] INFO:     174.1.60.1:44514 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:05] INFO:     174.1.60.1:44518 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:06] INFO:     174.1.60.1:44520 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:07] INFO:     174.1.60.1:44526 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:08] INFO:     174.1.60.1:44530 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:09] INFO:     174.1.60.1:44546 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:10] INFO:     174.1.60.1:44550 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:11] INFO:     174.1.60.1:45098 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:12] INFO:     127.0.0.1:59722 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:12] INFO:     174.1.60.1:45106 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:13] INFO:     174.1.60.1:45122 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:14] INFO:     174.1.60.1:45124 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:15] INFO:     174.1.60.1:45136 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:16] INFO:     174.1.60.1:45146 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:17] INFO:     174.1.60.1:45152 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:19] INFO:     174.1.60.1:45158 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:20] INFO:     174.1.60.1:45170 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:21] INFO:     174.1.60.1:45182 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:22] INFO:     174.1.60.1:39946 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:23] INFO:     174.1.60.1:39958 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:24] INFO:     174.1.60.1:39962 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:25] INFO:     174.1.60.1:39978 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:26] INFO:     174.1.60.1:39990 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:27] INFO:     174.1.60.1:39994 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:28] INFO:     174.1.60.1:39996 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:29] INFO:     174.1.60.1:40002 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:30] INFO:     174.1.60.1:40012 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:31] INFO:     174.1.60.1:46978 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:32] INFO:     174.1.60.1:46988 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:33] INFO:     174.1.60.1:46996 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:34] INFO:     174.1.60.1:47004 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:39] INFO:     174.1.60.1:47020 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:40] INFO:     174.1.60.1:47022 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:41] INFO:     174.1.60.1:59158 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:42] INFO:     174.1.60.1:59170 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:43] INFO:     174.1.60.1:59172 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:44] INFO:     174.1.60.1:59180 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:45] INFO:     174.1.60.1:59184 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:46] INFO:     174.1.60.1:59186 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:47] INFO:     174.1.60.1:59200 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:48] INFO:     174.1.60.1:59202 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:49] INFO:     174.1.60.1:59214 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:50] INFO:     174.1.60.1:59220 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:52] INFO:     174.1.60.1:40566 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:53] INFO:     174.1.60.1:40576 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:54] INFO:     174.1.60.1:40588 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:55] INFO:     174.1.60.1:40596 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:56] INFO:     174.1.60.1:40598 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:57] INFO:     174.1.60.1:40604 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:58] INFO:     174.1.60.1:40608 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:59] INFO:     174.1.60.1:40610 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:00] INFO:     174.1.60.1:40622 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:01] INFO:     174.1.60.1:41102 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:02] INFO:     174.1.60.1:41108 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:03] INFO:     174.1.60.1:41122 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:04] INFO:     174.1.60.1:41124 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:05] INFO:     174.1.60.1:41134 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:06] INFO:     174.1.60.1:41140 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:07] INFO:     174.1.60.1:41142 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:08] INFO:     174.1.60.1:41158 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:09] INFO:     174.1.60.1:41160 - "GET /health HTTP/1.1" 503 Service Unavailable
+[rank3]:[W827 04:35:10.359210278 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank2]:[W827 04:35:10.359272236 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank0]:[W827 04:35:10.360047094 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank1]:[W827 04:35:10.370610409 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[2026-08-27 04:35:10 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.28
+[2026-08-27 04:35:10 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.29
+[2026-08-27 04:35:10] INFO:     127.0.0.1:60364 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 04:35:10] Disaggregation warmup requests completed for all 1 DP ranks
+[2026-08-27 04:35:10] End of disaggregation warmup
+[2026-08-27 04:35:10] The server is fired up and ready to roll!
+[2026-08-27 04:35:29 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.39
+[2026-08-27 04:35:29 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.39
+[2026-08-27 04:35:31 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25.83
+[2026-08-27 04:35:31 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25.84
+[2026-08-27 04:35:32] INFO:     174.1.60.1:49902 - "GET /health HTTP/1.1" 200 OK
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/p_healthy_p_node1.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/p_healthy_p_node1.log
new file mode 100644
index 0000000..f2533ec
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/p_healthy_p_node1.log
@@ -0,0 +1,745 @@
+/usr/local/lib/python3.12/dist-packages/torchao/quantization/quant_api.py:1731: SyntaxWarning: invalid escape sequence '\.'
+  """Configuration class for applying different quantization configs to modules or parameters based on their fully qualified names (FQNs).
+/sgl-workspace/sglang/python/sglang/launch_server.py:56: UserWarning: 'python -m sglang.launch_server' is still supported, but 'sglang serve' is the recommended entrypoint.
+  Example: sglang serve --model-path <model> [options]
+  warnings.warn(
+[2026-08-27 04:28:43] Breakable CUDA graph is incompatible with MLA attention; disabling prefill CUDA graph.
+[2026-08-27 04:28:43] Attention backend not specified. Use triton backend by default.
+[2026-08-27 04:28:43] Pipeline parallelism is incompatible with overlap schedule.
+[2026-08-27 04:28:46] server_args=ServerArgs(model_path='/data/hf_models/Kimi-K3', tokenizer_path='/data/hf_models/Kimi-K3', tokenizer_mode='auto', tokenizer_backend='huggingface', tokenizer_worker_num=1, detokenizer_worker_num=1, skip_tokenizer_init=False, load_format='auto', model_loader_extra_config='{}', trust_remote_code=True, context_length=None, is_embedding=False, enable_multimodal=None, revision=None, model_impl='auto', model_config_parser='auto', json_model_override_args='{}', dtype='auto', quantization=None, quantization_param_path=None, kv_cache_dtype='auto', enable_fp32_lm_head=False, modelopt_quant=None, modelopt_checkpoint_restore_path=None, modelopt_checkpoint_save_path=None, modelopt_export_path=None, quantize_and_serve=False, rl_quant_profile=None, enable_tf32_matmul=False, mem_fraction_static=0.88, max_running_requests=None, max_queued_requests=None, max_total_tokens=None, chunked_prefill_size=8192, enable_dynamic_chunking=False, max_prefill_tokens=16384, prefill_max_requests=None, schedule_policy='fcfs', enable_priority_scheduling=False, disable_priority_preemption=False, default_priority_value=None, abort_on_priority_when_disabled=False, schedule_low_priority_values_first=False, priority_scheduling_preemption_threshold=10, retraction_policy='length', schedule_conservativeness=1.0, page_size=64, swa_full_tokens_ratio=0.8, disable_hybrid_swa_memory=False, radix_eviction_policy='lru', prefill_only_disable_kv_cache=False, disable_radix_cache=True, enable_page_major_kv_layout=False, enable_unified_memory=False, disable_chunked_prefix_cache=False, disable_overlap_schedule=True, num_continuous_decode_steps=1, scheduler_recv_interval=1, enable_mixed_chunk=False, nccl_port=None, dist_timeout=3600, dist_init_addr='174.1.60.1:20000', nnodes=4, node_rank=1, tp_size=4, dcp_size=1, pp_size=8, pp_max_micro_batch_size=None, pp_async_batch_depth=0, dp_size=1, load_balance_method='follow_bootstrap_room', attn_cp_size=1, moe_dp_size=1, dcp_comm_backend='ag_rs', dcp_replicate_q_proj=None, dwdp_size=1, enable_prefill_cp=False, cp_strategy=None, enable_dsa_cache_layer_split=False, enable_dsa_prefill_context_parallel=False, dsa_prefill_cp_mode='round-robin-split', enable_prefill_context_parallel=False, prefill_cp_mode='in-seq-split', enable_cp_decode_attn_tp=False, enable_dp_attention=False, enable_dp_attention_local_control_broadcast=False, enable_dp_lm_head=False, enable_attn_tp_input_scattered=False, disable_attn_tp_gather=False, enable_p2p_check=False, device='cuda', base_gpu_id=0, gpu_id_step=1, random_seed=577755302, watchdog_timeout=300, soft_watchdog_timeout=None, sleep_on_idle=False, use_ray=False, custom_sigquit_handler=None, numa_node=None, gc_threshold=None, host='0.0.0.0', port=30000, fastapi_root_path='', smg_grpc_mode=False, grpc_mode=False, grpc_port=None, sidecar=None, sidecar_args=None, skip_server_warmup=False, warmups=None, enable_http2=False, ssl_keyfile=None, ssl_certfile=None, ssl_ca_certs=None, ssl_keyfile_password=None, enable_ssl_refresh=False, api_key=None, admin_api_key=None, served_model_name='kimi-k3', weight_version='default', chat_template=None, hf_chat_template_name=None, completion_template=None, file_storage_path='sglang_storage', enable_cache_report=False, reasoning_parser=None, default_chat_template_kwargs=None, strip_thinking_cache=False, enable_strict_thinking=False, tool_call_parser=None, tool_server=None, sampling_defaults='model', asr_max_buffer_seconds=60, asr_max_concurrent_sessions=32, preferred_sampling_params=None, allow_auto_truncate=False, stream_interval=1, batch_notify_size=16, stream_response_default_include_usage=False, incremental_streaming_output=False, enable_streaming_session=False, enable_session_radix_cache=False, log_level='info', log_level_http=None, log_requests=False, log_requests_level=2, log_requests_format='text', log_requests_target=None, uvicorn_access_log_exclude_prefixes=[], crash_dump_folder=None, show_time_cost=False, enable_metrics=False, smg_http_sidecar_port=None, enable_mfu_metrics=False, enable_metrics_for_all_schedulers=False, load_snapshot_publish_interval=15, tokenizer_metrics_custom_labels_header='x-custom-labels', tokenizer_metrics_allowed_custom_labels=None, extra_metric_labels=None, bucket_time_to_first_token=None, bucket_inter_token_latency=None, bucket_e2e_request_latency=None, prompt_tokens_buckets=None, generation_tokens_buckets=None, gc_warning_threshold_secs=0.0, decode_log_interval=40, enable_request_time_stats_logging=False, kv_events_config=None, enable_forward_pass_metrics=False, forward_pass_metrics_worker_id='', forward_pass_metrics_ipc_name=None, enable_trace=False, trace_modules='request', otlp_traces_endpoint='localhost:4317', export_metrics_to_file=False, export_metrics_to_file_dir=None, stat_loggers=None, constrained_json_whitespace_pattern=None, constrained_json_disable_any_whitespace=False, attention_backend='triton', decode_attention_backend=None, prefill_attention_backend=None, sampling_backend='flashinfer', grammar_backend='xgrammar', radix_cache_backend=None, mm_attention_backend=None, fp8_gemm_runner_backend='auto', fp4_gemm_runner_backend='auto', bf16_gemm_backend='auto', dsa_prefill_backend=None, dsa_decode_backend=None, dsa_paged_mqa_logits_backend='auto', dsa_topk_backend='sgl-kernel', disable_flashinfer_autotune=False, flashinfer_autotune_skip_ops=None, mamba_backend='triton', cuda_graph_config=CudaGraphConfig(decode=PhaseConfig(backend='disabled', max_bs=16, bs=[1, 2, 4, 8, 12, 16], tc_compiler='eager', full_prefill_max_req=None), prefill=PhaseConfig(backend='disabled', max_bs=2048, bs=[4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], tc_compiler='eager', full_prefill_max_req=None)), cuda_graph_backend_decode=None, cuda_graph_backend_prefill=None, cuda_graph_max_bs_decode=16, cuda_graph_max_bs_prefill=None, cuda_graph_bs_decode=None, cuda_graph_bs_prefill=None, cuda_graph_tc_compiler=None, disable_prefill_cuda_graph=False, disable_decode_cuda_graph=False, disable_cuda_graph=False, disable_cuda_graph_padding=False, enable_profile_cuda_graph=False, enable_cudagraph_gc=False, debug_cuda_graph=False, enable_layerwise_nvtx_marker=False, enable_nccl_nvls=False, enable_symm_mem=False, triton_attention_reduce_in_fp32=False, triton_attention_num_kv_splits=8, triton_attention_split_tile_size=None, flashinfer_mla_disable_ragged=False, enable_fused_qk_norm_rope=False, enable_precise_embedding_interpolation=False, enable_fused_moe_sum_all_reduce=False, enable_deepseek_v4_fp4_indexer=False, disable_custom_all_reduce=False, enable_mscclpp=False, enable_torch_symm_mem=False, enable_scattered_sconv=False, pre_warm_nccl=False, enable_quant_communications=False, enable_flashinfer_allreduce_fusion=False, enforce_disable_flashinfer_allreduce_fusion=False, flashinfer_allreduce_fusion_backend=None, enable_aiter_allreduce_fusion=False, enable_torch_compile=False, enable_torch_compile_debug_mode=False, torch_compile_max_bs=32, torchao_config='', speculative_algorithm=None, speculative_draft_model_path=None, speculative_draft_model_revision=None, speculative_draft_load_format=None, speculative_num_steps=None, speculative_eagle_topk=None, speculative_num_draft_tokens=None, speculative_dflash_block_size=None, speculative_dspark_block_size=None, speculative_dspark_sps_table_path=None, speculative_dspark_confidence_sts_path=None, speculative_dspark_align_verify_tokens_to_graph_tier=False, speculative_accept_threshold_single=1.0, speculative_accept_threshold_acc=1.0, speculative_use_rejection_sampling=False, speculative_token_map=None, speculative_attention_mode='prefill', speculative_draft_attention_backend=None, speculative_draft_window_size=None, speculative_moe_runner_backend='flashinfer_mxfp4', speculative_moe_a2a_backend=None, speculative_draft_model_quantization=None, speculative_skip_dp_mlp_sync=False, enable_multi_layer_eagle=False, speculative_adaptive=False, speculative_adaptive_config=None, decoupled_spec_bind_endpoint=None, decoupled_spec_connect_endpoints=None, decoupled_spec_rank=None, decoupled_spec_role='null', spec_trace_dir=None, speculative_ngram_min_bfs_breadth=1, speculative_ngram_max_bfs_breadth=10, speculative_ngram_match_type='BFS', speculative_ngram_max_trie_depth=18, speculative_ngram_capacity=10000000, speculative_ngram_external_corpus_path=None, speculative_ngram_external_sam_budget=0, speculative_ngram_external_corpus_max_tokens=10000000, ep_size=4, moe_a2a_backend='none', moe_runner_backend='flashinfer_mxfp4', flashinfer_mxfp4_moe_precision='default', deepep_mode='auto', fuseep_mode=2, deepep_dispatcher_output_dtype='auto', ep_num_redundant_experts=0, ep_dispatch_algorithm=None, init_expert_location='trivial', enable_eplb=False, eplb_algorithm='auto', eplb_rebalance_num_iterations=1000, eplb_rebalance_layers_per_chunk=None, eplb_min_rebalancing_utilization_threshold=1.0, expert_distribution_recorder_mode=None, expert_distribution_recorder_buffer_size=1000, enable_expert_distribution_metrics=False, deepep_config=None, moe_dense_tp_size=None, elastic_ep_backend=None, enable_elastic_expert_backup=False, mooncake_ib_device=None, enable_waterfill=False, ep_join_mode=None, ep_join_rank_offset=0, elastic_ep_initial_size=None, max_ep_size=None, elastic_ep_scale_timeout=600, elastic_ep_rejoin=False, disable_flashinfer_cutlass_moe_fp4_allgather=False, disable_shared_experts_fusion=False, enforce_shared_experts_fusion=False, max_mamba_cache_size=None, mamba_ssm_dtype=None, mamba_max_states_per_path=-1, enable_mamba_cache_stochastic_rounding=False, mamba_cache_philox_rounds=0, mamba_full_memory_ratio=0.36, mamba_radix_cache_strategy='extra_buffer_lazy', uses_mamba_radix_cache=False, mamba_track_interval=256, enable_int8_mamba_checkpoint=False, int8_mamba_ckpt_size=None, linear_attn_backend='triton', linear_attn_decode_backend=None, linear_attn_prefill_backend=None, linear_attn_verify_backend=None, enable_linear_replayssm=False, linear_replayssm_cache_len=16, enable_linear_replayssm_spec=False, enable_hierarchical_cache=False, hicache_ratio=2.0, hicache_size=0, hicache_write_policy='write_through', hicache_io_backend='kernel', hicache_mem_layout='page_first', hicache_storage_backend=None, hicache_storage_prefetch_policy='timeout', hicache_storage_backend_extra_config=None, enable_hisparse=False, hisparse_config=None, enable_broadcast_mm_inputs_process=False, enable_prefix_mm_cache=False, mm_enable_dp_encoder=False, mm_process_config={}, mm_processor_worker_num=0, mm_io_worker_num=0, limit_mm_data_per_request=None, enable_mm_global_cache=False, disable_fast_image_processor=False, mm_feature_transport='cpu', keep_mm_feature_on_device=False, enable_lora=None, enable_lora_overlap_loading=None, max_lora_rank=None, lora_target_modules=None, lora_paths=None, max_loaded_loras=None, max_loras_per_batch=8, lora_eviction_policy='lru', lora_backend='csgmv', max_lora_chunk_size=16, experts_shared_outer_loras=None, lora_use_virtual_experts=False, lora_strict_loading=False, lora_drain_wait_threshold=0.0, enable_two_batch_overlap=False, enable_single_batch_overlap=False, tbo_token_distribution_threshold=0.48, cpu_offload_gb=0, offload_group_size=-1, offload_num_in_group=1, offload_prefetch_step=1, offload_mode='cpu', enable_lmcache=False, lmcache_config_file=None, enable_flexkv=False, flexkv_config_file=None, kt_weight_path=None, kt_method='AMXINT4', kt_cpuinfer=None, kt_threadpool_count=2, kt_num_gpu_experts=None, kt_max_deferred_experts_per_token=None, dllm_algorithm=None, dllm_algorithm_config=None, dllm_fdfo=True, disaggregation_mode='prefill', disaggregation_transfer_backend='mooncake', disaggregation_bootstrap_port=28800, disaggregation_ib_device='mlx5_0,mlx5_1,mlx5_2,mlx5_3', disaggregation_decode_enable_radix_cache=False, disaggregation_decode_enable_offload_kvcache=False, num_reserved_decode_tokens=512, disaggregation_decode_extra_slots=None, disaggregation_decode_polling_interval=1, optimistic_prefill_attempts=0, encoder_only=False, language_only=False, encoder_transfer_backend='zmq_to_tokenizer', encoder_urls=[], encoder_bootstrap_port=8997, encoder_register_urls=[], enable_adaptive_dispatch_to_encoder=False, enable_pdmux=False, pdmux_config_path=None, sm_group_num=8, custom_weight_loader=[], weight_loader_disable_mmap=False, weight_loader_prefetch_checkpoints=False, weight_loader_prefetch_num_threads=4, weight_loader_drop_cache_after_load=False, remote_instance_weight_loader_seed_instance_ip=None, remote_instance_weight_loader_seed_instance_service_port=None, remote_instance_weight_loader_send_weights_group_ports=None, remote_instance_weight_loader_backend='nccl', remote_instance_weight_loader_start_seed_via_transfer_engine=False, engine_info_bootstrap_port=6789, modelexpress_config=None, download_dir=None, model_checksum=None, delete_ckpt_after_loading=False, decrypted_config_file=None, decrypted_draft_config_file=None, checkpoint_engine_wait_weights_before_ready=False, enable_prefill_delayer=False, prefill_delayer_max_delay_passes=30, prefill_delayer_token_usage_low_watermark=None, prefill_delayer_forward_passes_buckets=None, prefill_delayer_wait_seconds_buckets=None, prefill_delayer_queue_min_ratio=None, prefill_delayer_max_delay_ms=None, min_free_slots_delay=None, enable_deterministic_inference=False, rl_on_policy_target=None, kv_canary='none', kv_canary_real_data='none', kv_canary_sweep_interval=0, enable_dynamic_batch_tokenizer=False, dynamic_batch_tokenizer_batch_size=32, dynamic_batch_tokenizer_batch_timeout=0.002, enable_tokenizer_batch_encode=False, disable_tokenizer_batch_decode=False, debug_tensor_dump_output_folder=None, debug_tensor_dump_layers=None, debug_tensor_dump_input_file=None, enable_memory_saver=False, enable_weights_cpu_backup=False, enable_draft_weights_cpu_backup=False, enable_custom_logit_processor=False, enable_return_hidden_states=False, enable_return_routed_experts=False, enable_return_indexer_topk=False, disable_outlines_disk_cache=False, enable_mis=False, weight_cache_mode='off', weight_cache_socket=None, weight_cache_timeout=1800, forward_hooks=None, msprobe_dump_config=None)
+[2026-08-27 04:28:46] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:46] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:46] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:46] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:46] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:46] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:46] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:46] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:02 PP3 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:02 PP3 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:03 PP3 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:03 PP3 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:03 PP3 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:03 PP3 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:03 PP2 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:03 PP2 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:03 PP2 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:03 PP2 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:04 PP2 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:04 PP2 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:04 PP2 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:04 PP2 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:04 PP3 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:04 PP3 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:05 PP3 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:05 PP2 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:05 PP3 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:05 PP3 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:05 PP2 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:06.044592   241 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:06.044694   241 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:06.044710   241 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.2 port: 12001
+I0827 04:29:06.044759   241 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.2:16343
+I0827 04:29:06.044883   241 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:06.053642   242 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:06.053738   242 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:06.053754   242 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.2 port: 12001
+I0827 04:29:06.053797   242 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.2:15170
+I0827 04:29:06.053925   242 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:06.058039   237 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:06.058135   237 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:06.058149   237 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.2 port: 12001
+I0827 04:29:06.058192   237 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.2:15486
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:06.058215   236 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:06.058310   236 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:06.058316   237 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:06.058326   236 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.2 port: 12001
+I0827 04:29:06.058367   236 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.2:16140
+I0827 04:29:06.058492   236 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:06.059734   239 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:06.059828   239 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:06.059844   239 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.2 port: 12001
+I0827 04:29:06.059890   239 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.2:15755
+I0827 04:29:06.060027   239 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:06.064433   237 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:06.064513   236 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:06.065588   239 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:06.069945   237 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:06.070852   236 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:06.071113   239 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:06.071978   241 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:06.079348   236 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:06.079967   242 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:06.080070   241 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:06.087044   241 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:06.087074   242 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:06.094024   242 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:06.095021   241 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:06.095822   237 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:06.097810   241 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:06.097839   241 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:06.097889   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.099939   239 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:06.100984   242 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:06.103683   242 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:06.103708   242 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:06.103756   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.103785   236 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:06.103833   237 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:06.106014   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:06.106982   236 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:06.107007   236 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:06.107053   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.107446   241 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:02
+I0827 04:29:06.107475   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.107936   239 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:06.107971   237 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:06.107997   237 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:06.108048   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.110760   239 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:06.110790   239 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:06.110847   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.111824   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:06.113214   242 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:02
+I0827 04:29:06.113242   242 rdma_context.cpp:205] Using SIEVE endpoint store
+[2026-08-27 04:29:06 PP2 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+I0827 04:29:06.114821   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:06.115196   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+[2026-08-27 04:29:06 PP2 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+I0827 04:29:06.116189   241 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:02
+I0827 04:29:06.116214   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.116621   237 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:02
+I0827 04:29:06.116650   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.118325   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:06.119874   239 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:02
+I0827 04:29:06.119906   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.121601   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:06.122952   242 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:02
+I0827 04:29:06.122977   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.127393   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:06.128841   239 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:02
+I0827 04:29:06.128867   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.132040   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:06.133428   236 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:02
+I0827 04:29:06.133458   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.136312   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:06.137817   239 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:02
+I0827 04:29:06.137846   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.144285   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:06.144373   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:06.145742   237 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:02
+I0827 04:29:06.145776   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.145833   241 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:02
+I0827 04:29:06.145866   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.152084   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:06.153820   242 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:02
+I0827 04:29:06.153860   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.154352   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:06.155728   241 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:02
+I0827 04:29:06.155835   241 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:06 PP3 TP2 EP2] Init torch distributed begin.
+I0827 04:29:06.163965   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:06.164443   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:06.164605   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:06.165319   236 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:02
+I0827 04:29:06.165345   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.165951   239 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:02
+I0827 04:29:06.166070   239 transfer_engine_impl.cpp:388] installTransport, type=rdma
+I0827 04:29:06.166186   242 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:02
+I0827 04:29:06.166311   242 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:06 PP3 TP0 EP0] Init torch distributed begin.
+[2026-08-27 04:29:06 PP3 TP3 EP3] Init torch distributed begin.
+I0827 04:29:06.172065   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:06.173699   237 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:02
+I0827 04:29:06.173743   237 rdma_context.cpp:205] Using SIEVE endpoint store
+[2026-08-27 04:29:06 PP3 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+I0827 04:29:06.192251   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:06.193763   236 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:02
+I0827 04:29:06.193797   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.200259   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:06.201810   237 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:02
+I0827 04:29:06.201937   237 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:06 PP2 TP2 EP2] Init torch distributed begin.
+I0827 04:29:06.220070   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:06.221505   236 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:02
+I0827 04:29:06.221653   236 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:06 PP2 TP1 EP1] Init torch distributed begin.
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:06.613179   235 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:06.613286   235 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:06.613301   235 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.2 port: 12001
+I0827 04:29:06.613344   235 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.2:16382
+I0827 04:29:06.613492   235 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:06.618471   235 topology.cpp:127] Device mlx5_0 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:06.618719   238 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:06.618824   238 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:06.618839   238 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.2 port: 12001
+I0827 04:29:06.618882   238 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.2:16170
+I0827 04:29:06.619020   238 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:06.624203   238 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:06.629066   238 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:06.633975   238 topology.cpp:127] Device mlx5_2 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:06.636366   240 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:06.636454   240 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:06.636466   240 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.2 port: 12001
+I0827 04:29:06.636509   240 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.2:15840
+I0827 04:29:06.636675   240 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:06.638876   238 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:06.640780   238 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:06.640802   238 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:06.640842   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.641396   240 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:06.643196   235 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:06.645701   240 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:06.649896   235 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:06.650310   240 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:06.654819   240 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:06.656375   240 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:06.656396   240 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:06.656432   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.657156   235 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:06.659863   235 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:06.659895   235 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:06.659950   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.661671   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:06.662956   240 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:02
+I0827 04:29:06.662979   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.664074   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:06.665762   238 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:02
+I0827 04:29:06.665796   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.667351   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:06.668808   235 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:02
+I0827 04:29:06.668838   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.676294   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:06.677721   235 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:02
+I0827 04:29:06.677748   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.692129   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:06.693621   240 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:02
+I0827 04:29:06.693648   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.696089   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:06.697444   238 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:02
+I0827 04:29:06.697470   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.707720   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:06.709362   235 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:02
+I0827 04:29:06.709394   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.719987   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:06.721463   240 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:02
+I0827 04:29:06.721489   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.728108   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:06.728855   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:06.729467   238 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:02
+I0827 04:29:06.729491   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.730392   240 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:02
+I0827 04:29:06.730533   240 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:06 PP3 TP1 EP1] Init torch distributed begin.
+I0827 04:29:06.736888   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:06.738317   238 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:02
+I0827 04:29:06.738422   238 transfer_engine_impl.cpp:388] installTransport, type=rdma
+I0827 04:29:06.738466   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+[2026-08-27 04:29:06 PP2 TP3 EP3] Init torch distributed begin.
+I0827 04:29:06.739869   235 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:02
+I0827 04:29:06.739982   235 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:06 PP2 TP0 EP0] Init torch distributed begin.
+[2026-08-27 04:29:26 PP3 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:26 PP2 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:27 PP2 TP3 EP3] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP2 TP1 EP1] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP2 TP2 EP2] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP2 TP0 EP0] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP3 TP3 EP3] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP3 TP0 EP0] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP3 TP2 EP2] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP3 TP1 EP1] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP3 TP3 EP3] Init torch distributed ends. elapsed=21.61 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP3 TP2 EP2] Init torch distributed ends. elapsed=21.62 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP3 TP1 EP1] Init torch distributed ends. elapsed=21.04 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP3 TP0 EP0] Init torch distributed ends. elapsed=21.61 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP2 TP2 EP2] Init torch distributed ends. elapsed=21.57 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP2 TP1 EP1] Init torch distributed ends. elapsed=21.55 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP2 TP0 EP0] Init torch distributed ends. elapsed=21.03 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP2 TP3 EP3] Init torch distributed ends. elapsed=21.04 s, mem usage=0.54 GB
+[2026-08-27 04:29:28 PP3 TP3 EP3] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP3 TP0 EP0] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP2 TP0 EP0] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP3 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP2 TP1 EP1] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP3 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP2 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP3 TP2 EP2] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP2 TP3 EP3] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP2 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP2 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP3 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP2 TP2 EP2] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP3 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP3 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP2 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP3 TP1 EP1] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP2 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP3 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP2 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP3 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP3 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP2 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP3 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP2 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP3 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP2 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP2 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP2 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP2 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP3 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP3 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:30:31 PP2 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:31 PP2 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:31 PP2 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:31 PP2 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:31 PP2 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:31 PP2 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:31 PP2 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:31 PP2 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP2 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:32 PP2 TP0 EP0] Load weight end. elapsed=64.92 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=36.14 GB, mem usage=45.91 GB.
+[2026-08-27 04:30:33 PP3 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:33 PP3 TP1 EP1] Load weight end. elapsed=65.31 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:34 PP2 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:34 PP2 TP1 EP1] Load weight end. elapsed=66.07 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=36.14 GB, mem usage=45.91 GB.
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:35 PP3 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:35 PP3 TP0 EP0] Load weight end. elapsed=67.41 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:36 PP2 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:36 PP2 TP3 EP3] Load weight end. elapsed=68.04 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=36.14 GB, mem usage=45.91 GB.
+[2026-08-27 04:30:36 PP2 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:36 PP2 TP2 EP2] Load weight end. elapsed=68.23 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=36.14 GB, mem usage=45.91 GB.
+[2026-08-27 04:30:36 PP2 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:36 PP2 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:36 PP2 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:36 PP2 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:36 PP2 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:36 PP2 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:36 PP2 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:36 PP2 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:36 PP3 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:36 PP3 TP3 EP3] Load weight end. elapsed=68.86 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:37 PP3 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:37 PP3 TP2 EP2] Load weight end. elapsed=69.68 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:38 PP3 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:38 PP3 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:38 PP3 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:38 PP3 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:38 PP3 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:38 PP3 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:38 PP3 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:38 PP3 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:54 PP3 TP3 EP3] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP3 TP2 EP2] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP3 TP1 EP1] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP2 TP3 EP3] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP3 TP0 EP0] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP2 TP2 EP2] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP2 TP1 EP1] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP2 TP0 EP0] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP2 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:54 PP2 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:54 PP3 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP3 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP2 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:54 PP3 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP3 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP2 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:54 PP3 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP3 TP3 EP3] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP2 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP2 TP2 EP2] Memory pool end. avail mem=23.25 GB
+[2026-08-27 04:30:54 PP2 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP2 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP2 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP3 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP3 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP2 TP3 EP3] Memory pool end. avail mem=23.25 GB
+[2026-08-27 04:30:54 PP2 TP1 EP1] Memory pool end. avail mem=23.25 GB
+[2026-08-27 04:30:54 PP3 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP3 TP1 EP1] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP3 TP2 EP2] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP2 TP0 EP0] Memory pool end. avail mem=23.25 GB
+[2026-08-27 04:30:54 PP3 TP0 EP0] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP3 TP3 EP3] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp3_pp3_dp0.json
+[2026-08-27 04:30:54 PP2 TP1 EP1] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp1_pp2_dp0.json
+[2026-08-27 04:30:54 PP2 TP2 EP2] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp2_pp2_dp0.json
+[2026-08-27 04:30:54 PP3 TP1 EP1] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp1_pp3_dp0.json
+[2026-08-27 04:30:54 PP2 TP3 EP3] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp3_pp2_dp0.json
+[2026-08-27 04:30:54 PP3 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:30:54 PP3 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:30:54 PP3 TP2 EP2] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp2_pp3_dp0.json
+[2026-08-27 04:30:54 PP2 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:30:54 PP2 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:30:54 PP3 TP0 EP0] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp0_pp3_dp0.json
+[2026-08-27 04:30:54 PP2 TP0 EP0] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp0_pp2_dp0.json
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO[TensorRT-LLM][INFO] Set logger level to INFO
+
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.65profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.60profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.79profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.51profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.86profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.27profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.62profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.98profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.86profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 12.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 13.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.93profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.92profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.09profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.63profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  6.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  7.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.33profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.58profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  8.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.56profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.23profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.63profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.36profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.09profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.97profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  8.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.30profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 13.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 13.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  8.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  8.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 11.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 11.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.23profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.48profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.23profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.53profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.27profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.59profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.62profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.97profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.60profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.46profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.12profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.68profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.77profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.62profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  6.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.87profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.30profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.40profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.83profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.94profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.59profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.66profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.29profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.23profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:01,  3.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.68profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.75profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  7.01profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.22profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.65profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.62profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.46profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.22profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  7.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:01<00:06,  1.03s/profile]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:01,  3.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.50profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.93profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.48profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.11profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.12profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.30profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.68profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  7.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 11.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  9.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 12.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 11.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 11.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 11.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.11profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.31profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.65profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.31profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.35profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.49profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:01<00:01,  2.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.65profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.68profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  7.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.09profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 11.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.88profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.98profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.50profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.62profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.51profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.91profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:01,  3.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.31profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.86profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.27profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.59profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.17profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.76profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.73profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  9.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.37profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.15profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.98profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.65profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.97profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.23profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.46profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  9.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.03profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.57profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.40profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.40profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.57profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.13profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.56profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  6.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.73profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.23profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  3.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.86profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.77profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.92profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.37profile/s]
+[2026-08-27 04:31:26 PP2 TP3 EP3] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.39profile/s]
+[2026-08-27 04:31:26 PP2 TP2 EP2] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.75profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.46profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.35profile/s]
+[2026-08-27 04:31:26 PP2 TP1 EP1] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.12profile/s]
+[2026-08-27 04:31:26 PP2 TP0 EP0] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.66profile/s][2026-08-27 04:31:26 PP2 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:26 PP2 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:26 PP2 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:26 PP2 TP0 EP0] max_total_num_tokens=3801408, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=54, context_len=1048576, available_gpu_mem=20.47 GB
+[2026-08-27 04:31:26 PP2 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:26 PP2 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:26 PP2 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:26 PP2 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:26 PP2 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.93profile/s]
+[2026-08-27 04:31:27 PP3 TP1 EP1] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.16profile/s]
+[2026-08-27 04:31:27 PP3 TP0 EP0] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.43profile/s]
+[2026-08-27 04:31:27 PP3 TP2 EP2] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.05profile/s]
+[2026-08-27 04:31:27 PP3 TP3 EP3] FlashInfer autotune completed.
+[2026-08-27 04:31:27 PP3 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:27 PP3 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:27 PP3 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:27 PP3 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:27 PP3 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:27 PP3 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:27 PP3 TP0 EP0] max_total_num_tokens=3801408, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=54, context_len=1048576, available_gpu_mem=16.54 GB
+[2026-08-27 04:31:27 PP3 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:27 PP3 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+W0827 04:31:30.590376  2521 memory_location.cpp:73] Failed to get NUMA node, addr: 0x513c5ec0, len: 6912: Operation not permitted [1]
+W0827 04:31:30.590432  2522 memory_location.cpp:73] Failed to get NUMA node, addr: 0x514a0900, len: 6912: Operation not permitted [1]
+W0827 04:31:30.590495  2525 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51226dc0, len: 55296: Operation not permitted [1]
+W0827 04:31:30.590492  2523 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5128db40, len: 6912: Operation not permitted [1]
+W0827 04:31:30.590533  2524 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5156ff80, len: 6912: Operation not permitted [1]
+W0827 04:31:30.590600  2526 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51320ac0, len: 55296: Operation not permitted [1]
+W0827 04:31:30.590636  2527 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51768540, len: 6912: Operation not permitted [1]
+W0827 04:31:30.590826  2528 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51483dc0, len: 13824: Operation not permitted [1]
+W0827 04:31:30.590914  2529 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51119380, len: 6912: Operation not permitted [1]
+W0827 04:31:30.590979  2530 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50fe1e00, len: 6912: Operation not permitted [1]
+W0827 04:31:30.846120  2547 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50ceebc0, len: 6912: Operation not permitted [1]
+W0827 04:31:30.846271  2548 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50c57640, len: 6912: Operation not permitted [1]
+W0827 04:31:30.846328  2549 memory_location.cpp:73] Failed to get NUMA node, addr: 0x502ed680, len: 6912: Operation not permitted [1]
+W0827 04:31:30.846354  2550 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50d1c800, len: 6912: Operation not permitted [1]
+W0827 04:31:30.846448  2551 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50510a40, len: 55296: Operation not permitted [1]
+W0827 04:31:30.846473  2552 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5051e280, len: 55296: Operation not permitted [1]
+W0827 04:31:30.846571  2553 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50d01340, len: 6912: Operation not permitted [1]
+W0827 04:31:30.846689  2554 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50ce3440, len: 13824: Operation not permitted [1]
+W0827 04:31:30.846858  2555 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5030f440, len: 6912: Operation not permitted [1]
+W0827 04:31:30.846926  2556 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50cae980, len: 6912: Operation not permitted [1]
+W0827 04:31:31.198441  2575 memory_location.cpp:73] Failed to get NUMA node, addr: 0x514f46c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.198501  2576 memory_location.cpp:73] Failed to get NUMA node, addr: 0x509c6400, len: 6912: Operation not permitted [1]
+W0827 04:31:31.198501  2577 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51513300, len: 6912: Operation not permitted [1]
+W0827 04:31:31.198679  2578 memory_location.cpp:73] Failed to get NUMA node, addr: 0x506eef40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.198787  2579 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50dc9180, len: 55296: Operation not permitted [1]
+W0827 04:31:31.198915  2580 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50904700, len: 55296: Operation not permitted [1]
+W0827 04:31:31.198982  2581 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50aa9440, len: 6912: Operation not permitted [1]
+W0827 04:31:31.199081  2582 memory_location.cpp:73] Failed to get NUMA node, addr: 0x515012c0, len: 13824: Operation not permitted [1]
+W0827 04:31:31.199148  2583 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5161ac80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.199185  2584 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50dd69c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.763885  2603 memory_location.cpp:73] Failed to get NUMA node, addr: 0x519f1500, len: 6912: Operation not permitted [1]
+W0827 04:31:31.763998  2605 memory_location.cpp:73] Failed to get NUMA node, addr: 0x517df940, len: 6912: Operation not permitted [1]
+W0827 04:31:31.763923  2604 memory_location.cpp:73] Failed to get NUMA node, addr: 0x514ab440, len: 6912: Operation not permitted [1]
+W0827 04:31:31.764139  2606 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51556180, len: 6912: Operation not permitted [1]
+W0827 04:31:31.764169  2608 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50ed8140, len: 55296: Operation not permitted [1]
+W0827 04:31:31.764184  2607 memory_location.cpp:73] Failed to get NUMA node, addr: 0x515df040, len: 55296: Operation not permitted [1]
+W0827 04:31:31.764356  2609 memory_location.cpp:73] Failed to get NUMA node, addr: 0x517e3580, len: 6912: Operation not permitted [1]
+W0827 04:31:31.764564  2610 memory_location.cpp:73] Failed to get NUMA node, addr: 0x519cdb80, len: 13824: Operation not permitted [1]
+W0827 04:31:31.764624  2611 memory_location.cpp:73] Failed to get NUMA node, addr: 0x519dbf80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.764714  2612 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50f08d80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.868036  2629 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50a7a680, len: 6912: Operation not permitted [1]
+W0827 04:31:31.868046  2630 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50b6e400, len: 6912: Operation not permitted [1]
+W0827 04:31:31.868139  2631 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50e5f580, len: 6912: Operation not permitted [1]
+W0827 04:31:31.868261  2632 memory_location.cpp:73] Failed to get NUMA node, addr: 0x503c7100, len: 6912: Operation not permitted [1]
+W0827 04:31:31.868330  2633 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50c94ac0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.868432  2635 memory_location.cpp:73] Failed to get NUMA node, addr: 0x509b2880, len: 6912: Operation not permitted [1]
+W0827 04:31:31.868502  2634 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50b14c80, len: 55296: Operation not permitted [1]
+W0827 04:31:31.868564  2636 memory_location.cpp:73] Failed to get NUMA node, addr: 0x503ee840, len: 13824: Operation not permitted [1]
+W0827 04:31:31.868708  2637 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51057d80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.868835  2638 memory_location.cpp:73] Failed to get NUMA node, addr: 0x509cd940, len: 6912: Operation not permitted [1]
+W0827 04:31:32.003983  2656 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50715b00, len: 6912: Operation not permitted [1]
+W0827 04:31:32.004096  2657 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51346e80, len: 6912: Operation not permitted [1]
+W0827 04:31:32.004005  2655 memory_location.cpp:73] Failed to get NUMA node, addr: 0x507176c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.004179  2658 memory_location.cpp:73] Failed to get NUMA node, addr: 0x507e3a40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.004302  2659 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5078aa80, len: 55296: Operation not permitted [1]
+W0827 04:31:32.004357  2660 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5135c980, len: 55296: Operation not permitted [1]
+W0827 04:31:32.004410  2661 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5099f180, len: 6912: Operation not permitted [1]
+W0827 04:31:32.004585  2662 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50b287c0, len: 13824: Operation not permitted [1]
+W0827 04:31:32.004739  2663 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5070b6c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.004812  2664 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50753540, len: 6912: Operation not permitted [1]
+W0827 04:31:32.144867  2688 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50aed700, len: 6912: Operation not permitted [1]
+W0827 04:31:32.144946  2689 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50792740, len: 6912: Operation not permitted [1]
+W0827 04:31:32.145030  2690 memory_location.cpp:73] Failed to get NUMA node, addr: 0x508fca40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.145135  2691 memory_location.cpp:73] Failed to get NUMA node, addr: 0x505d7c40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.145318  2693 memory_location.cpp:73] Failed to get NUMA node, addr: 0x505ec340, len: 55296: Operation not permitted [1]
+W0827 04:31:32.145387  2692 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50602cc0, len: 55296: Operation not permitted [1]
+W0827 04:31:32.145479  2694 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50748c00, len: 6912: Operation not permitted [1]
+W0827 04:31:32.145632  2695 memory_location.cpp:73] Failed to get NUMA node, addr: 0x507e3480, len: 13824: Operation not permitted [1]
+W0827 04:31:32.145767  2696 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50316400, len: 6912: Operation not permitted [1]
+W0827 04:31:32.145864  2697 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50788600, len: 6912: Operation not permitted [1]
+W0827 04:31:32.177534  2726 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fff3c80, len: 6912: Operation not permitted [1]
+W0827 04:31:32.177623  2728 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fad6780, len: 6912: Operation not permitted [1]
+W0827 04:31:32.177625  2727 memory_location.cpp:73] Failed to get NUMA node, addr: 0x501dc2c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.178279  2729 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50596480, len: 6912: Operation not permitted [1]
+W0827 04:31:32.178463  2730 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50222dc0, len: 55296: Operation not permitted [1]
+W0827 04:31:32.178498  2731 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5038f6c0, len: 55296: Operation not permitted [1]
+W0827 04:31:32.178581  2732 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4ffd7ec0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.178695  2733 memory_location.cpp:73] Failed to get NUMA node, addr: 0x506d84c0, len: 13824: Operation not permitted [1]
+W0827 04:31:32.178884  2734 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5078efc0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.178922  2735 memory_location.cpp:73] Failed to get NUMA node, addr: 0x502c7b00, len: 6912: Operation not permitted [1]
+[2026-08-27 04:31:32] Dummy health check server started in background thread at 0.0.0.0:30000
+[rank10]:[W827 04:32:32.605391940 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank11]:[W827 04:32:32.605815131 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank9]:[W827 04:32:32.606041691 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank8]:[W827 04:32:32.608313727 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank11]:[W827 04:33:19.698792619 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank8]:[W827 04:33:19.699392665 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank9]:[W827 04:33:19.699501937 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank10]:[W827 04:33:19.699589651 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank14]:[W827 04:33:19.728965278 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank12]:[W827 04:33:19.729026815 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank13]:[W827 04:33:19.729092243 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank15]:[W827 04:33:19.730463091 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank14]:[W827 04:33:25.763484084 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank13]:[W827 04:33:25.764518868 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank15]:[W827 04:33:25.764776298 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank12]:[W827 04:33:25.765271843 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[2026-08-27 04:35:10 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.29
+[2026-08-27 04:35:10 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.29
+[2026-08-27 04:35:29 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.39
+[2026-08-27 04:35:29 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.39
+[2026-08-27 04:35:31 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25.86
+[2026-08-27 04:35:31 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25.87
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/p_healthy_p_node2.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/p_healthy_p_node2.log
new file mode 100644
index 0000000..934b5d8
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/p_healthy_p_node2.log
@@ -0,0 +1,754 @@
+/usr/local/lib/python3.12/dist-packages/torchao/quantization/quant_api.py:1731: SyntaxWarning: invalid escape sequence '\.'
+  """Configuration class for applying different quantization configs to modules or parameters based on their fully qualified names (FQNs).
+/sgl-workspace/sglang/python/sglang/launch_server.py:56: UserWarning: 'python -m sglang.launch_server' is still supported, but 'sglang serve' is the recommended entrypoint.
+  Example: sglang serve --model-path <model> [options]
+  warnings.warn(
+[2026-08-27 04:28:43] Breakable CUDA graph is incompatible with MLA attention; disabling prefill CUDA graph.
+[2026-08-27 04:28:43] Attention backend not specified. Use triton backend by default.
+[2026-08-27 04:28:43] Pipeline parallelism is incompatible with overlap schedule.
+[2026-08-27 04:28:45] server_args=ServerArgs(model_path='/data/hf_models/Kimi-K3', tokenizer_path='/data/hf_models/Kimi-K3', tokenizer_mode='auto', tokenizer_backend='huggingface', tokenizer_worker_num=1, detokenizer_worker_num=1, skip_tokenizer_init=False, load_format='auto', model_loader_extra_config='{}', trust_remote_code=True, context_length=None, is_embedding=False, enable_multimodal=None, revision=None, model_impl='auto', model_config_parser='auto', json_model_override_args='{}', dtype='auto', quantization=None, quantization_param_path=None, kv_cache_dtype='auto', enable_fp32_lm_head=False, modelopt_quant=None, modelopt_checkpoint_restore_path=None, modelopt_checkpoint_save_path=None, modelopt_export_path=None, quantize_and_serve=False, rl_quant_profile=None, enable_tf32_matmul=False, mem_fraction_static=0.88, max_running_requests=None, max_queued_requests=None, max_total_tokens=None, chunked_prefill_size=8192, enable_dynamic_chunking=False, max_prefill_tokens=16384, prefill_max_requests=None, schedule_policy='fcfs', enable_priority_scheduling=False, disable_priority_preemption=False, default_priority_value=None, abort_on_priority_when_disabled=False, schedule_low_priority_values_first=False, priority_scheduling_preemption_threshold=10, retraction_policy='length', schedule_conservativeness=1.0, page_size=64, swa_full_tokens_ratio=0.8, disable_hybrid_swa_memory=False, radix_eviction_policy='lru', prefill_only_disable_kv_cache=False, disable_radix_cache=True, enable_page_major_kv_layout=False, enable_unified_memory=False, disable_chunked_prefix_cache=False, disable_overlap_schedule=True, num_continuous_decode_steps=1, scheduler_recv_interval=1, enable_mixed_chunk=False, nccl_port=None, dist_timeout=3600, dist_init_addr='174.1.60.1:20000', nnodes=4, node_rank=2, tp_size=4, dcp_size=1, pp_size=8, pp_max_micro_batch_size=None, pp_async_batch_depth=0, dp_size=1, load_balance_method='follow_bootstrap_room', attn_cp_size=1, moe_dp_size=1, dcp_comm_backend='ag_rs', dcp_replicate_q_proj=None, dwdp_size=1, enable_prefill_cp=False, cp_strategy=None, enable_dsa_cache_layer_split=False, enable_dsa_prefill_context_parallel=False, dsa_prefill_cp_mode='round-robin-split', enable_prefill_context_parallel=False, prefill_cp_mode='in-seq-split', enable_cp_decode_attn_tp=False, enable_dp_attention=False, enable_dp_attention_local_control_broadcast=False, enable_dp_lm_head=False, enable_attn_tp_input_scattered=False, disable_attn_tp_gather=False, enable_p2p_check=False, device='cuda', base_gpu_id=0, gpu_id_step=1, random_seed=336932344, watchdog_timeout=300, soft_watchdog_timeout=None, sleep_on_idle=False, use_ray=False, custom_sigquit_handler=None, numa_node=None, gc_threshold=None, host='0.0.0.0', port=30000, fastapi_root_path='', smg_grpc_mode=False, grpc_mode=False, grpc_port=None, sidecar=None, sidecar_args=None, skip_server_warmup=False, warmups=None, enable_http2=False, ssl_keyfile=None, ssl_certfile=None, ssl_ca_certs=None, ssl_keyfile_password=None, enable_ssl_refresh=False, api_key=None, admin_api_key=None, served_model_name='kimi-k3', weight_version='default', chat_template=None, hf_chat_template_name=None, completion_template=None, file_storage_path='sglang_storage', enable_cache_report=False, reasoning_parser=None, default_chat_template_kwargs=None, strip_thinking_cache=False, enable_strict_thinking=False, tool_call_parser=None, tool_server=None, sampling_defaults='model', asr_max_buffer_seconds=60, asr_max_concurrent_sessions=32, preferred_sampling_params=None, allow_auto_truncate=False, stream_interval=1, batch_notify_size=16, stream_response_default_include_usage=False, incremental_streaming_output=False, enable_streaming_session=False, enable_session_radix_cache=False, log_level='info', log_level_http=None, log_requests=False, log_requests_level=2, log_requests_format='text', log_requests_target=None, uvicorn_access_log_exclude_prefixes=[], crash_dump_folder=None, show_time_cost=False, enable_metrics=False, smg_http_sidecar_port=None, enable_mfu_metrics=False, enable_metrics_for_all_schedulers=False, load_snapshot_publish_interval=15, tokenizer_metrics_custom_labels_header='x-custom-labels', tokenizer_metrics_allowed_custom_labels=None, extra_metric_labels=None, bucket_time_to_first_token=None, bucket_inter_token_latency=None, bucket_e2e_request_latency=None, prompt_tokens_buckets=None, generation_tokens_buckets=None, gc_warning_threshold_secs=0.0, decode_log_interval=40, enable_request_time_stats_logging=False, kv_events_config=None, enable_forward_pass_metrics=False, forward_pass_metrics_worker_id='', forward_pass_metrics_ipc_name=None, enable_trace=False, trace_modules='request', otlp_traces_endpoint='localhost:4317', export_metrics_to_file=False, export_metrics_to_file_dir=None, stat_loggers=None, constrained_json_whitespace_pattern=None, constrained_json_disable_any_whitespace=False, attention_backend='triton', decode_attention_backend=None, prefill_attention_backend=None, sampling_backend='flashinfer', grammar_backend='xgrammar', radix_cache_backend=None, mm_attention_backend=None, fp8_gemm_runner_backend='auto', fp4_gemm_runner_backend='auto', bf16_gemm_backend='auto', dsa_prefill_backend=None, dsa_decode_backend=None, dsa_paged_mqa_logits_backend='auto', dsa_topk_backend='sgl-kernel', disable_flashinfer_autotune=False, flashinfer_autotune_skip_ops=None, mamba_backend='triton', cuda_graph_config=CudaGraphConfig(decode=PhaseConfig(backend='disabled', max_bs=16, bs=[1, 2, 4, 8, 12, 16], tc_compiler='eager', full_prefill_max_req=None), prefill=PhaseConfig(backend='disabled', max_bs=2048, bs=[4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], tc_compiler='eager', full_prefill_max_req=None)), cuda_graph_backend_decode=None, cuda_graph_backend_prefill=None, cuda_graph_max_bs_decode=16, cuda_graph_max_bs_prefill=None, cuda_graph_bs_decode=None, cuda_graph_bs_prefill=None, cuda_graph_tc_compiler=None, disable_prefill_cuda_graph=False, disable_decode_cuda_graph=False, disable_cuda_graph=False, disable_cuda_graph_padding=False, enable_profile_cuda_graph=False, enable_cudagraph_gc=False, debug_cuda_graph=False, enable_layerwise_nvtx_marker=False, enable_nccl_nvls=False, enable_symm_mem=False, triton_attention_reduce_in_fp32=False, triton_attention_num_kv_splits=8, triton_attention_split_tile_size=None, flashinfer_mla_disable_ragged=False, enable_fused_qk_norm_rope=False, enable_precise_embedding_interpolation=False, enable_fused_moe_sum_all_reduce=False, enable_deepseek_v4_fp4_indexer=False, disable_custom_all_reduce=False, enable_mscclpp=False, enable_torch_symm_mem=False, enable_scattered_sconv=False, pre_warm_nccl=False, enable_quant_communications=False, enable_flashinfer_allreduce_fusion=False, enforce_disable_flashinfer_allreduce_fusion=False, flashinfer_allreduce_fusion_backend=None, enable_aiter_allreduce_fusion=False, enable_torch_compile=False, enable_torch_compile_debug_mode=False, torch_compile_max_bs=32, torchao_config='', speculative_algorithm=None, speculative_draft_model_path=None, speculative_draft_model_revision=None, speculative_draft_load_format=None, speculative_num_steps=None, speculative_eagle_topk=None, speculative_num_draft_tokens=None, speculative_dflash_block_size=None, speculative_dspark_block_size=None, speculative_dspark_sps_table_path=None, speculative_dspark_confidence_sts_path=None, speculative_dspark_align_verify_tokens_to_graph_tier=False, speculative_accept_threshold_single=1.0, speculative_accept_threshold_acc=1.0, speculative_use_rejection_sampling=False, speculative_token_map=None, speculative_attention_mode='prefill', speculative_draft_attention_backend=None, speculative_draft_window_size=None, speculative_moe_runner_backend='flashinfer_mxfp4', speculative_moe_a2a_backend=None, speculative_draft_model_quantization=None, speculative_skip_dp_mlp_sync=False, enable_multi_layer_eagle=False, speculative_adaptive=False, speculative_adaptive_config=None, decoupled_spec_bind_endpoint=None, decoupled_spec_connect_endpoints=None, decoupled_spec_rank=None, decoupled_spec_role='null', spec_trace_dir=None, speculative_ngram_min_bfs_breadth=1, speculative_ngram_max_bfs_breadth=10, speculative_ngram_match_type='BFS', speculative_ngram_max_trie_depth=18, speculative_ngram_capacity=10000000, speculative_ngram_external_corpus_path=None, speculative_ngram_external_sam_budget=0, speculative_ngram_external_corpus_max_tokens=10000000, ep_size=4, moe_a2a_backend='none', moe_runner_backend='flashinfer_mxfp4', flashinfer_mxfp4_moe_precision='default', deepep_mode='auto', fuseep_mode=2, deepep_dispatcher_output_dtype='auto', ep_num_redundant_experts=0, ep_dispatch_algorithm=None, init_expert_location='trivial', enable_eplb=False, eplb_algorithm='auto', eplb_rebalance_num_iterations=1000, eplb_rebalance_layers_per_chunk=None, eplb_min_rebalancing_utilization_threshold=1.0, expert_distribution_recorder_mode=None, expert_distribution_recorder_buffer_size=1000, enable_expert_distribution_metrics=False, deepep_config=None, moe_dense_tp_size=None, elastic_ep_backend=None, enable_elastic_expert_backup=False, mooncake_ib_device=None, enable_waterfill=False, ep_join_mode=None, ep_join_rank_offset=0, elastic_ep_initial_size=None, max_ep_size=None, elastic_ep_scale_timeout=600, elastic_ep_rejoin=False, disable_flashinfer_cutlass_moe_fp4_allgather=False, disable_shared_experts_fusion=False, enforce_shared_experts_fusion=False, max_mamba_cache_size=None, mamba_ssm_dtype=None, mamba_max_states_per_path=-1, enable_mamba_cache_stochastic_rounding=False, mamba_cache_philox_rounds=0, mamba_full_memory_ratio=0.36, mamba_radix_cache_strategy='extra_buffer_lazy', uses_mamba_radix_cache=False, mamba_track_interval=256, enable_int8_mamba_checkpoint=False, int8_mamba_ckpt_size=None, linear_attn_backend='triton', linear_attn_decode_backend=None, linear_attn_prefill_backend=None, linear_attn_verify_backend=None, enable_linear_replayssm=False, linear_replayssm_cache_len=16, enable_linear_replayssm_spec=False, enable_hierarchical_cache=False, hicache_ratio=2.0, hicache_size=0, hicache_write_policy='write_through', hicache_io_backend='kernel', hicache_mem_layout='page_first', hicache_storage_backend=None, hicache_storage_prefetch_policy='timeout', hicache_storage_backend_extra_config=None, enable_hisparse=False, hisparse_config=None, enable_broadcast_mm_inputs_process=False, enable_prefix_mm_cache=False, mm_enable_dp_encoder=False, mm_process_config={}, mm_processor_worker_num=0, mm_io_worker_num=0, limit_mm_data_per_request=None, enable_mm_global_cache=False, disable_fast_image_processor=False, mm_feature_transport='cpu', keep_mm_feature_on_device=False, enable_lora=None, enable_lora_overlap_loading=None, max_lora_rank=None, lora_target_modules=None, lora_paths=None, max_loaded_loras=None, max_loras_per_batch=8, lora_eviction_policy='lru', lora_backend='csgmv', max_lora_chunk_size=16, experts_shared_outer_loras=None, lora_use_virtual_experts=False, lora_strict_loading=False, lora_drain_wait_threshold=0.0, enable_two_batch_overlap=False, enable_single_batch_overlap=False, tbo_token_distribution_threshold=0.48, cpu_offload_gb=0, offload_group_size=-1, offload_num_in_group=1, offload_prefetch_step=1, offload_mode='cpu', enable_lmcache=False, lmcache_config_file=None, enable_flexkv=False, flexkv_config_file=None, kt_weight_path=None, kt_method='AMXINT4', kt_cpuinfer=None, kt_threadpool_count=2, kt_num_gpu_experts=None, kt_max_deferred_experts_per_token=None, dllm_algorithm=None, dllm_algorithm_config=None, dllm_fdfo=True, disaggregation_mode='prefill', disaggregation_transfer_backend='mooncake', disaggregation_bootstrap_port=28800, disaggregation_ib_device='mlx5_0,mlx5_1,mlx5_2,mlx5_3', disaggregation_decode_enable_radix_cache=False, disaggregation_decode_enable_offload_kvcache=False, num_reserved_decode_tokens=512, disaggregation_decode_extra_slots=None, disaggregation_decode_polling_interval=1, optimistic_prefill_attempts=0, encoder_only=False, language_only=False, encoder_transfer_backend='zmq_to_tokenizer', encoder_urls=[], encoder_bootstrap_port=8997, encoder_register_urls=[], enable_adaptive_dispatch_to_encoder=False, enable_pdmux=False, pdmux_config_path=None, sm_group_num=8, custom_weight_loader=[], weight_loader_disable_mmap=False, weight_loader_prefetch_checkpoints=False, weight_loader_prefetch_num_threads=4, weight_loader_drop_cache_after_load=False, remote_instance_weight_loader_seed_instance_ip=None, remote_instance_weight_loader_seed_instance_service_port=None, remote_instance_weight_loader_send_weights_group_ports=None, remote_instance_weight_loader_backend='nccl', remote_instance_weight_loader_start_seed_via_transfer_engine=False, engine_info_bootstrap_port=6789, modelexpress_config=None, download_dir=None, model_checksum=None, delete_ckpt_after_loading=False, decrypted_config_file=None, decrypted_draft_config_file=None, checkpoint_engine_wait_weights_before_ready=False, enable_prefill_delayer=False, prefill_delayer_max_delay_passes=30, prefill_delayer_token_usage_low_watermark=None, prefill_delayer_forward_passes_buckets=None, prefill_delayer_wait_seconds_buckets=None, prefill_delayer_queue_min_ratio=None, prefill_delayer_max_delay_ms=None, min_free_slots_delay=None, enable_deterministic_inference=False, rl_on_policy_target=None, kv_canary='none', kv_canary_real_data='none', kv_canary_sweep_interval=0, enable_dynamic_batch_tokenizer=False, dynamic_batch_tokenizer_batch_size=32, dynamic_batch_tokenizer_batch_timeout=0.002, enable_tokenizer_batch_encode=False, disable_tokenizer_batch_decode=False, debug_tensor_dump_output_folder=None, debug_tensor_dump_layers=None, debug_tensor_dump_input_file=None, enable_memory_saver=False, enable_weights_cpu_backup=False, enable_draft_weights_cpu_backup=False, enable_custom_logit_processor=False, enable_return_hidden_states=False, enable_return_routed_experts=False, enable_return_indexer_topk=False, disable_outlines_disk_cache=False, enable_mis=False, weight_cache_mode='off', weight_cache_socket=None, weight_cache_timeout=1800, forward_hooks=None, msprobe_dump_config=None)
+[2026-08-27 04:28:45] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:45] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:45] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:45] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:45] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:45] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:45] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:45] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:01 PP4 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:01 PP4 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:01 PP5 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:01 PP5 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:01 PP4 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:01 PP4 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:01 PP5 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:01 PP5 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:01 PP4 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:01 PP4 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:01 PP4 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:01 PP4 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:02 PP5 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:02 PP5 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:02 PP5 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:02 PP5 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:03 PP5 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:03 PP5 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:03 PP4 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:03 PP4 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:03 PP5 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:03 PP5 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:03 PP4 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:03 PP4 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:05.195727   240 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:05.195828   240 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:05.195843   240 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.3 port: 12001
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:05.195847   236 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:05.195899   240 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.3:15652
+I0827 04:29:05.195935   236 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:05.195948   236 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.3 port: 12001
+I0827 04:29:05.195989   236 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.3:15032
+I0827 04:29:05.196039   240 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:05.196110   236 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:05.206050   241 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:05.206164   241 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:05.206182   241 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.3 port: 12001
+I0827 04:29:05.206233   241 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.3:16487
+I0827 04:29:05.206414   241 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:05.206894   238 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:05.207010   238 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:05.207029   238 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.3 port: 12001
+I0827 04:29:05.207082   238 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.3:15213
+I0827 04:29:05.207248   238 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:05.213657   238 topology.cpp:127] Device mlx5_0 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:05.215921   237 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:05.216045   237 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:05.216063   237 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.3 port: 12001
+I0827 04:29:05.216125   237 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.3:16450
+I0827 04:29:05.216308   237 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:05.219630   235 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:05.219745   235 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:05.219763   235 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.3 port: 12001
+I0827 04:29:05.219815   235 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.3:15658
+I0827 04:29:05.219810   236 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:05.220000   235 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:05.222787   237 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:05.223301   240 topology.cpp:127] Device mlx5_0 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:05.223796   239 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+W0827 04:29:05.223798   242 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:05.223948   242 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:05.223948   239 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:05.223969   242 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.3 port: 12001
+I0827 04:29:05.223973   239 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.3 port: 12001
+I0827 04:29:05.224030   242 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.3:16182
+I0827 04:29:05.224030   239 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.3:15581
+I0827 04:29:05.224237   242 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:05.224242   239 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:05.226227   235 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:05.227162   236 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:05.228888   237 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:05.229261   240 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:05.232141   235 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:05.232787   239 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:05.232861   242 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:05.233434   241 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:05.233938   236 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:05.236400   237 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:05.236641   240 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:05.238986   235 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:05.241036   236 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:05.241355   238 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:05.241638   241 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:05.242061   239 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:05.242151   242 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:05.244179   236 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:05.244223   236 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:05.244287   237 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:05.244297   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.244899   240 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:05.245494   235 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:05.249099   237 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:05.249130   237 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:05.249187   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.251125   235 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:05.251137   240 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:05.251168   235 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:05.251168   240 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:05.251241   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.251241   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.251624   239 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:05.251729   242 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:05.251823   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:05.253651   236 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:03
+I0827 04:29:05.253687   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.256857   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:05.257697   239 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:05.258617   237 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:03
+I0827 04:29:05.258652   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.258757   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:05.258822   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:05.260108   239 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:05.260136   239 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:05.260191   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.260488   235 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:03
+I0827 04:29:05.260502   240 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:03
+I0827 04:29:05.260530   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.260533   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.262187   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:05.264094   236 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:03
+I0827 04:29:05.264137   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.266639   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:05.268234   237 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:03
+I0827 04:29:05.268261   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.268549   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:05.269187   241 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:05.269232   238 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:05.270207   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:05.270285   240 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:03
+I0827 04:29:05.270316   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.271827   235 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:03
+I0827 04:29:05.271884   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.275274   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:05.276836   238 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:05.276907   242 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:05.276996   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:05.277058   236 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:03
+I0827 04:29:05.277101   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.278084   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:05.278632   237 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:03
+I0827 04:29:05.278667   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.279789   240 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:03
+I0827 04:29:05.279817   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.282152   238 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:05.282193   238 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:05.282260   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.283128   242 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:05.283200   242 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:05.283263   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.286593   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:05.287469   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:05.288283   239 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:03
+I0827 04:29:05.288322   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.289064   240 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:03
+I0827 04:29:05.289194   240 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:05 PP5 TP1 EP1] Init torch distributed begin.
+I0827 04:29:05.290841   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:05.292457   242 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:03
+I0827 04:29:05.292488   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.292999   241 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:05.296203   241 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:05.296272   241 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:05.296341   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.300206   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:05.301280   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:05.301775   242 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:03
+I0827 04:29:05.301801   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.303004   235 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:03
+I0827 04:29:05.303042   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.303999   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:05.305142   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:05.305728   241 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:03
+I0827 04:29:05.305768   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.305974   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:05.306758   237 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:03
+I0827 04:29:05.306921   237 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:05 PP4 TP2 EP2] Init torch distributed begin.
+I0827 04:29:05.307497   236 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:03
+I0827 04:29:05.307627   236 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:05 PP4 TP1 EP1] Init torch distributed begin.
+I0827 04:29:05.309624   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:05.309773   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:05.311504   238 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:03
+I0827 04:29:05.311542   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.311590   242 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:03
+I0827 04:29:05.311633   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.313076   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:05.314762   239 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:03
+I0827 04:29:05.314813   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.319522   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:05.319566   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:05.321249   242 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:03
+I0827 04:29:05.321398   242 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:05 PP5 TP3 EP3] Init torch distributed begin.
+I0827 04:29:05.326727   238 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:03
+I0827 04:29:05.326759   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.329339   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:05.331007   235 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:03
+I0827 04:29:05.331142   235 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:05 PP4 TP0 EP0] Init torch distributed begin.
+I0827 04:29:05.334666   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:05.336285   238 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:03
+I0827 04:29:05.336313   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.337296   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:05.338828   241 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:03
+I0827 04:29:05.338877   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.344906   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:05.346731   239 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:03
+I0827 04:29:05.346776   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.365299   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:05.366928   238 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:03
+I0827 04:29:05.367062   238 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:05 PP4 TP3 EP3] Init torch distributed begin.
+I0827 04:29:05.368464   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:05.369644   241 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:03
+I0827 04:29:05.369663   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.377357   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:05.379019   239 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:03
+I0827 04:29:05.379146   239 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:05 PP5 TP0 EP0] Init torch distributed begin.
+I0827 04:29:05.396831   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:05.398531   241 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:03
+I0827 04:29:05.398662   241 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:05 PP5 TP2 EP2] Init torch distributed begin.
+[2026-08-27 04:29:26 PP5 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:26 PP4 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:27 PP4 TP1 EP1] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP4 TP3 EP3] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP4 TP2 EP2] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP4 TP0 EP0] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP5 TP2 EP2] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP5 TP3 EP3] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP5 TP0 EP0] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP5 TP1 EP1] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP5 TP0 EP0] Init torch distributed ends. elapsed=22.37 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP5 TP3 EP3] Init torch distributed ends. elapsed=22.42 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP5 TP2 EP2] Init torch distributed ends. elapsed=22.35 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP4 TP1 EP1] Init torch distributed ends. elapsed=22.44 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP5 TP1 EP1] Init torch distributed ends. elapsed=22.46 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP4 TP3 EP3] Init torch distributed ends. elapsed=22.38 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP4 TP2 EP2] Init torch distributed ends. elapsed=22.44 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP4 TP0 EP0] Init torch distributed ends. elapsed=22.41 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP5 TP0 EP0] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP4 TP1 EP1] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP4 TP3 EP3] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP4 TP2 EP2] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP5 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP4 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP4 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP5 TP2 EP2] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP4 TP0 EP0] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP4 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP5 TP3 EP3] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP5 TP1 EP1] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP4 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP5 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP5 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP5 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP4 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP5 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP4 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP5 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP5 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP4 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP4 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP5 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP5 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP4 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP4 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP4 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP5 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP5 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP4 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP5 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:30:33 PP4 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:33 PP4 TP2 EP2] Load weight end. elapsed=65.50 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:34 PP4 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:34 PP4 TP3 EP3] Load weight end. elapsed=66.61 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP4 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:36 PP4 TP1 EP1] Load weight end. elapsed=68.45 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:36 PP5 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:36 PP5 TP0 EP0] Load weight end. elapsed=69.00 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:37 PP5 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:37 PP5 TP3 EP3] Load weight end. elapsed=69.63 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:37 PP5 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:37 PP5 TP1 EP1] Load weight end. elapsed=69.72 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:38 PP5 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:38 PP5 TP2 EP2] Load weight end. elapsed=70.08 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:38 PP5 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:38 PP5 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:38 PP5 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:38 PP5 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:38 PP5 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:38 PP5 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:38 PP5 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:38 PP5 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:47 PP4 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:47 PP4 TP0 EP0] Load weight end. elapsed=79.31 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:47 PP4 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:47 PP4 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:47 PP4 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:47 PP4 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:47 PP4 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:47 PP4 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:47 PP4 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:47 PP4 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:54 PP5 TP3 EP3] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP5 TP1 EP1] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP5 TP2 EP2] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP5 TP0 EP0] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP4 TP3 EP3] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP4 TP2 EP2] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP4 TP1 EP1] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP4 TP0 EP0] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP5 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP5 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP4 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP4 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP4 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP5 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP4 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP5 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP5 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP5 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP4 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP5 TP3 EP3] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP4 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP4 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP4 TP1 EP1] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP5 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP5 TP2 EP2] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP4 TP3 EP3] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP4 TP0 EP0] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP5 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP5 TP1 EP1] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP5 TP0 EP0] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP4 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP4 TP2 EP2] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP4 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:30:54 PP4 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:30:54 PP4 TP0 EP0] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp0_pp4_dp0.json
+[2026-08-27 04:30:54 PP4 TP1 EP1] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp1_pp4_dp0.json
+[2026-08-27 04:30:54 PP5 TP2 EP2] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp2_pp5_dp0.json
+[2026-08-27 04:30:54 PP5 TP3 EP3] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp3_pp5_dp0.json
+[2026-08-27 04:30:54 PP5 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:30:54 PP5 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:30:54 PP5 TP0 EP0] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp0_pp5_dp0.json
+[2026-08-27 04:30:54 PP4 TP2 EP2] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp2_pp4_dp0.json
+[2026-08-27 04:30:54 PP4 TP3 EP3] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp3_pp4_dp0.json
+[2026-08-27 04:30:54 PP5 TP1 EP1] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp1_pp5_dp0.json
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO[TensorRT-LLM][INFO] Set logger level to INFO
+
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.37profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.12profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.87profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.32profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.91profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.77profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.94profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.88profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.20profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.87profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  7.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 10.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.46profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.67profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.91profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.29profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.49profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  7.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.30profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.94profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.47profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.12profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.06profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.35profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  9.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:01<00:02,  2.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.89profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.58profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.32profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.70profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.20profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.27profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  6.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  7.03profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.56profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.73profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.88profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.37profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:05,  1.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:01<00:02,  2.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.05profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.19profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.77profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.18profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.25profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  3.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.18profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.20profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.37profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  9.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.27profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  8.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.43profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.89profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.35profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.96profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.90profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.30profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.72profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.18profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.33profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.51profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  2.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.03profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.68profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.16profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.01profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.24profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 10.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:02,  2.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.06profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.83profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.35profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.87profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.19profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  3.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.06profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.86profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 13.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 13.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.13profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:01,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:01,  3.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.47profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  6.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.47profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.57profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.22profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.17profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.50profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.71profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.19profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.55profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.90profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:05,  1.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:01<00:06,  1.00s/profile]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.33profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.19profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.03profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.76profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.28profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.32profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.15profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.79profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.31profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.18profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 11.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:01,  3.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.06profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:01,  3.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  6.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  3.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.05profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.60profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.50profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.92profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.00profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.29profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 11.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.09profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.43profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.11profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.91profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  7.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:02,  2.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.18profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.78profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.75profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.89profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.64profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.05profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.47profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.47profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.65profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:05,  1.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.85profile/s]
+[2026-08-27 04:31:28 PP5 TP1 EP1] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.21profile/s]
+[2026-08-27 04:31:28 PP5 TP0 EP0] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.74profile/s]
+[2026-08-27 04:31:28 PP5 TP2 EP2] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.77profile/s]
+[2026-08-27 04:31:28 PP5 TP3 EP3] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.19profile/s][2026-08-27 04:31:28 PP5 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:28 PP5 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:28 PP5 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:28 PP5 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:28 PP5 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:28 PP5 TP0 EP0] max_total_num_tokens=3801408, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=54, context_len=1048576, available_gpu_mem=16.30 GB
+[2026-08-27 04:31:28 PP5 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:28 PP5 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:28 PP5 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.87profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.90profile/s]
+[2026-08-27 04:31:29 PP4 TP3 EP3] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.87profile/s]
+[2026-08-27 04:31:29 PP4 TP2 EP2] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.93profile/s]
+[2026-08-27 04:31:29 PP4 TP0 EP0] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.43profile/s]
+[2026-08-27 04:31:29 PP4 TP1 EP1] FlashInfer autotune completed.
+[2026-08-27 04:31:29 PP4 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:29 PP4 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:29 PP4 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:29 PP4 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:29 PP4 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:29 PP4 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:29 PP4 TP0 EP0] max_total_num_tokens=3801408, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=54, context_len=1048576, available_gpu_mem=16.30 GB
+[2026-08-27 04:31:29 PP4 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:29 PP4 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+W0827 04:31:30.513926  2527 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50a93b80, len: 6912: Operation not permitted [1]
+W0827 04:31:30.514082  2528 memory_location.cpp:73] Failed to get NUMA node, addr: 0x507df700, len: 6912: Operation not permitted [1]
+W0827 04:31:30.514106  2529 memory_location.cpp:73] Failed to get NUMA node, addr: 0x508aa940, len: 6912: Operation not permitted [1]
+W0827 04:31:30.514214  2530 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50f67bc0, len: 6912: Operation not permitted [1]
+W0827 04:31:30.514272  2531 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50a2f680, len: 55296: Operation not permitted [1]
+W0827 04:31:30.514393  2532 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50afecc0, len: 55296: Operation not permitted [1]
+W0827 04:31:30.514506  2533 memory_location.cpp:73] Failed to get NUMA node, addr: 0x502be580, len: 6912: Operation not permitted [1]
+W0827 04:31:30.514607  2534 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5054d3c0, len: 13824: Operation not permitted [1]
+W0827 04:31:30.514652  2535 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50ad4580, len: 6912: Operation not permitted [1]
+W0827 04:31:30.514743  2536 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50eb9880, len: 6912: Operation not permitted [1]
+W0827 04:31:31.680995  2556 memory_location.cpp:73] Failed to get NUMA node, addr: 0x501c3300, len: 6912: Operation not permitted [1]
+W0827 04:31:31.680985  2557 memory_location.cpp:73] Failed to get NUMA node, addr: 0x504ecec0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.680985  2555 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd4bac0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681169  2558 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50706900, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681272  2559 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5006d700, len: 55296: Operation not permitted [1]
+W0827 04:31:31.681360  2560 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50041cc0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.681468  2561 memory_location.cpp:73] Failed to get NUMA node, addr: 0x506f5a40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681562  2562 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fdfdf40, len: 13824: Operation not permitted [1]
+W0827 04:31:31.681607  2563 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd16080, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681685  2564 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fcfa300, len: 6912: Operation not permitted [1]
+W0827 04:31:31.733497  2583 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51554040, len: 6912: Operation not permitted [1]
+W0827 04:31:31.733552  2584 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5110a8c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.733618  2585 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51ca8e80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.733795  2586 memory_location.cpp:73] Failed to get NUMA node, addr: 0x52017680, len: 6912: Operation not permitted [1]
+W0827 04:31:31.733909  2587 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5174af80, len: 55296: Operation not permitted [1]
+W0827 04:31:31.734004  2588 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51770440, len: 55296: Operation not permitted [1]
+W0827 04:31:31.734061  2589 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51118000, len: 6912: Operation not permitted [1]
+W0827 04:31:31.734179  2590 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51564f40, len: 13824: Operation not permitted [1]
+W0827 04:31:31.734277  2591 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51f69480, len: 6912: Operation not permitted [1]
+W0827 04:31:31.734400  2592 memory_location.cpp:73] Failed to get NUMA node, addr: 0x517c2940, len: 6912: Operation not permitted [1]
+W0827 04:31:31.741586  2611 memory_location.cpp:73] Failed to get NUMA node, addr: 0x519520c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.741609  2612 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50ff8c40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.741679  2613 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51a2ad80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.741792  2614 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51b5b300, len: 6912: Operation not permitted [1]
+W0827 04:31:31.741938  2615 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50fe5b80, len: 55296: Operation not permitted [1]
+W0827 04:31:31.742043  2616 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51625140, len: 55296: Operation not permitted [1]
+W0827 04:31:31.742168  2617 memory_location.cpp:73] Failed to get NUMA node, addr: 0x516bf640, len: 6912: Operation not permitted [1]
+W0827 04:31:31.742278  2618 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51b73e80, len: 13824: Operation not permitted [1]
+W0827 04:31:31.742414  2619 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51ef60c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.742429  2620 memory_location.cpp:73] Failed to get NUMA node, addr: 0x513e95c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.900663  2641 memory_location.cpp:73] Failed to get NUMA node, addr: 0x517e8c80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.900679  2640 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5170b840, len: 6912: Operation not permitted [1]
+W0827 04:31:31.900738  2642 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5131d0c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.900669  2639 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51372400, len: 6912: Operation not permitted [1]
+W0827 04:31:31.900923  2643 memory_location.cpp:73] Failed to get NUMA node, addr: 0x512b7e00, len: 55296: Operation not permitted [1]
+W0827 04:31:31.900976  2644 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51380200, len: 55296: Operation not permitted [1]
+W0827 04:31:31.901000  2646 memory_location.cpp:73] Failed to get NUMA node, addr: 0x513151c0, len: 13824: Operation not permitted [1]
+W0827 04:31:31.901011  2645 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51194f40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.901038  2647 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51520740, len: 6912: Operation not permitted [1]
+W0827 04:31:31.901073  2648 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51724980, len: 6912: Operation not permitted [1]
+W0827 04:31:31.931138  2668 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51061000, len: 6912: Operation not permitted [1]
+W0827 04:31:31.931178  2669 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50d74c00, len: 6912: Operation not permitted [1]
+W0827 04:31:31.931098  2667 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50b2fb40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.931272  2670 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50de9000, len: 6912: Operation not permitted [1]
+W0827 04:31:31.931349  2671 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5097dc80, len: 55296: Operation not permitted [1]
+W0827 04:31:31.931519  2672 memory_location.cpp:73] Failed to get NUMA node, addr: 0x515aa000, len: 55296: Operation not permitted [1]
+W0827 04:31:31.931557  2675 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51269700, len: 6912: Operation not permitted [1]
+W0827 04:31:31.931556  2673 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5097bd00, len: 6912: Operation not permitted [1]
+W0827 04:31:31.931576  2674 memory_location.cpp:73] Failed to get NUMA node, addr: 0x514ee800, len: 13824: Operation not permitted [1]
+W0827 04:31:31.931759  2676 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50deac80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.983201  2696 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50819380, len: 6912: Operation not permitted [1]
+W0827 04:31:31.983201  2695 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50600a80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.983378  2697 memory_location.cpp:73] Failed to get NUMA node, addr: 0x506ddf40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.983438  2698 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50716800, len: 6912: Operation not permitted [1]
+W0827 04:31:31.983490  2699 memory_location.cpp:73] Failed to get NUMA node, addr: 0x501575c0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.983624  2700 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50863340, len: 55296: Operation not permitted [1]
+W0827 04:31:31.983737  2701 memory_location.cpp:73] Failed to get NUMA node, addr: 0x508d7a40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.983930  2702 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50668c40, len: 13824: Operation not permitted [1]
+W0827 04:31:31.984035  2703 memory_location.cpp:73] Failed to get NUMA node, addr: 0x506841c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.984107  2704 memory_location.cpp:73] Failed to get NUMA node, addr: 0x508f16c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.001853  2725 memory_location.cpp:73] Failed to get NUMA node, addr: 0x507c8d40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.001852  2724 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50b39240, len: 6912: Operation not permitted [1]
+W0827 04:31:32.002048  2727 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50a8ecc0, len: 55296: Operation not permitted [1]
+W0827 04:31:32.001888  2726 memory_location.cpp:73] Failed to get NUMA node, addr: 0x512a8140, len: 6912: Operation not permitted [1]
+W0827 04:31:32.002044  2728 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50a68880, len: 55296: Operation not permitted [1]
+W0827 04:31:32.002257  2729 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5087b880, len: 6912: Operation not permitted [1]
+W0827 04:31:32.002264  2731 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50fc6980, len: 6912: Operation not permitted [1]
+W0827 04:31:32.002266  2730 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51331100, len: 13824: Operation not permitted [1]
+W0827 04:31:32.002470  2732 memory_location.cpp:73] Failed to get NUMA node, addr: 0x513384c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.002780  2723 memory_location.cpp:73] Failed to get NUMA node, addr: 0x512c6940, len: 6912: Operation not permitted [1]
+[2026-08-27 04:31:32] Dummy health check server started in background thread at 0.0.0.0:30000
+[rank16]:[W827 04:33:25.972222579 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank17]:[W827 04:33:25.980800100 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank19]:[W827 04:33:25.981413562 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank18]:[W827 04:33:26.644294178 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank16]:[W827 04:34:10.467297472 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank18]:[W827 04:34:10.467837454 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank17]:[W827 04:34:10.468151853 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank19]:[W827 04:34:10.468427857 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank22]:[W827 04:34:10.495396864 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank23]:[W827 04:34:10.495456053 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank21]:[W827 04:34:10.495795969 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank20]:[W827 04:34:10.495814123 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank22]:[W827 04:34:16.703081720 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank20]:[W827 04:34:16.703915664 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank21]:[W827 04:34:16.704033167 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank23]:[W827 04:34:16.704314444 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[2026-08-27 04:35:10 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.29
+[2026-08-27 04:35:10 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.29
+[2026-08-27 04:35:29 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.39
+[2026-08-27 04:35:29 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.39
+[2026-08-27 04:35:31 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25.89
+[2026-08-27 04:35:31 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25.91
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/p_healthy_p_node3.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/p_healthy_p_node3.log
new file mode 100644
index 0000000..4835a32
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/p_healthy_p_node3.log
@@ -0,0 +1,757 @@
+/usr/local/lib/python3.12/dist-packages/torchao/quantization/quant_api.py:1731: SyntaxWarning: invalid escape sequence '\.'
+  """Configuration class for applying different quantization configs to modules or parameters based on their fully qualified names (FQNs).
+/sgl-workspace/sglang/python/sglang/launch_server.py:56: UserWarning: 'python -m sglang.launch_server' is still supported, but 'sglang serve' is the recommended entrypoint.
+  Example: sglang serve --model-path <model> [options]
+  warnings.warn(
+[2026-08-27 04:28:46] Breakable CUDA graph is incompatible with MLA attention; disabling prefill CUDA graph.
+[2026-08-27 04:28:47] Attention backend not specified. Use triton backend by default.
+[2026-08-27 04:28:47] Pipeline parallelism is incompatible with overlap schedule.
+[2026-08-27 04:28:49] server_args=ServerArgs(model_path='/data/hf_models/Kimi-K3', tokenizer_path='/data/hf_models/Kimi-K3', tokenizer_mode='auto', tokenizer_backend='huggingface', tokenizer_worker_num=1, detokenizer_worker_num=1, skip_tokenizer_init=False, load_format='auto', model_loader_extra_config='{}', trust_remote_code=True, context_length=None, is_embedding=False, enable_multimodal=None, revision=None, model_impl='auto', model_config_parser='auto', json_model_override_args='{}', dtype='auto', quantization=None, quantization_param_path=None, kv_cache_dtype='auto', enable_fp32_lm_head=False, modelopt_quant=None, modelopt_checkpoint_restore_path=None, modelopt_checkpoint_save_path=None, modelopt_export_path=None, quantize_and_serve=False, rl_quant_profile=None, enable_tf32_matmul=False, mem_fraction_static=0.88, max_running_requests=None, max_queued_requests=None, max_total_tokens=None, chunked_prefill_size=8192, enable_dynamic_chunking=False, max_prefill_tokens=16384, prefill_max_requests=None, schedule_policy='fcfs', enable_priority_scheduling=False, disable_priority_preemption=False, default_priority_value=None, abort_on_priority_when_disabled=False, schedule_low_priority_values_first=False, priority_scheduling_preemption_threshold=10, retraction_policy='length', schedule_conservativeness=1.0, page_size=64, swa_full_tokens_ratio=0.8, disable_hybrid_swa_memory=False, radix_eviction_policy='lru', prefill_only_disable_kv_cache=False, disable_radix_cache=True, enable_page_major_kv_layout=False, enable_unified_memory=False, disable_chunked_prefix_cache=False, disable_overlap_schedule=True, num_continuous_decode_steps=1, scheduler_recv_interval=1, enable_mixed_chunk=False, nccl_port=None, dist_timeout=3600, dist_init_addr='174.1.60.1:20000', nnodes=4, node_rank=3, tp_size=4, dcp_size=1, pp_size=8, pp_max_micro_batch_size=None, pp_async_batch_depth=0, dp_size=1, load_balance_method='follow_bootstrap_room', attn_cp_size=1, moe_dp_size=1, dcp_comm_backend='ag_rs', dcp_replicate_q_proj=None, dwdp_size=1, enable_prefill_cp=False, cp_strategy=None, enable_dsa_cache_layer_split=False, enable_dsa_prefill_context_parallel=False, dsa_prefill_cp_mode='round-robin-split', enable_prefill_context_parallel=False, prefill_cp_mode='in-seq-split', enable_cp_decode_attn_tp=False, enable_dp_attention=False, enable_dp_attention_local_control_broadcast=False, enable_dp_lm_head=False, enable_attn_tp_input_scattered=False, disable_attn_tp_gather=False, enable_p2p_check=False, device='cuda', base_gpu_id=0, gpu_id_step=1, random_seed=927276655, watchdog_timeout=300, soft_watchdog_timeout=None, sleep_on_idle=False, use_ray=False, custom_sigquit_handler=None, numa_node=None, gc_threshold=None, host='0.0.0.0', port=30000, fastapi_root_path='', smg_grpc_mode=False, grpc_mode=False, grpc_port=None, sidecar=None, sidecar_args=None, skip_server_warmup=False, warmups=None, enable_http2=False, ssl_keyfile=None, ssl_certfile=None, ssl_ca_certs=None, ssl_keyfile_password=None, enable_ssl_refresh=False, api_key=None, admin_api_key=None, served_model_name='kimi-k3', weight_version='default', chat_template=None, hf_chat_template_name=None, completion_template=None, file_storage_path='sglang_storage', enable_cache_report=False, reasoning_parser=None, default_chat_template_kwargs=None, strip_thinking_cache=False, enable_strict_thinking=False, tool_call_parser=None, tool_server=None, sampling_defaults='model', asr_max_buffer_seconds=60, asr_max_concurrent_sessions=32, preferred_sampling_params=None, allow_auto_truncate=False, stream_interval=1, batch_notify_size=16, stream_response_default_include_usage=False, incremental_streaming_output=False, enable_streaming_session=False, enable_session_radix_cache=False, log_level='info', log_level_http=None, log_requests=False, log_requests_level=2, log_requests_format='text', log_requests_target=None, uvicorn_access_log_exclude_prefixes=[], crash_dump_folder=None, show_time_cost=False, enable_metrics=False, smg_http_sidecar_port=None, enable_mfu_metrics=False, enable_metrics_for_all_schedulers=False, load_snapshot_publish_interval=15, tokenizer_metrics_custom_labels_header='x-custom-labels', tokenizer_metrics_allowed_custom_labels=None, extra_metric_labels=None, bucket_time_to_first_token=None, bucket_inter_token_latency=None, bucket_e2e_request_latency=None, prompt_tokens_buckets=None, generation_tokens_buckets=None, gc_warning_threshold_secs=0.0, decode_log_interval=40, enable_request_time_stats_logging=False, kv_events_config=None, enable_forward_pass_metrics=False, forward_pass_metrics_worker_id='', forward_pass_metrics_ipc_name=None, enable_trace=False, trace_modules='request', otlp_traces_endpoint='localhost:4317', export_metrics_to_file=False, export_metrics_to_file_dir=None, stat_loggers=None, constrained_json_whitespace_pattern=None, constrained_json_disable_any_whitespace=False, attention_backend='triton', decode_attention_backend=None, prefill_attention_backend=None, sampling_backend='flashinfer', grammar_backend='xgrammar', radix_cache_backend=None, mm_attention_backend=None, fp8_gemm_runner_backend='auto', fp4_gemm_runner_backend='auto', bf16_gemm_backend='auto', dsa_prefill_backend=None, dsa_decode_backend=None, dsa_paged_mqa_logits_backend='auto', dsa_topk_backend='sgl-kernel', disable_flashinfer_autotune=False, flashinfer_autotune_skip_ops=None, mamba_backend='triton', cuda_graph_config=CudaGraphConfig(decode=PhaseConfig(backend='disabled', max_bs=16, bs=[1, 2, 4, 8, 12, 16], tc_compiler='eager', full_prefill_max_req=None), prefill=PhaseConfig(backend='disabled', max_bs=2048, bs=[4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], tc_compiler='eager', full_prefill_max_req=None)), cuda_graph_backend_decode=None, cuda_graph_backend_prefill=None, cuda_graph_max_bs_decode=16, cuda_graph_max_bs_prefill=None, cuda_graph_bs_decode=None, cuda_graph_bs_prefill=None, cuda_graph_tc_compiler=None, disable_prefill_cuda_graph=False, disable_decode_cuda_graph=False, disable_cuda_graph=False, disable_cuda_graph_padding=False, enable_profile_cuda_graph=False, enable_cudagraph_gc=False, debug_cuda_graph=False, enable_layerwise_nvtx_marker=False, enable_nccl_nvls=False, enable_symm_mem=False, triton_attention_reduce_in_fp32=False, triton_attention_num_kv_splits=8, triton_attention_split_tile_size=None, flashinfer_mla_disable_ragged=False, enable_fused_qk_norm_rope=False, enable_precise_embedding_interpolation=False, enable_fused_moe_sum_all_reduce=False, enable_deepseek_v4_fp4_indexer=False, disable_custom_all_reduce=False, enable_mscclpp=False, enable_torch_symm_mem=False, enable_scattered_sconv=False, pre_warm_nccl=False, enable_quant_communications=False, enable_flashinfer_allreduce_fusion=False, enforce_disable_flashinfer_allreduce_fusion=False, flashinfer_allreduce_fusion_backend=None, enable_aiter_allreduce_fusion=False, enable_torch_compile=False, enable_torch_compile_debug_mode=False, torch_compile_max_bs=32, torchao_config='', speculative_algorithm=None, speculative_draft_model_path=None, speculative_draft_model_revision=None, speculative_draft_load_format=None, speculative_num_steps=None, speculative_eagle_topk=None, speculative_num_draft_tokens=None, speculative_dflash_block_size=None, speculative_dspark_block_size=None, speculative_dspark_sps_table_path=None, speculative_dspark_confidence_sts_path=None, speculative_dspark_align_verify_tokens_to_graph_tier=False, speculative_accept_threshold_single=1.0, speculative_accept_threshold_acc=1.0, speculative_use_rejection_sampling=False, speculative_token_map=None, speculative_attention_mode='prefill', speculative_draft_attention_backend=None, speculative_draft_window_size=None, speculative_moe_runner_backend='flashinfer_mxfp4', speculative_moe_a2a_backend=None, speculative_draft_model_quantization=None, speculative_skip_dp_mlp_sync=False, enable_multi_layer_eagle=False, speculative_adaptive=False, speculative_adaptive_config=None, decoupled_spec_bind_endpoint=None, decoupled_spec_connect_endpoints=None, decoupled_spec_rank=None, decoupled_spec_role='null', spec_trace_dir=None, speculative_ngram_min_bfs_breadth=1, speculative_ngram_max_bfs_breadth=10, speculative_ngram_match_type='BFS', speculative_ngram_max_trie_depth=18, speculative_ngram_capacity=10000000, speculative_ngram_external_corpus_path=None, speculative_ngram_external_sam_budget=0, speculative_ngram_external_corpus_max_tokens=10000000, ep_size=4, moe_a2a_backend='none', moe_runner_backend='flashinfer_mxfp4', flashinfer_mxfp4_moe_precision='default', deepep_mode='auto', fuseep_mode=2, deepep_dispatcher_output_dtype='auto', ep_num_redundant_experts=0, ep_dispatch_algorithm=None, init_expert_location='trivial', enable_eplb=False, eplb_algorithm='auto', eplb_rebalance_num_iterations=1000, eplb_rebalance_layers_per_chunk=None, eplb_min_rebalancing_utilization_threshold=1.0, expert_distribution_recorder_mode=None, expert_distribution_recorder_buffer_size=1000, enable_expert_distribution_metrics=False, deepep_config=None, moe_dense_tp_size=None, elastic_ep_backend=None, enable_elastic_expert_backup=False, mooncake_ib_device=None, enable_waterfill=False, ep_join_mode=None, ep_join_rank_offset=0, elastic_ep_initial_size=None, max_ep_size=None, elastic_ep_scale_timeout=600, elastic_ep_rejoin=False, disable_flashinfer_cutlass_moe_fp4_allgather=False, disable_shared_experts_fusion=False, enforce_shared_experts_fusion=False, max_mamba_cache_size=None, mamba_ssm_dtype=None, mamba_max_states_per_path=-1, enable_mamba_cache_stochastic_rounding=False, mamba_cache_philox_rounds=0, mamba_full_memory_ratio=0.36, mamba_radix_cache_strategy='extra_buffer_lazy', uses_mamba_radix_cache=False, mamba_track_interval=256, enable_int8_mamba_checkpoint=False, int8_mamba_ckpt_size=None, linear_attn_backend='triton', linear_attn_decode_backend=None, linear_attn_prefill_backend=None, linear_attn_verify_backend=None, enable_linear_replayssm=False, linear_replayssm_cache_len=16, enable_linear_replayssm_spec=False, enable_hierarchical_cache=False, hicache_ratio=2.0, hicache_size=0, hicache_write_policy='write_through', hicache_io_backend='kernel', hicache_mem_layout='page_first', hicache_storage_backend=None, hicache_storage_prefetch_policy='timeout', hicache_storage_backend_extra_config=None, enable_hisparse=False, hisparse_config=None, enable_broadcast_mm_inputs_process=False, enable_prefix_mm_cache=False, mm_enable_dp_encoder=False, mm_process_config={}, mm_processor_worker_num=0, mm_io_worker_num=0, limit_mm_data_per_request=None, enable_mm_global_cache=False, disable_fast_image_processor=False, mm_feature_transport='cpu', keep_mm_feature_on_device=False, enable_lora=None, enable_lora_overlap_loading=None, max_lora_rank=None, lora_target_modules=None, lora_paths=None, max_loaded_loras=None, max_loras_per_batch=8, lora_eviction_policy='lru', lora_backend='csgmv', max_lora_chunk_size=16, experts_shared_outer_loras=None, lora_use_virtual_experts=False, lora_strict_loading=False, lora_drain_wait_threshold=0.0, enable_two_batch_overlap=False, enable_single_batch_overlap=False, tbo_token_distribution_threshold=0.48, cpu_offload_gb=0, offload_group_size=-1, offload_num_in_group=1, offload_prefetch_step=1, offload_mode='cpu', enable_lmcache=False, lmcache_config_file=None, enable_flexkv=False, flexkv_config_file=None, kt_weight_path=None, kt_method='AMXINT4', kt_cpuinfer=None, kt_threadpool_count=2, kt_num_gpu_experts=None, kt_max_deferred_experts_per_token=None, dllm_algorithm=None, dllm_algorithm_config=None, dllm_fdfo=True, disaggregation_mode='prefill', disaggregation_transfer_backend='mooncake', disaggregation_bootstrap_port=28800, disaggregation_ib_device='mlx5_0,mlx5_1,mlx5_2,mlx5_3', disaggregation_decode_enable_radix_cache=False, disaggregation_decode_enable_offload_kvcache=False, num_reserved_decode_tokens=512, disaggregation_decode_extra_slots=None, disaggregation_decode_polling_interval=1, optimistic_prefill_attempts=0, encoder_only=False, language_only=False, encoder_transfer_backend='zmq_to_tokenizer', encoder_urls=[], encoder_bootstrap_port=8997, encoder_register_urls=[], enable_adaptive_dispatch_to_encoder=False, enable_pdmux=False, pdmux_config_path=None, sm_group_num=8, custom_weight_loader=[], weight_loader_disable_mmap=False, weight_loader_prefetch_checkpoints=False, weight_loader_prefetch_num_threads=4, weight_loader_drop_cache_after_load=False, remote_instance_weight_loader_seed_instance_ip=None, remote_instance_weight_loader_seed_instance_service_port=None, remote_instance_weight_loader_send_weights_group_ports=None, remote_instance_weight_loader_backend='nccl', remote_instance_weight_loader_start_seed_via_transfer_engine=False, engine_info_bootstrap_port=6789, modelexpress_config=None, download_dir=None, model_checksum=None, delete_ckpt_after_loading=False, decrypted_config_file=None, decrypted_draft_config_file=None, checkpoint_engine_wait_weights_before_ready=False, enable_prefill_delayer=False, prefill_delayer_max_delay_passes=30, prefill_delayer_token_usage_low_watermark=None, prefill_delayer_forward_passes_buckets=None, prefill_delayer_wait_seconds_buckets=None, prefill_delayer_queue_min_ratio=None, prefill_delayer_max_delay_ms=None, min_free_slots_delay=None, enable_deterministic_inference=False, rl_on_policy_target=None, kv_canary='none', kv_canary_real_data='none', kv_canary_sweep_interval=0, enable_dynamic_batch_tokenizer=False, dynamic_batch_tokenizer_batch_size=32, dynamic_batch_tokenizer_batch_timeout=0.002, enable_tokenizer_batch_encode=False, disable_tokenizer_batch_decode=False, debug_tensor_dump_output_folder=None, debug_tensor_dump_layers=None, debug_tensor_dump_input_file=None, enable_memory_saver=False, enable_weights_cpu_backup=False, enable_draft_weights_cpu_backup=False, enable_custom_logit_processor=False, enable_return_hidden_states=False, enable_return_routed_experts=False, enable_return_indexer_topk=False, disable_outlines_disk_cache=False, enable_mis=False, weight_cache_mode='off', weight_cache_socket=None, weight_cache_timeout=1800, forward_hooks=None, msprobe_dump_config=None)
+[2026-08-27 04:28:49] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:49] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:49] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:49] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:49] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:49] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:49] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:49] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:05 PP6 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:05 PP6 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:07 PP6 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:07 PP6 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:07 PP6 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:07 PP7 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:07 PP7 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:07 PP7 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:07 PP7 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:07 PP6 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:07 PP6 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:07.662256   238 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:07.662379   238 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:07.662392   238 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.4 port: 12001
+I0827 04:29:07.662429   238 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.4:15914
+I0827 04:29:07.662514   238 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:07.685189   238 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:07.689950   238 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:07.695539   238 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:07.701294   238 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:07.703691   238 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:07.703715   238 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:07.703760   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:07.710492   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:07.711953   238 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:04
+I0827 04:29:07.711989   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:07.735716   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:07.737195   238 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:04
+I0827 04:29:07.737223   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:07.763653   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:07.765146   238 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:04
+I0827 04:29:07.765172   238 rdma_context.cpp:205] Using SIEVE endpoint store
+[2026-08-27 04:29:07 PP7 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:07 PP7 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:07 PP7 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:07 PP7 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+I0827 04:29:07.791589   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:07.793025   238 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:04
+I0827 04:29:07.793139   238 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:07 PP6 TP3 EP3] Init torch distributed begin.
+[2026-08-27 04:29:07 PP6 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:07 PP6 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:08 PP6 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:08.916008   235 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:08.916072   235 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:08.916082   235 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.4 port: 12001
+I0827 04:29:08.916121   235 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.4:15737
+I0827 04:29:08.916215   235 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:08.941840   235 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:08.950320   235 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:08.979278   235 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:08.988204   235 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:08.991855   235 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:08.991879   235 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:08.991928   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:08.999985   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:09.001416   235 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:04
+I0827 04:29:09.001446   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:09.027477   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:09.028954   235 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:04
+I0827 04:29:09.028980   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:09.055524   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:09.056981   235 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:04
+I0827 04:29:09.057020   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:09.083544   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:09.084911   235 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:04
+I0827 04:29:09.085031   235 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:09 PP6 TP0 EP0] Init torch distributed begin.
+[2026-08-27 04:29:09 PP7 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:09 PP6 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:09 PP7 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:09 PP6 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:09 PP7 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:09 PP7 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:10.000486   242 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:10.000609   242 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:10.000628   242 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.4 port: 12001
+I0827 04:29:10.000689   242 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.4:16890
+I0827 04:29:10.000851   242 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:10.006592   236 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:10.006711   236 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:10.006728   236 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.4 port: 12001
+I0827 04:29:10.006786   236 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.4:15893
+I0827 04:29:10.006963   236 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:10.022238   241 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:10.022354   241 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:10.022370   241 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.4 port: 12001
+I0827 04:29:10.022423   241 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.4:16277
+I0827 04:29:10.022593   241 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:10.032052   242 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:10.032101   236 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:10.039772   236 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:10.039858   242 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:10.047292   236 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:10.047341   242 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:10.047785   241 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:10.054806   242 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:10.054838   236 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:10.055280   241 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:10.059140   236 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:10.059171   236 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:10.059175   242 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:10.059201   242 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:10.059227   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.059260   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.062610   241 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:10.066787   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:10.066826   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:10.068500   242 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:04
+I0827 04:29:10.068532   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.068558   236 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:04
+I0827 04:29:10.068590   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.069866   241 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:10.072566   241 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:10.072592   241 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:10.072649   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.080149   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:10.081809   241 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:04
+I0827 04:29:10.081843   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.095701   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:10.097368   236 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:04
+I0827 04:29:10.097400   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.099974   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:10.101672   242 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:04
+I0827 04:29:10.101717   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.107705   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:10.109316   241 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:04
+I0827 04:29:10.109347   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.123868   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:10.125504   236 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:04
+I0827 04:29:10.125536   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.127957   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:10.129673   242 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:04
+I0827 04:29:10.129703   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.139813   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:10.141404   241 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:04
+I0827 04:29:10.141431   241 rdma_context.cpp:205] Using SIEVE endpoint store
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:10.150050   237 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:10.150159   237 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:10.150173   237 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.4 port: 12001
+I0827 04:29:10.150219   237 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.4:15413
+I0827 04:29:10.150358   237 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:10.151880   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:10.153475   236 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:04
+I0827 04:29:10.153601   236 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:10 PP6 TP1 EP1] Init torch distributed begin.
+I0827 04:29:10.155849   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:10.155864   237 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:10.157462   242 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:04
+I0827 04:29:10.157585   242 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:10 PP7 TP3 EP3] Init torch distributed begin.
+I0827 04:29:10.171923   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:10.172516   240 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:10.172667   240 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:10.172688   240 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.4 port: 12001
+I0827 04:29:10.172762   240 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.4:15705
+I0827 04:29:10.172977   240 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:10.173573   241 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:04
+I0827 04:29:10.173692   241 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:10 PP7 TP2 EP2] Init torch distributed begin.
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:10.177717   239 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:10.177865   239 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:10.177886   239 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.4 port: 12001
+I0827 04:29:10.177956   239 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.4:16192
+I0827 04:29:10.178146   239 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:10.183638   237 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:10.187090   239 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:10.191001   237 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:10.195258   239 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:10.198271   237 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:10.200984   237 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:10.201025   237 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:10.201079   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.203014   239 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:10.203156   240 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:10.209589   240 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:10.210305   239 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:10.212793   239 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:10.212821   239 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:10.212878   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.216092   240 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:10.219946   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:10.221297   239 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:04
+I0827 04:29:10.221323   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.222187   240 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:10.224284   240 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:10.224304   240 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:10.224344   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.230640   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:10.231778   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:10.231937   240 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:04
+I0827 04:29:10.231962   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.233362   237 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:04
+I0827 04:29:10.233397   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.236478   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:10.237651   240 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:04
+I0827 04:29:10.237670   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.250197   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:10.251711   239 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:04
+I0827 04:29:10.251737   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.263743   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:10.265311   237 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:04
+I0827 04:29:10.265339   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.267139   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:10.268607   240 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:04
+I0827 04:29:10.268632   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.273075   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:10.274682   237 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:04
+I0827 04:29:10.274710   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.276602   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:10.278275   239 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:04
+I0827 04:29:10.278314   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.295709   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:10.297385   240 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:04
+I0827 04:29:10.297510   240 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:10 PP7 TP1 EP1] Init torch distributed begin.
+I0827 04:29:10.299893   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:10.301563   237 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:04
+I0827 04:29:10.301689   237 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:10 PP6 TP2 EP2] Init torch distributed begin.
+I0827 04:29:10.303737   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:10.305307   239 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:04
+I0827 04:29:10.305433   239 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:10 PP7 TP0 EP0] Init torch distributed begin.
+[2026-08-27 04:29:26 PP7 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:26 PP6 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:27 PP6 TP0 EP0] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP6 TP3 EP3] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP6 TP2 EP2] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP6 TP1 EP1] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP7 TP1 EP1] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP7 TP0 EP0] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP7 TP3 EP3] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP7 TP2 EP2] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP7 TP1 EP1] Init torch distributed ends. elapsed=17.45 s, mem usage=0.46 GB
+[2026-08-27 04:29:27 PP7 TP3 EP3] Init torch distributed ends. elapsed=17.59 s, mem usage=0.46 GB
+[2026-08-27 04:29:27 PP7 TP2 EP2] Init torch distributed ends. elapsed=17.57 s, mem usage=0.46 GB
+[2026-08-27 04:29:27 PP7 TP0 EP0] Init torch distributed ends. elapsed=17.44 s, mem usage=0.46 GB
+[2026-08-27 04:29:27 PP6 TP1 EP1] Init torch distributed ends. elapsed=17.59 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP6 TP3 EP3] Init torch distributed ends. elapsed=19.95 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP6 TP0 EP0] Init torch distributed ends. elapsed=18.66 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP6 TP2 EP2] Init torch distributed ends. elapsed=17.44 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP6 TP1 EP1] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP7 TP1 EP1] Load weight begin. avail mem=82.12 GB
+[2026-08-27 04:29:27 PP6 TP0 EP0] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP6 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP7 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP7 TP3 EP3] Load weight begin. avail mem=82.12 GB
+[2026-08-27 04:29:27 PP6 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP7 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP6 TP2 EP2] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP6 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP7 TP2 EP2] Load weight begin. avail mem=82.12 GB
+[2026-08-27 04:29:28 PP7 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP6 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP7 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP6 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP7 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP7 TP0 EP0] Load weight begin. avail mem=82.12 GB
+[2026-08-27 04:29:28 PP6 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP7 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP7 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP6 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP7 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP6 TP3 EP3] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP6 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP6 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP6 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP7 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP7 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP7 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP6 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP7 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP6 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:30:40 PP7 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:40 PP7 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:40 PP7 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:40 PP7 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:40 PP7 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:40 PP7 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:40 PP7 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:40 PP7 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:41 PP7 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:41 PP7 TP1 EP1] Load weight end. elapsed=73.17 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=31.91 GB, mem usage=50.21 GB.
+[2026-08-27 04:30:41 PP7 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:41 PP7 TP0 EP0] Load weight end. elapsed=73.80 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=31.91 GB, mem usage=50.21 GB.
+[2026-08-27 04:30:42 PP6 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:42 PP6 TP1 EP1] Load weight end. elapsed=74.47 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:45 PP6 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:45 PP6 TP2 EP2] Load weight end. elapsed=77.20 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:50 PP7 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:50 PP7 TP3 EP3] Load weight end. elapsed=82.79 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=31.91 GB, mem usage=50.21 GB.
+[2026-08-27 04:30:51 PP6 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:51 PP6 TP3 EP3] Load weight end. elapsed=83.19 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:51 PP7 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:51 PP7 TP2 EP2] Load weight end. elapsed=83.68 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=31.91 GB, mem usage=50.21 GB.
+[2026-08-27 04:30:51 PP7 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:51 PP7 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:52 PP7 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:52 PP7 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:52 PP7 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:52 PP7 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:52 PP7 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:52 PP7 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:53 PP6 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:53 PP6 TP0 EP0] Load weight end. elapsed=85.47 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:53 PP6 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:53 PP6 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:53 PP6 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:53 PP6 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:53 PP6 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:53 PP6 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:53 PP6 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:53 PP6 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:54 PP7 TP3 EP3] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP7 TP2 EP2] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP7 TP1 EP1] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP7 TP0 EP0] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP6 TP3 EP3] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP6 TP2 EP2] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP6 TP1 EP1] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP6 TP0 EP0] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP6 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP6 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP7 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:54 PP6 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP7 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:54 PP7 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:54 PP7 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:54 PP6 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP6 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP6 TP3 EP3] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP6 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP6 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP6 TP2 EP2] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP6 TP1 EP1] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP7 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 16.31 GB
+[2026-08-27 04:30:54 PP7 TP2 EP2] Memory pool end. avail mem=14.95 GB
+[2026-08-27 04:30:54 PP6 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP6 TP0 EP0] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP7 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 16.31 GB
+[2026-08-27 04:30:54 PP7 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 16.31 GB
+[2026-08-27 04:30:54 PP7 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 16.31 GB
+[2026-08-27 04:30:54 PP7 TP3 EP3] Memory pool end. avail mem=14.95 GB
+[2026-08-27 04:30:54 PP7 TP0 EP0] Memory pool end. avail mem=14.95 GB
+[2026-08-27 04:30:54 PP7 TP1 EP1] Memory pool end. avail mem=14.95 GB
+[2026-08-27 04:30:54 PP7 TP2 EP2] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp2_pp7_dp0.json
+[2026-08-27 04:30:54 PP6 TP3 EP3] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp3_pp6_dp0.json
+[2026-08-27 04:30:54 PP6 TP1 EP1] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp1_pp6_dp0.json
+[2026-08-27 04:30:54 PP7 TP3 EP3] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp3_pp7_dp0.json
+[2026-08-27 04:30:54 PP6 TP2 EP2] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp2_pp6_dp0.json
+[2026-08-27 04:30:54 PP6 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:30:54 PP6 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:30:54 PP6 TP0 EP0] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp0_pp6_dp0.json
+[2026-08-27 04:30:54 PP7 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:30:54 PP7 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:30:54 PP7 TP0 EP0] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp0_pp7_dp0.json
+[2026-08-27 04:30:54 PP7 TP1 EP1] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp1_pp7_dp0.json
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 11.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 13.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.76profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.62profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.66profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.72profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.68profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.55profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.12profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  6.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  7.16profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.92profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.15profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.59profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.56profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 19.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  8.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.67profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.40profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.88profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  8.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  6.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.83profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.50profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.32profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.92profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.77profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.94profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.57profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 12.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 12.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 13.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.36profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.66profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  6.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.72profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  6.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.22profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.65profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.13profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.03profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.46profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.59profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.16profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  6.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.96profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.76profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.67profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  8.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.27profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.88profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.01profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.57profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.12profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 11.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  2.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.68profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.90profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.43profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.92profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  8.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.66profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  6.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  7.13profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.32profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.71profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:05,  1.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.16profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.31profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.77profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.28profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.48profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.87profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.40profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 13.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.35profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.98profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.88profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.73profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  3.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.88profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.09profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.20profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.97profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.83profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.47profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.40profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.60profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  8.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  7.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.97profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.05profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.05profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  6.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.49profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.72profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.19profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 10.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.91profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.64profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.58profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  3.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.35profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  9.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 12.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.09profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:02,  2.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 12.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.15profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.76profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.49profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.03profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.79profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.96profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.47profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.65profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.55profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.03profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  9.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.63profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.98profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:05,  1.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:01,  3.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.50profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  6.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.94profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.13profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.78profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.53profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.35profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.56profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.36profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:01<00:06,  1.06s/profile]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  8.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.91profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.66profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  8.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  7.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:02,  2.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.83profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.01profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.66profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.13profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.59profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  8.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  8.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.71profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.29profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.97profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.88profile/s]
+[2026-08-27 04:31:27 PP6 TP3 EP3] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.86profile/s]
+[2026-08-27 04:31:27 PP6 TP1 EP1] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.24profile/s]
+[2026-08-27 04:31:27 PP6 TP2 EP2] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.33profile/s]
+[2026-08-27 04:31:27 PP6 TP0 EP0] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.56profile/s][2026-08-27 04:31:27 PP6 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:27 PP6 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:27 PP6 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:27 PP6 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:27 PP6 TP0 EP0] max_total_num_tokens=3801408, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=54, context_len=1048576, available_gpu_mem=16.52 GB
+[2026-08-27 04:31:27 PP6 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:27 PP6 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:27 PP6 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:27 PP6 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
+[2026-08-27 04:31:28 PP7 TP2 EP2] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.32profile/s]
+[2026-08-27 04:31:28 PP7 TP3 EP3] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.60profile/s]
+[2026-08-27 04:31:28 PP7 TP0 EP0] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.60profile/s]
+[2026-08-27 04:31:28 PP7 TP1 EP1] FlashInfer autotune completed.
+[2026-08-27 04:31:28 PP7 TP2 EP2] multimem all-gather disabled (no multicast for world_size=4)
+[2026-08-27 04:31:28 PP7 TP3 EP3] multimem all-gather disabled (no multicast for world_size=4)
+[2026-08-27 04:31:28 PP7 TP1 EP1] multimem all-gather disabled (no multicast for world_size=4)
+[2026-08-27 04:31:28 PP7 TP0 EP0] multimem all-gather disabled (no multicast for world_size=4)
+[2026-08-27 04:31:28 PP7 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:28 PP7 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:28 PP7 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:28 PP7 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:28 PP7 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:28 PP7 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:28 PP7 TP0 EP0] max_total_num_tokens=3801408, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=54, context_len=1048576, available_gpu_mem=12.14 GB
+[2026-08-27 04:31:28 PP7 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:28 PP7 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+W0827 04:31:31.959518  2596 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5efc8ec0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.959637  2598 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ee654c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.959640  2597 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5eda3c80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.959542  2595 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5efc6f80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.959741  2599 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5eeaf500, len: 55296: Operation not permitted [1]
+W0827 04:31:31.959794  2600 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ef4e380, len: 55296: Operation not permitted [1]
+W0827 04:31:31.959827  2601 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5f548500, len: 6912: Operation not permitted [1]
+W0827 04:31:31.959918  2602 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5eec3bc0, len: 13824: Operation not permitted [1]
+W0827 04:31:31.960085  2603 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5f01bcc0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.960222  2604 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5f4c8ac0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.980162  2623 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60ba35c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.980152  2624 memory_location.cpp:73] Failed to get NUMA node, addr: 0x609ccf40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.980242  2625 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60d38600, len: 6912: Operation not permitted [1]
+W0827 04:31:31.980361  2626 memory_location.cpp:73] Failed to get NUMA node, addr: 0x609546c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.980516  2628 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60c49a00, len: 55296: Operation not permitted [1]
+W0827 04:31:31.980537  2627 memory_location.cpp:73] Failed to get NUMA node, addr: 0x607c6800, len: 55296: Operation not permitted [1]
+W0827 04:31:31.980547  2629 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60d11800, len: 6912: Operation not permitted [1]
+W0827 04:31:31.980834  2630 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60a1b740, len: 13824: Operation not permitted [1]
+W0827 04:31:31.981053  2632 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60ce5000, len: 6912: Operation not permitted [1]
+W0827 04:31:31.981062  2631 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60aef300, len: 6912: Operation not permitted [1]
+W0827 04:31:32.117022  2651 memory_location.cpp:73] Failed to get NUMA node, addr: 0x74c79d40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.117132  2652 memory_location.cpp:73] Failed to get NUMA node, addr: 0x74f99700, len: 6912: Operation not permitted [1]
+W0827 04:31:32.117317  2653 memory_location.cpp:73] Failed to get NUMA node, addr: 0x7546c740, len: 6912: Operation not permitted [1]
+W0827 04:31:32.117331  2654 memory_location.cpp:73] Failed to get NUMA node, addr: 0x74f4f140, len: 6912: Operation not permitted [1]
+W0827 04:31:32.117414  2656 memory_location.cpp:73] Failed to get NUMA node, addr: 0x750037c0, len: 55296: Operation not permitted [1]
+W0827 04:31:32.117445  2655 memory_location.cpp:73] Failed to get NUMA node, addr: 0x74bf1200, len: 55296: Operation not permitted [1]
+W0827 04:31:32.117705  2657 memory_location.cpp:73] Failed to get NUMA node, addr: 0x74f80080, len: 6912: Operation not permitted [1]
+W0827 04:31:32.117738  2658 memory_location.cpp:73] Failed to get NUMA node, addr: 0x74fbd2c0, len: 13824: Operation not permitted [1]
+W0827 04:31:32.117790  2659 memory_location.cpp:73] Failed to get NUMA node, addr: 0x74fa5e40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.118012  2660 memory_location.cpp:73] Failed to get NUMA node, addr: 0x74c24880, len: 6912: Operation not permitted [1]
+W0827 04:31:32.137427  2677 memory_location.cpp:73] Failed to get NUMA node, addr: 0x685bea40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.137485  2678 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6865c240, len: 6912: Operation not permitted [1]
+W0827 04:31:32.137612  2680 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6832f280, len: 6912: Operation not permitted [1]
+W0827 04:31:32.137683  2679 memory_location.cpp:73] Failed to get NUMA node, addr: 0x68511ec0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.137791  2681 memory_location.cpp:73] Failed to get NUMA node, addr: 0x687ae280, len: 55296: Operation not permitted [1]
+W0827 04:31:32.137800  2682 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6863aec0, len: 55296: Operation not permitted [1]
+W0827 04:31:32.137928  2683 memory_location.cpp:73] Failed to get NUMA node, addr: 0x685f4540, len: 6912: Operation not permitted [1]
+W0827 04:31:32.138145  2685 memory_location.cpp:73] Failed to get NUMA node, addr: 0x68b70f80, len: 6912: Operation not permitted [1]
+W0827 04:31:32.138144  2684 memory_location.cpp:73] Failed to get NUMA node, addr: 0x686e87c0, len: 13824: Operation not permitted [1]
+W0827 04:31:32.138219  2686 memory_location.cpp:73] Failed to get NUMA node, addr: 0x685491c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.187127  2706 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8eaeff40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.187187  2707 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8eaa1740, len: 6912: Operation not permitted [1]
+W0827 04:31:32.187136  2705 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8e5f2600, len: 6912: Operation not permitted [1]
+W0827 04:31:32.187413  2709 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ec4eb00, len: 55296: Operation not permitted [1]
+W0827 04:31:32.187405  2708 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ea3b800, len: 6912: Operation not permitted [1]
+W0827 04:31:32.187651  2710 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ed92ec0, len: 55296: Operation not permitted [1]
+W0827 04:31:32.187799  2711 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8eb3fbc0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.188133  2712 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ed0e1c0, len: 13824: Operation not permitted [1]
+W0827 04:31:32.188231  2713 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8e9d20c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.188272  2714 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8f2c7a80, len: 6912: Operation not permitted [1]
+W0827 04:31:32.252096  2732 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72b6b180, len: 6912: Operation not permitted [1]
+W0827 04:31:32.252050  2731 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72be0600, len: 6912: Operation not permitted [1]
+W0827 04:31:32.252147  2733 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72716c00, len: 6912: Operation not permitted [1]
+W0827 04:31:32.252276  2734 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72739940, len: 6912: Operation not permitted [1]
+W0827 04:31:32.252379  2735 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72da8a80, len: 55296: Operation not permitted [1]
+W0827 04:31:32.252439  2736 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72d15800, len: 55296: Operation not permitted [1]
+W0827 04:31:32.252678  2737 memory_location.cpp:73] Failed to get NUMA node, addr: 0x7315db40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.252701  2738 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72e0e700, len: 13824: Operation not permitted [1]
+W0827 04:31:32.252745  2739 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72f5f480, len: 6912: Operation not permitted [1]
+W0827 04:31:32.252832  2740 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72b295c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.268508  2760 memory_location.cpp:73] Failed to get NUMA node, addr: 0x69eefc00, len: 6912: Operation not permitted [1]
+W0827 04:31:32.268532  2759 memory_location.cpp:73] Failed to get NUMA node, addr: 0x69ec8680, len: 6912: Operation not permitted [1]
+W0827 04:31:32.268577  2761 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6a02e280, len: 6912: Operation not permitted [1]
+W0827 04:31:32.268708  2762 memory_location.cpp:73] Failed to get NUMA node, addr: 0x69d75dc0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.268766  2763 memory_location.cpp:73] Failed to get NUMA node, addr: 0x69fc4880, len: 55296: Operation not permitted [1]
+W0827 04:31:32.268886  2764 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6a0c8540, len: 55296: Operation not permitted [1]
+W0827 04:31:32.268975  2765 memory_location.cpp:73] Failed to get NUMA node, addr: 0x69c75940, len: 6912: Operation not permitted [1]
+W0827 04:31:32.269125  2766 memory_location.cpp:73] Failed to get NUMA node, addr: 0x698ae1c0, len: 13824: Operation not permitted [1]
+W0827 04:31:32.269232  2767 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6a5457c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.269354  2768 memory_location.cpp:73] Failed to get NUMA node, addr: 0x69c8fe80, len: 6912: Operation not permitted [1]
+W0827 04:31:32.281724  2785 memory_location.cpp:73] Failed to get NUMA node, addr: 0x678e47c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.281751  2786 memory_location.cpp:73] Failed to get NUMA node, addr: 0x67a39b80, len: 6912: Operation not permitted [1]
+W0827 04:31:32.281852  2787 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6798d9c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.281994  2788 memory_location.cpp:73] Failed to get NUMA node, addr: 0x67981dc0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.282085  2789 memory_location.cpp:73] Failed to get NUMA node, addr: 0x67823300, len: 55296: Operation not permitted [1]
+W0827 04:31:32.282146  2790 memory_location.cpp:73] Failed to get NUMA node, addr: 0x67a3bb00, len: 55296: Operation not permitted [1]
+W0827 04:31:32.282214  2791 memory_location.cpp:73] Failed to get NUMA node, addr: 0x67cdad40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.282285  2792 memory_location.cpp:73] Failed to get NUMA node, addr: 0x68028e40, len: 13824: Operation not permitted [1]
+W0827 04:31:32.282544  2794 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6796cf40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.282545  2793 memory_location.cpp:73] Failed to get NUMA node, addr: 0x67fa6600, len: 6912: Operation not permitted [1]
+[2026-08-27 04:31:32] Dummy health check server started in background thread at 0.0.0.0:30000
+[rank27]:[W827 04:34:16.127713860 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank25]:[W827 04:34:16.128748517 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank24]:[W827 04:34:16.129831915 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank26]:[W827 04:34:16.136239904 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank26]:[W827 04:35:04.964905528 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank25]:[W827 04:35:04.966225881 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank27]:[W827 04:35:04.966239242 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank24]:[W827 04:35:04.966301664 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank31]:[W827 04:35:04.990701389 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank28]:[W827 04:35:04.991914270 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank29]:[W827 04:35:04.994546241 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank30]:[W827 04:35:04.999020079 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank31]:[W827 04:35:10.873647776 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank30]:[W827 04:35:10.873901815 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank28]:[W827 04:35:10.874227564 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank29]:[W827 04:35:10.885318244 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[2026-08-27 04:35:10 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.29
+[2026-08-27 04:35:10 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.29
+[2026-08-27 04:35:29 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.39
+[2026-08-27 04:35:29 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.39
+[2026-08-27 04:35:31 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25.93
+[2026-08-27 04:35:31 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25.95
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/p_healthy_router.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/p_healthy_router.log
new file mode 100644
index 0000000..1190be5
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/p_healthy_router.log
@@ -0,0 +1 @@
+Error response from daemon: No such container: kimi3_pro6000_pd_pp8_standard_router
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/p_starting_d_node0.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/p_starting_d_node0.log
new file mode 100644
index 0000000..48f2248
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/p_starting_d_node0.log
@@ -0,0 +1 @@
+Error response from daemon: No such container: kimi3_pro6000_pd_pp8_standard_decode_node0
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/p_starting_d_node1.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/p_starting_d_node1.log
new file mode 100644
index 0000000..140acb3
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/p_starting_d_node1.log
@@ -0,0 +1 @@
+Error response from daemon: No such container: kimi3_pro6000_pd_pp8_standard_decode_node1
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/p_starting_d_node2.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/p_starting_d_node2.log
new file mode 100644
index 0000000..87ead82
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/p_starting_d_node2.log
@@ -0,0 +1 @@
+Error response from daemon: No such container: kimi3_pro6000_pd_pp8_standard_decode_node2
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/p_starting_d_node3.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/p_starting_d_node3.log
new file mode 100644
index 0000000..32cb479
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/p_starting_d_node3.log
@@ -0,0 +1 @@
+Error response from daemon: No such container: kimi3_pro6000_pd_pp8_standard_decode_node3
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/p_starting_mc_master.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/p_starting_mc_master.log
new file mode 100644
index 0000000..1462bdf
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/p_starting_mc_master.log
@@ -0,0 +1,47 @@
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+I0827 04:28:16.835302    99 master.cpp:1366] Master service started on port 50051, max_threads=16, enable_metric_reporting=1, metrics_port=9003, default_kv_lease_ttl=10000, default_kv_soft_pin_ttl=1800000, allow_evict_soft_pinned_objects=1, eviction_ratio=0.05, eviction_high_watermark_ratio=0.9, enable_ha=0, enable_offload=0, enable_kv_events=0, kv_events_bind_endpoint=, kv_events_backend_id=, offload_on_evict=0, offload_force_evict=0, offloading_queue_limit=50000, offload_cap_ratio=0.5, ha_backend_type=etcd, ha_backend_connstring=, etcd_endpoints=, client_ttl=10, rpc_thread_num=16, rpc_port=50051, rpc_address=0.0.0.0, rpc_interface=, rpc_conn_timeout_seconds=0, rpc_enable_tcp_no_delay=1, rpc protocol=tcp, cluster_id=mooncake_cluster, root_fs_dir=, global_file_segment_size=9223372036854775807, memory_allocator=offset, enable_http_metadata_server=0, http_metadata_server_port=8080, http_metadata_server_host=0.0.0.0, enable_metadata_cleanup_on_timeout=0, put_start_discard_timeout_sec=30, put_start_release_timeout_sec=600, max_total_finished_tasks=10000, max_total_pending_tasks=10000, max_total_processing_tasks=10000, pending_task_timeout_sec=300, processing_task_timeout_sec=300, enable_snapshot=0, enable_snapshot_restore=0, snapshot_interval_seconds=600, snapshot_backup_dir=, snapshot_object_store_type=, snapshot_catalog_store_type=, snapshot_retention_count=2, max_retry_attempts=10, enable_cxl=0, cxl_path=/dev/dax0.0, cxl_size=8589934592
+I0827 04:28:16.866606   108 master_service.cpp:1520] Task cleanup thread started
+I0827 04:28:16.867491    99 master_admin_service.cpp:311] Master admin server started on port 9003
+I0827 04:28:16.869802   111 master_admin_service.cpp:302] Master Admin Metrics: role=standby, state=starting, service_ready=false, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:28:26.870203   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:28:36.870568   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:28:46.870873   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:28:56.871232   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:29:06.871615   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:29:16.871969   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:29:26.872316   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:29:36.872687   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:29:46.873068   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:29:56.873467   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:30:06.873914   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:30:16.874214   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:30:26.874567   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:30:36.874953   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:30:46.875332   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:30:56.875741   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:31:06.876119   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:31:16.876497   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:31:26.876866   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:31:36.877233   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:31:46.877605   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:31:56.877948   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:32:06.878437   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:32:16.879210   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:32:26.879711   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:32:36.880080   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:32:46.880565   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:32:56.881004   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:33:06.881398   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:33:16.881786   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:33:26.882170   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:33:36.882550   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:33:46.882939   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:33:56.883315   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:34:06.883746   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:34:16.884127   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:34:26.884568   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:34:36.884953   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:34:46.885334   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:34:56.885706   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:35:06.886085   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:35:16.886464   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/p_starting_p_node0.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/p_starting_p_node0.log
new file mode 100644
index 0000000..4974ca3
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/p_starting_p_node0.log
@@ -0,0 +1,926 @@
+/usr/local/lib/python3.12/dist-packages/torchao/quantization/quant_api.py:1731: SyntaxWarning: invalid escape sequence '\.'
+  """Configuration class for applying different quantization configs to modules or parameters based on their fully qualified names (FQNs).
+/sgl-workspace/sglang/python/sglang/launch_server.py:56: UserWarning: 'python -m sglang.launch_server' is still supported, but 'sglang serve' is the recommended entrypoint.
+  Example: sglang serve --model-path <model> [options]
+  warnings.warn(
+[2026-08-27 04:28:54] Breakable CUDA graph is incompatible with MLA attention; disabling prefill CUDA graph.
+[2026-08-27 04:28:55] Attention backend not specified. Use triton backend by default.
+[2026-08-27 04:28:55] Pipeline parallelism is incompatible with overlap schedule.
+[2026-08-27 04:28:58] server_args=ServerArgs(model_path='/data/hf_models/Kimi-K3', tokenizer_path='/data/hf_models/Kimi-K3', tokenizer_mode='auto', tokenizer_backend='huggingface', tokenizer_worker_num=1, detokenizer_worker_num=1, skip_tokenizer_init=False, load_format='auto', model_loader_extra_config='{}', trust_remote_code=True, context_length=None, is_embedding=False, enable_multimodal=None, revision=None, model_impl='auto', model_config_parser='auto', json_model_override_args='{}', dtype='auto', quantization=None, quantization_param_path=None, kv_cache_dtype='auto', enable_fp32_lm_head=False, modelopt_quant=None, modelopt_checkpoint_restore_path=None, modelopt_checkpoint_save_path=None, modelopt_export_path=None, quantize_and_serve=False, rl_quant_profile=None, enable_tf32_matmul=False, mem_fraction_static=0.88, max_running_requests=None, max_queued_requests=None, max_total_tokens=None, chunked_prefill_size=8192, enable_dynamic_chunking=False, max_prefill_tokens=16384, prefill_max_requests=None, schedule_policy='fcfs', enable_priority_scheduling=False, disable_priority_preemption=False, default_priority_value=None, abort_on_priority_when_disabled=False, schedule_low_priority_values_first=False, priority_scheduling_preemption_threshold=10, retraction_policy='length', schedule_conservativeness=1.0, page_size=64, swa_full_tokens_ratio=0.8, disable_hybrid_swa_memory=False, radix_eviction_policy='lru', prefill_only_disable_kv_cache=False, disable_radix_cache=True, enable_page_major_kv_layout=False, enable_unified_memory=False, disable_chunked_prefix_cache=False, disable_overlap_schedule=True, num_continuous_decode_steps=1, scheduler_recv_interval=1, enable_mixed_chunk=False, nccl_port=None, dist_timeout=3600, dist_init_addr='174.1.60.1:20000', nnodes=4, node_rank=0, tp_size=4, dcp_size=1, pp_size=8, pp_max_micro_batch_size=None, pp_async_batch_depth=0, dp_size=1, load_balance_method='follow_bootstrap_room', attn_cp_size=1, moe_dp_size=1, dcp_comm_backend='ag_rs', dcp_replicate_q_proj=None, dwdp_size=1, enable_prefill_cp=False, cp_strategy=None, enable_dsa_cache_layer_split=False, enable_dsa_prefill_context_parallel=False, dsa_prefill_cp_mode='round-robin-split', enable_prefill_context_parallel=False, prefill_cp_mode='in-seq-split', enable_cp_decode_attn_tp=False, enable_dp_attention=False, enable_dp_attention_local_control_broadcast=False, enable_dp_lm_head=False, enable_attn_tp_input_scattered=False, disable_attn_tp_gather=False, enable_p2p_check=False, device='cuda', base_gpu_id=0, gpu_id_step=1, random_seed=984607766, watchdog_timeout=300, soft_watchdog_timeout=None, sleep_on_idle=False, use_ray=False, custom_sigquit_handler=None, numa_node=None, gc_threshold=None, host='0.0.0.0', port=30000, fastapi_root_path='', smg_grpc_mode=False, grpc_mode=False, grpc_port=None, sidecar=None, sidecar_args=None, skip_server_warmup=False, warmups=None, enable_http2=False, ssl_keyfile=None, ssl_certfile=None, ssl_ca_certs=None, ssl_keyfile_password=None, enable_ssl_refresh=False, api_key=None, admin_api_key=None, served_model_name='kimi-k3', weight_version='default', chat_template=None, hf_chat_template_name=None, completion_template=None, file_storage_path='sglang_storage', enable_cache_report=False, reasoning_parser=None, default_chat_template_kwargs=None, strip_thinking_cache=False, enable_strict_thinking=False, tool_call_parser=None, tool_server=None, sampling_defaults='model', asr_max_buffer_seconds=60, asr_max_concurrent_sessions=32, preferred_sampling_params=None, allow_auto_truncate=False, stream_interval=1, batch_notify_size=16, stream_response_default_include_usage=False, incremental_streaming_output=False, enable_streaming_session=False, enable_session_radix_cache=False, log_level='info', log_level_http=None, log_requests=False, log_requests_level=2, log_requests_format='text', log_requests_target=None, uvicorn_access_log_exclude_prefixes=[], crash_dump_folder=None, show_time_cost=False, enable_metrics=False, smg_http_sidecar_port=None, enable_mfu_metrics=False, enable_metrics_for_all_schedulers=False, load_snapshot_publish_interval=15, tokenizer_metrics_custom_labels_header='x-custom-labels', tokenizer_metrics_allowed_custom_labels=None, extra_metric_labels=None, bucket_time_to_first_token=None, bucket_inter_token_latency=None, bucket_e2e_request_latency=None, prompt_tokens_buckets=None, generation_tokens_buckets=None, gc_warning_threshold_secs=0.0, decode_log_interval=40, enable_request_time_stats_logging=False, kv_events_config=None, enable_forward_pass_metrics=False, forward_pass_metrics_worker_id='', forward_pass_metrics_ipc_name=None, enable_trace=False, trace_modules='request', otlp_traces_endpoint='localhost:4317', export_metrics_to_file=False, export_metrics_to_file_dir=None, stat_loggers=None, constrained_json_whitespace_pattern=None, constrained_json_disable_any_whitespace=False, attention_backend='triton', decode_attention_backend=None, prefill_attention_backend=None, sampling_backend='flashinfer', grammar_backend='xgrammar', radix_cache_backend=None, mm_attention_backend=None, fp8_gemm_runner_backend='auto', fp4_gemm_runner_backend='auto', bf16_gemm_backend='auto', dsa_prefill_backend=None, dsa_decode_backend=None, dsa_paged_mqa_logits_backend='auto', dsa_topk_backend='sgl-kernel', disable_flashinfer_autotune=False, flashinfer_autotune_skip_ops=None, mamba_backend='triton', cuda_graph_config=CudaGraphConfig(decode=PhaseConfig(backend='disabled', max_bs=16, bs=[1, 2, 4, 8, 12, 16], tc_compiler='eager', full_prefill_max_req=None), prefill=PhaseConfig(backend='disabled', max_bs=2048, bs=[4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], tc_compiler='eager', full_prefill_max_req=None)), cuda_graph_backend_decode=None, cuda_graph_backend_prefill=None, cuda_graph_max_bs_decode=16, cuda_graph_max_bs_prefill=None, cuda_graph_bs_decode=None, cuda_graph_bs_prefill=None, cuda_graph_tc_compiler=None, disable_prefill_cuda_graph=False, disable_decode_cuda_graph=False, disable_cuda_graph=False, disable_cuda_graph_padding=False, enable_profile_cuda_graph=False, enable_cudagraph_gc=False, debug_cuda_graph=False, enable_layerwise_nvtx_marker=False, enable_nccl_nvls=False, enable_symm_mem=False, triton_attention_reduce_in_fp32=False, triton_attention_num_kv_splits=8, triton_attention_split_tile_size=None, flashinfer_mla_disable_ragged=False, enable_fused_qk_norm_rope=False, enable_precise_embedding_interpolation=False, enable_fused_moe_sum_all_reduce=False, enable_deepseek_v4_fp4_indexer=False, disable_custom_all_reduce=False, enable_mscclpp=False, enable_torch_symm_mem=False, enable_scattered_sconv=False, pre_warm_nccl=False, enable_quant_communications=False, enable_flashinfer_allreduce_fusion=False, enforce_disable_flashinfer_allreduce_fusion=False, flashinfer_allreduce_fusion_backend=None, enable_aiter_allreduce_fusion=False, enable_torch_compile=False, enable_torch_compile_debug_mode=False, torch_compile_max_bs=32, torchao_config='', speculative_algorithm=None, speculative_draft_model_path=None, speculative_draft_model_revision=None, speculative_draft_load_format=None, speculative_num_steps=None, speculative_eagle_topk=None, speculative_num_draft_tokens=None, speculative_dflash_block_size=None, speculative_dspark_block_size=None, speculative_dspark_sps_table_path=None, speculative_dspark_confidence_sts_path=None, speculative_dspark_align_verify_tokens_to_graph_tier=False, speculative_accept_threshold_single=1.0, speculative_accept_threshold_acc=1.0, speculative_use_rejection_sampling=False, speculative_token_map=None, speculative_attention_mode='prefill', speculative_draft_attention_backend=None, speculative_draft_window_size=None, speculative_moe_runner_backend='flashinfer_mxfp4', speculative_moe_a2a_backend=None, speculative_draft_model_quantization=None, speculative_skip_dp_mlp_sync=False, enable_multi_layer_eagle=False, speculative_adaptive=False, speculative_adaptive_config=None, decoupled_spec_bind_endpoint=None, decoupled_spec_connect_endpoints=None, decoupled_spec_rank=None, decoupled_spec_role='null', spec_trace_dir=None, speculative_ngram_min_bfs_breadth=1, speculative_ngram_max_bfs_breadth=10, speculative_ngram_match_type='BFS', speculative_ngram_max_trie_depth=18, speculative_ngram_capacity=10000000, speculative_ngram_external_corpus_path=None, speculative_ngram_external_sam_budget=0, speculative_ngram_external_corpus_max_tokens=10000000, ep_size=4, moe_a2a_backend='none', moe_runner_backend='flashinfer_mxfp4', flashinfer_mxfp4_moe_precision='default', deepep_mode='auto', fuseep_mode=2, deepep_dispatcher_output_dtype='auto', ep_num_redundant_experts=0, ep_dispatch_algorithm=None, init_expert_location='trivial', enable_eplb=False, eplb_algorithm='auto', eplb_rebalance_num_iterations=1000, eplb_rebalance_layers_per_chunk=None, eplb_min_rebalancing_utilization_threshold=1.0, expert_distribution_recorder_mode=None, expert_distribution_recorder_buffer_size=1000, enable_expert_distribution_metrics=False, deepep_config=None, moe_dense_tp_size=None, elastic_ep_backend=None, enable_elastic_expert_backup=False, mooncake_ib_device=None, enable_waterfill=False, ep_join_mode=None, ep_join_rank_offset=0, elastic_ep_initial_size=None, max_ep_size=None, elastic_ep_scale_timeout=600, elastic_ep_rejoin=False, disable_flashinfer_cutlass_moe_fp4_allgather=False, disable_shared_experts_fusion=False, enforce_shared_experts_fusion=False, max_mamba_cache_size=None, mamba_ssm_dtype=None, mamba_max_states_per_path=-1, enable_mamba_cache_stochastic_rounding=False, mamba_cache_philox_rounds=0, mamba_full_memory_ratio=0.36, mamba_radix_cache_strategy='extra_buffer_lazy', uses_mamba_radix_cache=False, mamba_track_interval=256, enable_int8_mamba_checkpoint=False, int8_mamba_ckpt_size=None, linear_attn_backend='triton', linear_attn_decode_backend=None, linear_attn_prefill_backend=None, linear_attn_verify_backend=None, enable_linear_replayssm=False, linear_replayssm_cache_len=16, enable_linear_replayssm_spec=False, enable_hierarchical_cache=False, hicache_ratio=2.0, hicache_size=0, hicache_write_policy='write_through', hicache_io_backend='kernel', hicache_mem_layout='page_first', hicache_storage_backend=None, hicache_storage_prefetch_policy='timeout', hicache_storage_backend_extra_config=None, enable_hisparse=False, hisparse_config=None, enable_broadcast_mm_inputs_process=False, enable_prefix_mm_cache=False, mm_enable_dp_encoder=False, mm_process_config={}, mm_processor_worker_num=0, mm_io_worker_num=0, limit_mm_data_per_request=None, enable_mm_global_cache=False, disable_fast_image_processor=False, mm_feature_transport='cpu', keep_mm_feature_on_device=False, enable_lora=None, enable_lora_overlap_loading=None, max_lora_rank=None, lora_target_modules=None, lora_paths=None, max_loaded_loras=None, max_loras_per_batch=8, lora_eviction_policy='lru', lora_backend='csgmv', max_lora_chunk_size=16, experts_shared_outer_loras=None, lora_use_virtual_experts=False, lora_strict_loading=False, lora_drain_wait_threshold=0.0, enable_two_batch_overlap=False, enable_single_batch_overlap=False, tbo_token_distribution_threshold=0.48, cpu_offload_gb=0, offload_group_size=-1, offload_num_in_group=1, offload_prefetch_step=1, offload_mode='cpu', enable_lmcache=False, lmcache_config_file=None, enable_flexkv=False, flexkv_config_file=None, kt_weight_path=None, kt_method='AMXINT4', kt_cpuinfer=None, kt_threadpool_count=2, kt_num_gpu_experts=None, kt_max_deferred_experts_per_token=None, dllm_algorithm=None, dllm_algorithm_config=None, dllm_fdfo=True, disaggregation_mode='prefill', disaggregation_transfer_backend='mooncake', disaggregation_bootstrap_port=28800, disaggregation_ib_device='mlx5_0,mlx5_1,mlx5_2,mlx5_3', disaggregation_decode_enable_radix_cache=False, disaggregation_decode_enable_offload_kvcache=False, num_reserved_decode_tokens=512, disaggregation_decode_extra_slots=None, disaggregation_decode_polling_interval=1, optimistic_prefill_attempts=0, encoder_only=False, language_only=False, encoder_transfer_backend='zmq_to_tokenizer', encoder_urls=[], encoder_bootstrap_port=8997, encoder_register_urls=[], enable_adaptive_dispatch_to_encoder=False, enable_pdmux=False, pdmux_config_path=None, sm_group_num=8, custom_weight_loader=[], weight_loader_disable_mmap=False, weight_loader_prefetch_checkpoints=False, weight_loader_prefetch_num_threads=4, weight_loader_drop_cache_after_load=False, remote_instance_weight_loader_seed_instance_ip=None, remote_instance_weight_loader_seed_instance_service_port=None, remote_instance_weight_loader_send_weights_group_ports=None, remote_instance_weight_loader_backend='nccl', remote_instance_weight_loader_start_seed_via_transfer_engine=False, engine_info_bootstrap_port=6789, modelexpress_config=None, download_dir=None, model_checksum=None, delete_ckpt_after_loading=False, decrypted_config_file=None, decrypted_draft_config_file=None, checkpoint_engine_wait_weights_before_ready=False, enable_prefill_delayer=False, prefill_delayer_max_delay_passes=30, prefill_delayer_token_usage_low_watermark=None, prefill_delayer_forward_passes_buckets=None, prefill_delayer_wait_seconds_buckets=None, prefill_delayer_queue_min_ratio=None, prefill_delayer_max_delay_ms=None, min_free_slots_delay=None, enable_deterministic_inference=False, rl_on_policy_target=None, kv_canary='none', kv_canary_real_data='none', kv_canary_sweep_interval=0, enable_dynamic_batch_tokenizer=False, dynamic_batch_tokenizer_batch_size=32, dynamic_batch_tokenizer_batch_timeout=0.002, enable_tokenizer_batch_encode=False, disable_tokenizer_batch_decode=False, debug_tensor_dump_output_folder=None, debug_tensor_dump_layers=None, debug_tensor_dump_input_file=None, enable_memory_saver=False, enable_weights_cpu_backup=False, enable_draft_weights_cpu_backup=False, enable_custom_logit_processor=False, enable_return_hidden_states=False, enable_return_routed_experts=False, enable_return_indexer_topk=False, disable_outlines_disk_cache=False, enable_mis=False, weight_cache_mode='off', weight_cache_socket=None, weight_cache_timeout=1800, forward_hooks=None, msprobe_dump_config=None)
+[2026-08-27 04:28:58] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:58] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:58] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:58] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:58] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:58] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:58] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:58] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:01] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:01] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:01] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:03] Multimodal processor concurrency enabled with 2 isolated worker threads (auto).
+[2026-08-27 04:29:03] CommonKVBootstrapServer started successfully on 0.0.0.0:28800
+[2026-08-27 04:29:03] No HuggingFace chat template found
+[2026-08-27 04:29:03] No chat template found, defaulting to 'string' content format
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:15 PP0 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:15 PP0 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:16] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:16] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:16 PP1 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:16 PP1 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+/sgl-workspace/sglang/python/sglang/srt/utils/hf_transformers/tokenizer.py:423: UserWarning: Using a slow tokenizer. This might cause a significant slowdown. Consider using a fast tokenizer instead.
+  warnings.warn(
+[2026-08-27 04:29:16] Applying special tokens cache patch for Kimi tokenizer: <class 'transformers_modules.Kimi_hyphen_K3.caf40b2773f9029e.tokenization_kimi.TikTokenTokenizer'>
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:16 PP1 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:16 PP1 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:16 PP0 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:16 PP0 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:16 PP1 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:16 PP1 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:17 PP0 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:17 PP0 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:17 PP1 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:17 PP1 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:17 PP0 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:17 PP0 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:17 PP0 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:17.587481   237 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:17.587555   237 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:17.587566   237 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.1 port: 12001
+I0827 04:29:17.587602   237 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.1:16830
+I0827 04:29:17.587723   237 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:17.613691   237 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:17.620473   237 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:17.627439   237 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:17.653980   237 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:17.656663   237 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:17.656692   237 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:17.656746   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:17.663833   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:17.665359   237 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:01
+I0827 04:29:17.665387   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:17.690173   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:17.691654   237 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:01
+I0827 04:29:17.691689   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:17.718215   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:17.719650   237 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:01
+I0827 04:29:17.719686   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:17.746297   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:17.747779   237 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:01
+I0827 04:29:17.747891   237 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:17 PP0 TP1 EP1] Init torch distributed begin.
+[2026-08-27 04:29:17 PP1 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:18.219898   241 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:18.219964   241 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:18.219973   241 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.1 port: 12001
+I0827 04:29:18.220007   241 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.1:15425
+I0827 04:29:18.220109   241 transfer_engine_impl.cpp:258] Auto-discovering topology...
+[2026-08-27 04:29:18 PP1 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+I0827 04:29:18.244532   241 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:18.252967   241 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:18.261787   241 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:18.289944   241 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:18.293610   241 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:18.293633   241 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:18.293681   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.301677   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:18.303126   241 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:01
+I0827 04:29:18.303155   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.330181   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:18.331590   241 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:01
+I0827 04:29:18.331614   241 rdma_context.cpp:205] Using SIEVE endpoint store
+[2026-08-27 04:29:18 PP0 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+I0827 04:29:18.358213   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:18.359581   241 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:01
+I0827 04:29:18.359604   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.386204   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:18.387562   241 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:01
+I0827 04:29:18.387681   241 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:18 PP1 TP1 EP1] Init torch distributed begin.
+[2026-08-27 04:29:18 PP1 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:18.588883   242 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:18.588970   242 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:18.588984   242 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.1 port: 12001
+I0827 04:29:18.589022   242 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.1:15650
+I0827 04:29:18.589143   242 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:18.594419   242 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:18.599439   242 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:18.604553   242 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:18.609825   242 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:18.612036   242 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:18.612056   242 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:18.612097   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.618021   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:18.619308   242 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:01
+I0827 04:29:18.619333   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.646080   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:18.647465   242 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:01
+I0827 04:29:18.647495   242 rdma_context.cpp:205] Using SIEVE endpoint store
+[2026-08-27 04:29:18 PP0 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+I0827 04:29:18.674199   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:18.675598   242 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:01
+I0827 04:29:18.675621   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.682866   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:18.684276   242 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:01
+I0827 04:29:18.684391   242 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:18 PP1 TP2 EP2] Init torch distributed begin.
+[2026-08-27 04:29:18 PP1 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:18.695660   238 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:18.695755   238 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:18.695767   238 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.1 port: 12001
+I0827 04:29:18.695806   238 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.1:15123
+I0827 04:29:18.695909   238 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:18.701093   238 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:18.706161   238 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:18.711323   238 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:18.737910   238 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:18.740556   238 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:18.740602   238 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:18.740651   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.747717   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:18.749193   238 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:01
+I0827 04:29:18.749220   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.774145   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:18.775534   238 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:01
+I0827 04:29:18.775558   238 rdma_context.cpp:205] Using SIEVE endpoint store
+[2026-08-27 04:29:18 PP0 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+I0827 04:29:18.802196   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:18.803623   238 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:01
+I0827 04:29:18.803649   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.810976   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:18.812341   238 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:01
+I0827 04:29:18.812443   238 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:18 PP0 TP2 EP2] Init torch distributed begin.
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:19.084213   243 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:19.084332   243 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:19.084348   243 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.1 port: 12001
+I0827 04:29:19.084400   243 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.1:15411
+I0827 04:29:19.084602   243 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:19.114387   243 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:19.121773   243 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:19.146405   243 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:19.153759   243 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:19.156515   243 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:19.156548   243 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:19.156603   243 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.163982   243 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:19.165694   243 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:01
+I0827 04:29:19.165726   243 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.173307   243 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:19.174904   243 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:01
+I0827 04:29:19.174932   243 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.206454   243 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:19.208142   243 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:01
+I0827 04:29:19.208168   243 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.215909   243 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:19.217583   243 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:01
+I0827 04:29:19.217706   243 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:19 PP1 TP3 EP3] Init torch distributed begin.
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:19.239588   236 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:19.239699   236 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:19.239715   236 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.1 port: 12001
+I0827 04:29:19.239763   236 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.1:15605
+I0827 04:29:19.239933   236 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:19.240546   240 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:19.240644   240 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:19.240658   240 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.1 port: 12001
+I0827 04:29:19.240702   240 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.1:15487
+I0827 04:29:19.240862   240 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:19.242301   239 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:19.242408   239 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:19.242422   239 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.1 port: 12001
+I0827 04:29:19.242470   239 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.1:15552
+I0827 04:29:19.242636   239 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:19.246201   240 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:19.248451   239 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:19.251204   240 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:19.254226   239 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:19.256254   240 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:19.259852   239 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:19.261372   240 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:19.263411   240 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:19.263432   240 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:19.263473   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.265838   239 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:19.268059   239 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:19.268082   239 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:19.268126   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.269106   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:19.270298   236 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:19.270401   240 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:01
+I0827 04:29:19.270423   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.274480   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:19.275991   239 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:01
+I0827 04:29:19.276019   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.276877   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:19.278193   240 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:01
+I0827 04:29:19.278220   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.297820   236 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:19.302485   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:19.302623   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:19.303958   239 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:01
+I0827 04:29:19.303983   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.304121   240 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:01
+I0827 04:29:19.304147   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.311621   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:19.313341   240 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:01
+I0827 04:29:19.313478   240 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:19 PP1 TP0 EP0] Init torch distributed begin.
+I0827 04:29:19.322126   236 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:19.330281   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:19.331759   239 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:01
+I0827 04:29:19.331784   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.339274   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:19.340766   239 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:01
+I0827 04:29:19.340884   239 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:19 PP0 TP3 EP3] Init torch distributed begin.
+I0827 04:29:19.345958   236 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:19.348632   236 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:19.348659   236 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:19.348712   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.355962   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:19.417273   236 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:01
+I0827 04:29:19.417309   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.424894   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:19.426437   236 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:01
+I0827 04:29:19.426491   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.434231   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:19.435868   236 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:01
+I0827 04:29:19.435896   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.443578   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:19.445148   236 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:01
+I0827 04:29:19.445264   236 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:19 PP0 TP0 EP0] Init torch distributed begin.
+[2026-08-27 04:29:26 PP0 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:26 PP1 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:27 PP1 TP2 EP2] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP1 TP0 EP0] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP1 TP3 EP3] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP1 TP1 EP1] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP0 TP3 EP3] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP0 TP0 EP0] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP0 TP1 EP1] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP0 TP2 EP2] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP0 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:27 PP0 TP3 EP3] sglang is using nccl==2.28.9
+[2026-08-27 04:29:27 PP0 TP1 EP1] sglang is using nccl==2.28.9
+[2026-08-27 04:29:27 PP0 TP2 EP2] sglang is using nccl==2.28.9
+[2026-08-27 04:29:27 PP0 TP0 EP0] Init torch distributed ends. elapsed=8.22 s, mem usage=0.46 GB
+[2026-08-27 04:29:27 PP1 TP3 EP3] Init torch distributed ends. elapsed=8.45 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP1 TP1 EP1] Init torch distributed ends. elapsed=9.28 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP1 TP2 EP2] Init torch distributed ends. elapsed=8.98 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP1 TP0 EP0] Init torch distributed ends. elapsed=8.35 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP0 TP2 EP2] Init torch distributed ends. elapsed=8.85 s, mem usage=0.46 GB
+[2026-08-27 04:29:27 PP0 TP3 EP3] Init torch distributed ends. elapsed=8.32 s, mem usage=0.46 GB
+[2026-08-27 04:29:27 PP0 TP1 EP1] Init torch distributed ends. elapsed=9.92 s, mem usage=0.46 GB
+[2026-08-27 04:29:27 PP1 TP0 EP0] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP1 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP0 TP0 EP0] Load weight begin. avail mem=82.12 GB
+[2026-08-27 04:29:27 PP1 TP3 EP3] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP1 TP2 EP2] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP0 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP1 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP1 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP1 TP1 EP1] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP0 TP2 EP2] Load weight begin. avail mem=82.12 GB
+[2026-08-27 04:29:27 PP1 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP0 TP3 EP3] Load weight begin. avail mem=82.12 GB
+[2026-08-27 04:29:27 PP0 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP0 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP1 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:27 PP1 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:27 PP0 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:27 PP1 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:27 PP1 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:27 PP1 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:27 PP0 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:27 PP1 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:27 PP1 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:27 PP1 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:27 PP0 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:27 PP0 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:27 PP0 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:27 PP0 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:27 PP0 TP1 EP1] Load weight begin. avail mem=82.12 GB
+[2026-08-27 04:29:27 PP0 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP0 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP0 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+
Multi-thread loading shards:   0% Completed | 0/96 [00:00<?, ?it/s]
Multi-thread loading shards:   1% Completed | 1/96 [00:00<01:01,  1.55it/s]
Multi-thread loading shards:   2% Completed | 2/96 [00:09<08:49,  5.63s/it]
Multi-thread loading shards:   3% Completed | 3/96 [00:12<06:35,  4.25s/it]
Multi-thread loading shards:   4% Completed | 4/96 [00:14<05:26,  3.55s/it]
Multi-thread loading shards:   5% Completed | 5/96 [00:17<04:43,  3.11s/it]
Multi-thread loading shards:   6% Completed | 6/96 [00:19<04:17,  2.86s/it]
Multi-thread loading shards:   7% Completed | 7/96 [00:21<04:00,  2.71s/it]
Multi-thread loading shards:   8% Completed | 8/96 [00:24<03:49,  2.61s/it]
Multi-thread loading shards:   9% Completed | 9/96 [00:26<03:44,  2.58s/it]
Multi-thread loading shards:  10% Completed | 10/96 [00:29<03:40,  2.56s/it]
Multi-thread loading shards:  11% Completed | 11/96 [00:31<03:36,  2.54s/it]
Multi-thread loading shards:  12% Completed | 12/96 [00:32<02:43,  1.95s/it]
Multi-thread loading shards:  14% Completed | 13/96 [00:32<02:05,  1.52s/it]
Multi-thread loading shards:  15% Completed | 14/96 [00:33<01:33,  1.14s/it]
Multi-thread loading shards:  16% Completed | 15/96 [00:33<01:21,  1.01s/it]
Multi-thread loading shards:  17% Completed | 16/96 [00:34<01:19,  1.00it/s]
Multi-thread loading shards:  18% Completed | 17/96 [00:35<01:02,  1.26it/s]
Multi-thread loading shards:  19% Completed | 18/96 [00:35<00:51,  1.53it/s]
Multi-thread loading shards:  20% Completed | 19/96 [00:35<00:42,  1.80it/s]
Multi-thread loading shards:  21% Completed | 20/96 [00:36<00:37,  2.01it/s]
Multi-thread loading shards:  22% Completed | 21/96 [00:36<00:33,  2.23it/s]
Multi-thread loading shards:  23% Completed | 22/96 [00:36<00:30,  2.43it/s]
Multi-thread loading shards:  24% Completed | 23/96 [00:37<00:27,  2.68it/s]
Multi-thread loading shards:  25% Completed | 24/96 [00:37<00:25,  2.80it/s]
Multi-thread loading shards:  26% Completed | 25/96 [00:37<00:24,  2.89it/s]
Multi-thread loading shards:  27% Completed | 26/96 [00:38<00:23,  2.99it/s]
Multi-thread loading shards:  28% Completed | 27/96 [00:38<00:22,  3.01it/s]
Multi-thread loading shards:  29% Completed | 28/96 [00:38<00:22,  2.99it/s]
Multi-thread loading shards:  30% Completed | 29/96 [00:39<00:22,  2.98it/s]
Multi-thread loading shards:  31% Completed | 30/96 [00:39<00:22,  2.98it/s]
Multi-thread loading shards:  32% Completed | 31/96 [00:39<00:21,  3.01it/s]
Multi-thread loading shards:  33% Completed | 32/96 [00:40<00:21,  3.00it/s]
Multi-thread loading shards:  34% Completed | 33/96 [00:41<00:35,  1.79it/s]
Multi-thread loading shards:  35% Completed | 34/96 [00:41<00:30,  2.02it/s]
Multi-thread loading shards:  36% Completed | 35/96 [00:41<00:27,  2.23it/s]
Multi-thread loading shards:  38% Completed | 36/96 [00:42<00:24,  2.41it/s]
Multi-thread loading shards:  39% Completed | 37/96 [00:42<00:22,  2.57it/s]
Multi-thread loading shards:  40% Completed | 38/96 [00:42<00:21,  2.64it/s]
Multi-thread loading shards:  41% Completed | 39/96 [00:43<00:20,  2.73it/s]
Multi-thread loading shards:  42% Completed | 40/96 [00:43<00:20,  2.80it/s]
Multi-thread loading shards:  43% Completed | 41/96 [00:43<00:19,  2.79it/s]
Multi-thread loading shards:  44% Completed | 42/96 [00:44<00:19,  2.80it/s]
Multi-thread loading shards:  45% Completed | 43/96 [00:44<00:22,  2.39it/s]
Multi-thread loading shards:  46% Completed | 44/96 [00:45<00:20,  2.51it/s]
Multi-thread loading shards:  47% Completed | 45/96 [00:45<00:19,  2.64it/s]
Multi-thread loading shards:  48% Completed | 46/96 [00:45<00:18,  2.67it/s]
Multi-thread loading shards:  49% Completed | 47/96 [00:46<00:18,  2.61it/s]
Multi-thread loading shards:  50% Completed | 48/96 [00:46<00:17,  2.72it/s]
Multi-thread loading shards:  51% Completed | 49/96 [00:47<00:16,  2.84it/s]
Multi-thread loading shards:  52% Completed | 50/96 [00:47<00:16,  2.81it/s]
Multi-thread loading shards:  53% Completed | 51/96 [00:47<00:15,  2.93it/s]
Multi-thread loading shards:  54% Completed | 52/96 [00:48<00:21,  2.05it/s]
Multi-thread loading shards:  55% Completed | 53/96 [00:49<00:33,  1.30it/s]
Multi-thread loading shards:  56% Completed | 54/96 [00:50<00:27,  1.55it/s]
Multi-thread loading shards:  57% Completed | 55/96 [00:50<00:22,  1.82it/s]
Multi-thread loading shards:  58% Completed | 56/96 [00:50<00:19,  2.06it/s]
Multi-thread loading shards:  59% Completed | 57/96 [00:51<00:17,  2.25it/s]
Multi-thread loading shards:  60% Completed | 58/96 [00:51<00:15,  2.40it/s]
Multi-thread loading shards:  61% Completed | 59/96 [00:51<00:14,  2.56it/s]
Multi-thread loading shards:  62% Completed | 60/96 [00:52<00:13,  2.67it/s]
Multi-thread loading shards:  64% Completed | 61/96 [00:52<00:12,  2.77it/s]
Multi-thread loading shards:  65% Completed | 62/96 [00:52<00:11,  2.84it/s]
Multi-thread loading shards:  66% Completed | 63/96 [00:53<00:11,  2.89it/s]
Multi-thread loading shards:  67% Completed | 64/96 [00:53<00:11,  2.87it/s]
Multi-thread loading shards:  68% Completed | 65/96 [00:54<00:10,  2.89it/s]
Multi-thread loading shards:  69% Completed | 66/96 [00:54<00:10,  2.91it/s]
Multi-thread loading shards:  70% Completed | 67/96 [00:54<00:10,  2.90it/s]
Multi-thread loading shards:  71% Completed | 68/96 [00:55<00:09,  2.90it/s]
Multi-thread loading shards:  72% Completed | 69/96 [00:55<00:09,  2.89it/s]
Multi-thread loading shards:  73% Completed | 70/96 [00:56<00:14,  1.85it/s]
Multi-thread loading shards:  74% Completed | 71/96 [00:56<00:11,  2.10it/s]
Multi-thread loading shards:  75% Completed | 72/96 [00:57<00:10,  2.23it/s]
Multi-thread loading shards:  76% Completed | 73/96 [00:57<00:09,  2.39it/s]
Multi-thread loading shards:  77% Completed | 74/96 [00:57<00:08,  2.48it/s]
Multi-thread loading shards:  78% Completed | 75/96 [00:58<00:09,  2.15it/s]
Multi-thread loading shards:  79% Completed | 76/96 [00:59<00:10,  1.89it/s]
Multi-thread loading shards:  80% Completed | 77/96 [00:59<00:09,  2.05it/s]
Multi-thread loading shards:  81% Completed | 78/96 [00:59<00:08,  2.24it/s]
Multi-thread loading shards:  82% Completed | 79/96 [01:00<00:07,  2.39it/s]
Multi-thread loading shards:  83% Completed | 80/96 [01:00<00:06,  2.51it/s]
Multi-thread loading shards:  84% Completed | 81/96 [01:00<00:05,  2.65it/s]
Multi-thread loading shards:  85% Completed | 82/96 [01:01<00:04,  2.84it/s]
Multi-thread loading shards:  86% Completed | 83/96 [01:02<00:06,  1.99it/s]
Multi-thread loading shards:  88% Completed | 84/96 [01:02<00:06,  1.99it/s]
Multi-thread loading shards:  90% Completed | 86/96 [01:02<00:03,  3.27it/s]
Multi-thread loading shards:  92% Completed | 88/96 [01:02<00:01,  4.72it/s]
Multi-thread loading shards:  94% Completed | 90/96 [01:02<00:00,  6.49it/s]
Multi-thread loading shards:  96% Completed | 92/96 [01:03<00:00,  8.37it/s]
Multi-thread loading shards:  98% Completed | 94/96 [01:03<00:00,  8.91it/s]
Multi-thread loading shards: 100% Completed | 96/96 [01:03<00:00,  9.35it/s]
Multi-thread loading shards: 100% Completed | 96/96 [01:03<00:00,  1.51it/s]
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:33 PP0 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:33 PP0 TP2 EP2] Load weight end. elapsed=65.63 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=39.03 GB, mem usage=43.09 GB.
+[2026-08-27 04:30:33 PP0 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:33 PP0 TP0 EP0] Load weight end. elapsed=65.81 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=39.03 GB, mem usage=43.09 GB.
+[2026-08-27 04:30:35 PP0 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:35 PP0 TP3 EP3] Load weight end. elapsed=67.37 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=39.03 GB, mem usage=43.09 GB.
+[2026-08-27 04:30:35 PP0 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:35 PP0 TP1 EP1] Load weight end. elapsed=67.46 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=39.03 GB, mem usage=43.09 GB.
+[2026-08-27 04:30:35 PP0 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:35 PP0 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:35 PP0 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:35 PP0 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:35 PP0 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:35 PP0 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:35 PP0 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:35 PP0 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:36 PP1 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:36 PP1 TP3 EP3] Load weight end. elapsed=68.32 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=36.12 GB, mem usage=45.92 GB.
+[2026-08-27 04:30:36 PP1 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP1 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP1 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP1 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP1 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP1 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP1 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP1 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:37 PP1 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:37 PP1 TP1 EP1] Load weight end. elapsed=69.55 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=36.12 GB, mem usage=45.92 GB.
+[2026-08-27 04:30:37 PP1 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:37 PP1 TP0 EP0] Load weight end. elapsed=69.66 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=36.12 GB, mem usage=45.92 GB.
+[2026-08-27 04:30:39 PP1 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:39 PP1 TP2 EP2] Load weight end. elapsed=71.33 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=36.12 GB, mem usage=45.92 GB.
+[2026-08-27 04:30:39 PP1 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:39 PP1 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:39 PP1 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:39 PP1 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:39 PP1 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:39 PP1 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:39 PP1 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:39 PP1 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:53 PP0 TP0 EP0] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:53 PP1 TP3 EP3] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:53 PP1 TP2 EP2] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:53 PP1 TP1 EP1] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:53 PP1 TP0 EP0] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:53 PP0 TP2 EP2] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:53 PP0 TP3 EP3] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:53 PP0 TP1 EP1] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:53 PP0 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:53 PP1 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:53 PP1 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:53 PP1 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:53 PP0 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:53 PP1 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:53 PP0 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:53 PP0 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:53 PP0 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 8.16 GB
+[2026-08-27 04:30:53 PP0 TP0 EP0] Memory pool end. avail mem=30.14 GB
+[2026-08-27 04:30:53 PP0 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 8.16 GB
+[2026-08-27 04:30:53 PP0 TP2 EP2] Memory pool end. avail mem=30.14 GB
+[2026-08-27 04:30:53 PP1 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:53 PP1 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:53 PP1 TP0 EP0] Memory pool end. avail mem=23.24 GB
+[2026-08-27 04:30:53 PP1 TP2 EP2] Memory pool end. avail mem=23.24 GB
+[2026-08-27 04:30:53 PP0 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 8.16 GB
+[2026-08-27 04:30:53 PP1 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:53 PP0 TP3 EP3] Memory pool end. avail mem=30.14 GB
+[2026-08-27 04:30:53 PP1 TP1 EP1] Memory pool end. avail mem=23.24 GB
+[2026-08-27 04:30:53 PP1 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:53 PP0 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 8.16 GB
+[2026-08-27 04:30:53 PP1 TP3 EP3] Memory pool end. avail mem=23.24 GB
+[2026-08-27 04:30:53 PP0 TP1 EP1] Memory pool end. avail mem=30.14 GB
+[2026-08-27 04:30:54 PP0 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:30:54 PP0 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:30:54 PP0 TP0 EP0] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp0_pp0_dp0.json
+[2026-08-27 04:30:54 PP1 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:30:54 PP1 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:30:54 PP0 TP2 EP2] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp2_pp0_dp0.json
+[2026-08-27 04:30:54 PP1 TP2 EP2] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp2_pp1_dp0.json
+[2026-08-27 04:30:54 PP1 TP0 EP0] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp0_pp1_dp0.json
+[2026-08-27 04:30:54 PP1 TP1 EP1] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp1_pp1_dp0.json
+[2026-08-27 04:30:54 PP1 TP3 EP3] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp3_pp1_dp0.json
+[2026-08-27 04:30:54 PP0 TP1 EP1] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp1_pp0_dp0.json
+[2026-08-27 04:30:54 PP0 TP3 EP3] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp3_pp0_dp0.json
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.46profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.32profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.51profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  6.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.09profile/s]
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.07profile/s]
+[TensorRT-LLM][INFO] Set logger level to INFO
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.48profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.11profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.29profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.37profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.16profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.82profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.18profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.58profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:01<00:02,  1.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.66profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.03profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.49profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:01<00:01,  2.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.25profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:01,  2.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:02<00:00,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:02<00:00,  3.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.20profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.01profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  9.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  9.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  9.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 12.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 12.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 11.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 11.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.78profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.16profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  3.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.75profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.50profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.92profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.68profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  3.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.97profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.87profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 10.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.51profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.74profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 11.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.30profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.70profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.72profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.63profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.53profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.29profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.58profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  8.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 10.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.71profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.62profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.43profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.18profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.29profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.37profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.53profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.37profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.35profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.00profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.71profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.60profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.78profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.79profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  8.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.86profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  8.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.16profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.78profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  3.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.15profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.94profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.93profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.25profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.23profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.97profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.79profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.96profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.64profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.12profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.75profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.12profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.25profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.91profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.05profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.66profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.90profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  9.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.73profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.55profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.49profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.13profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.74profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.17profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  3.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.72profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.57profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.59profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.64profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.83profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 13.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.49profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 12.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  9.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  7.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.32profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.93profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.29profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.72profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.75profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.46profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.51profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  3.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.90profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.16profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.72profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.63profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.83profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.70profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.59profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.55profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  6.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.93profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.33profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.75profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.30profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  6.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  7.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.79profile/s]
+[2026-08-27 04:31:21 PP0 TP1 EP1] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.53profile/s][2026-08-27 04:31:21 PP0 TP0 EP0] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.97profile/s]
+[2026-08-27 04:31:21 PP0 TP3 EP3] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.43profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.06profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s][2026-08-27 04:31:21 PP0 TP2 EP2] FlashInfer autotune completed.
+[2026-08-27 04:31:21 PP0 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:21 PP0 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:21 PP0 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:21 PP0 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:21 PP0 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:21 PP0 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:21 PP0 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:21 PP0 TP0 EP0] max_total_num_tokens=3801408, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=54, context_len=1048576, available_gpu_mem=27.35 GB
+[2026-08-27 04:31:21 PP0 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.86profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.58profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:01<00:07,  1.18s/profile]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  2.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.27profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.55profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  3.96profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.86profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  6.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  7.01profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.23profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.05profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  6.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  7.06profile/s]
+[2026-08-27 04:31:26 PP1 TP1 EP1] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.40profile/s]
+[2026-08-27 04:31:26 PP1 TP2 EP2] FlashInfer autotune completed.
+[2026-08-27 04:31:26 PP1 TP0 EP0] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.81profile/s]
+[2026-08-27 04:31:26 PP1 TP3 EP3] FlashInfer autotune completed.
+[2026-08-27 04:31:26 PP1 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:26 PP1 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:26 PP1 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:26 PP1 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:26 PP1 TP0 EP0] max_total_num_tokens=3801408, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=54, context_len=1048576, available_gpu_mem=20.45 GB
+[2026-08-27 04:31:26 PP1 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:26 PP1 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:26 PP1 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:26 PP1 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+W0827 04:31:30.805795  2607 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fe48940, len: 6912: Operation not permitted [1]
+W0827 04:31:30.805797  2608 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4f684400, len: 6912: Operation not permitted [1]
+W0827 04:31:30.805945  2609 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4f962700, len: 6912: Operation not permitted [1]
+W0827 04:31:30.805979  2610 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4f9e8a00, len: 6912: Operation not permitted [1]
+W0827 04:31:30.806130  2612 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fe5f640, len: 55296: Operation not permitted [1]
+W0827 04:31:30.806133  2611 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4f83f000, len: 55296: Operation not permitted [1]
+W0827 04:31:30.806264  2613 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fe328c0, len: 6912: Operation not permitted [1]
+W0827 04:31:30.806334  2614 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fa75cc0, len: 13824: Operation not permitted [1]
+W0827 04:31:30.806449  2615 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fa87f80, len: 6912: Operation not permitted [1]
+W0827 04:31:30.806493  2616 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fa29a80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.203712  2635 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50c65e80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.203859  2636 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50c6bb40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.203975  2637 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50d1ffc0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.204025  2638 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50a0c300, len: 6912: Operation not permitted [1]
+W0827 04:31:31.204075  2640 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50de1940, len: 55296: Operation not permitted [1]
+W0827 04:31:31.204082  2639 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50dc93c0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.204438  2641 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50c87940, len: 6912: Operation not permitted [1]
+W0827 04:31:31.204561  2642 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50cffb00, len: 13824: Operation not permitted [1]
+W0827 04:31:31.204641  2644 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50dc7380, len: 6912: Operation not permitted [1]
+W0827 04:31:31.204722  2643 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50c243c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.323861  2661 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50b13c40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.323863  2662 memory_location.cpp:73] Failed to get NUMA node, addr: 0x511191c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.323912  2663 memory_location.cpp:73] Failed to get NUMA node, addr: 0x511b37c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.324069  2664 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50b2da80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.324110  2666 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5115c980, len: 55296: Operation not permitted [1]
+W0827 04:31:31.324155  2665 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5116ac40, len: 55296: Operation not permitted [1]
+W0827 04:31:31.324479  2667 memory_location.cpp:73] Failed to get NUMA node, addr: 0x511a7e80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.324587  2668 memory_location.cpp:73] Failed to get NUMA node, addr: 0x511b8cc0, len: 13824: Operation not permitted [1]
+W0827 04:31:31.324621  2669 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50fc6800, len: 6912: Operation not permitted [1]
+W0827 04:31:31.324682  2670 memory_location.cpp:73] Failed to get NUMA node, addr: 0x511bc780, len: 6912: Operation not permitted [1]
+W0827 04:31:31.543082  2689 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4902e440, len: 6912: Operation not permitted [1]
+W0827 04:31:31.543203  2690 memory_location.cpp:73] Failed to get NUMA node, addr: 0x49484200, len: 6912: Operation not permitted [1]
+W0827 04:31:31.543314  2691 memory_location.cpp:73] Failed to get NUMA node, addr: 0x49119600, len: 6912: Operation not permitted [1]
+W0827 04:31:31.543375  2692 memory_location.cpp:73] Failed to get NUMA node, addr: 0x493dc780, len: 6912: Operation not permitted [1]
+W0827 04:31:31.543467  2693 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51ac5fc0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.543601  2694 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51a9e780, len: 55296: Operation not permitted [1]
+W0827 04:31:31.543751  2695 memory_location.cpp:73] Failed to get NUMA node, addr: 0x49149a40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.543912  2696 memory_location.cpp:73] Failed to get NUMA node, addr: 0x518c0840, len: 13824: Operation not permitted [1]
+W0827 04:31:31.543947  2697 memory_location.cpp:73] Failed to get NUMA node, addr: 0x49303540, len: 6912: Operation not permitted [1]
+W0827 04:31:31.543990  2698 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51951a80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681525  2717 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50376e40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681525  2718 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5058ee40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681525  2716 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5008b980, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681548  2715 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fe0e840, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681727  2719 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5040eb80, len: 55296: Operation not permitted [1]
+W0827 04:31:31.681779  2724 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50209ac0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681811  2720 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5041c3c0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.681828  2722 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5048cc40, len: 13824: Operation not permitted [1]
+W0827 04:31:31.681844  2723 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fde7a00, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681851  2721 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50386580, len: 6912: Operation not permitted [1]
+W0827 04:31:31.730942  2745 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fce4240, len: 6912: Operation not permitted [1]
+W0827 04:31:31.730918  2744 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fed7d40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.730950  2743 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fdcaf80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.731007  2748 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5074c400, len: 55296: Operation not permitted [1]
+W0827 04:31:31.731073  2747 memory_location.cpp:73] Failed to get NUMA node, addr: 0x507bd9c0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.731086  2746 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4f931480, len: 6912: Operation not permitted [1]
+W0827 04:31:31.731294  2749 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd958c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.731361  2751 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd31d80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.731361  2750 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4ffabe00, len: 13824: Operation not permitted [1]
+W0827 04:31:31.731566  2752 memory_location.cpp:73] Failed to get NUMA node, addr: 0x508527c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.808167  2778 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd3ce40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.808190  2777 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd20740, len: 6912: Operation not permitted [1]
+W0827 04:31:31.808192  2776 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4ff86c80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.808637  2782 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fe88780, len: 6912: Operation not permitted [1]
+W0827 04:31:31.808724  2780 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fe48c80, len: 55296: Operation not permitted [1]
+W0827 04:31:31.808760  2781 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fe564c0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.809056  2785 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4ffa7e40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.809049  2779 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4feaa740, len: 6912: Operation not permitted [1]
+W0827 04:31:31.809075  2784 memory_location.cpp:73] Failed to get NUMA node, addr: 0x502fabc0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.809239  2783 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4ffa1f80, len: 13824: Operation not permitted [1]
+W0827 04:31:31.851475  2807 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4ff6fcc0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.851596  2808 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fc0a280, len: 6912: Operation not permitted [1]
+W0827 04:31:31.851615  2809 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd7d700, len: 6912: Operation not permitted [1]
+W0827 04:31:31.851732  2810 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd761c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.851951  2811 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fde6cc0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.851970  2812 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4feeb000, len: 55296: Operation not permitted [1]
+W0827 04:31:31.851962  2813 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fbe2680, len: 6912: Operation not permitted [1]
+W0827 04:31:31.851987  2816 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fbf07c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.852048  2815 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4ffa9940, len: 6912: Operation not permitted [1]
+W0827 04:31:31.852046  2814 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd235c0, len: 13824: Operation not permitted [1]
+[2026-08-27 04:31:32] INFO:     Started server process [1]
+[2026-08-27 04:31:32] INFO:     Waiting for application startup.
+[2026-08-27 04:31:32] Calling super().encode with {'add_special_tokens': False}
+[2026-08-27 04:31:32] Calling super().encode with {'add_special_tokens': False}
+[2026-08-27 04:31:32] INFO:     Application startup complete.
+[2026-08-27 04:31:32] INFO:     Uvicorn running on http://0.0.0.0:30000 (Press CTRL+C to quit)
+[2026-08-27 04:31:33] INFO:     127.0.0.1:60348 - "GET /model_info HTTP/1.1" 200 OK
+[2026-08-27 04:31:33] Start of pd disaggregation warmup ...
+[2026-08-27 04:31:37] INFO:     174.1.60.1:33088 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:38] INFO:     174.1.60.1:33092 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:39] INFO:     174.1.60.1:33106 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:40] INFO:     174.1.60.1:33110 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:41] INFO:     174.1.60.1:42028 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:42] INFO:     174.1.60.1:42030 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:43] INFO:     174.1.60.1:42046 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:44] INFO:     174.1.60.1:42050 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:45] INFO:     174.1.60.1:42060 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:46] INFO:     174.1.60.1:42072 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:47] INFO:     174.1.60.1:42086 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:48] INFO:     174.1.60.1:42092 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:49] INFO:     174.1.60.1:42098 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:50] INFO:     174.1.60.1:42100 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:51] INFO:     174.1.60.1:46852 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:52] INFO:     174.1.60.1:46856 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:53] INFO:     174.1.60.1:46872 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:54] INFO:     174.1.60.1:46876 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:55] INFO:     174.1.60.1:46886 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:56] INFO:     174.1.60.1:46890 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:57] INFO:     174.1.60.1:46898 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:58] INFO:     174.1.60.1:46912 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:59] INFO:     174.1.60.1:46928 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:00] INFO:     174.1.60.1:46936 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:01] INFO:     174.1.60.1:57672 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:02] INFO:     174.1.60.1:57676 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:03] INFO:     174.1.60.1:57686 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:04] INFO:     174.1.60.1:57688 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:05] INFO:     174.1.60.1:57696 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:06] INFO:     174.1.60.1:57708 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:12] INFO:     174.1.60.1:39714 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:13] INFO:     174.1.60.1:39728 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:14] INFO:     174.1.60.1:39736 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:15] INFO:     174.1.60.1:39746 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:16] INFO:     174.1.60.1:39760 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:17] INFO:     174.1.60.1:39772 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:18] INFO:     174.1.60.1:39782 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:19] INFO:     174.1.60.1:39786 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:20] INFO:     174.1.60.1:39800 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:21] INFO:     174.1.60.1:41036 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:22] INFO:     174.1.60.1:41052 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:23] INFO:     174.1.60.1:41058 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:24] INFO:     174.1.60.1:41068 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:25] INFO:     174.1.60.1:41074 - "GET /health HTTP/1.1" 503 Service Unavailable
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank1]:[W827 04:32:26.019157362 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank0]:[W827 04:32:26.019963648 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank3]:[W827 04:32:26.020028119 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank2]:[W827 04:32:26.020057305 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank6]:[W827 04:32:26.045976535 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank7]:[W827 04:32:26.046176399 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank5]:[W827 04:32:26.048033943 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank4]:[W827 04:32:26.048331224 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[2026-08-27 04:32:26] INFO:     174.1.60.1:41078 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:27] INFO:     174.1.60.1:41090 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:28] INFO:     174.1.60.1:41102 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:29] INFO:     174.1.60.1:41104 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:30] INFO:     174.1.60.1:41120 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:31] INFO:     174.1.60.1:57002 - "GET /health HTTP/1.1" 503 Service Unavailable
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank6]:[W827 04:32:31.709756552 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank7]:[W827 04:32:31.709839299 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank5]:[W827 04:32:31.709991082 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank4]:[W827 04:32:31.710123265 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[2026-08-27 04:32:32] INFO:     174.1.60.1:57016 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:33] INFO:     174.1.60.1:57030 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:34] INFO:     174.1.60.1:57046 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:35] INFO:     174.1.60.1:57056 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:36] INFO:     174.1.60.1:57066 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:37] INFO:     174.1.60.1:57070 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:38] INFO:     174.1.60.1:57072 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:39] INFO:     174.1.60.1:57082 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:40] INFO:     174.1.60.1:57098 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:41] INFO:     174.1.60.1:56606 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:50] INFO:     174.1.60.1:56620 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:51] INFO:     174.1.60.1:45658 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:52] INFO:     174.1.60.1:45672 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:53] INFO:     174.1.60.1:45674 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:54] INFO:     174.1.60.1:45690 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:55] INFO:     174.1.60.1:45704 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:56] INFO:     174.1.60.1:45708 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:57] INFO:     174.1.60.1:45714 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:58] INFO:     174.1.60.1:45722 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:00] INFO:     174.1.60.1:45736 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:01] INFO:     174.1.60.1:45746 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:02] INFO:     174.1.60.1:33310 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:03] INFO:     174.1.60.1:33320 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:04] INFO:     174.1.60.1:33324 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:05] INFO:     174.1.60.1:33328 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:06] INFO:     174.1.60.1:33336 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:07] INFO:     174.1.60.1:33350 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:08] INFO:     174.1.60.1:33354 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:09] INFO:     174.1.60.1:33356 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:10] INFO:     174.1.60.1:33368 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:11] INFO:     174.1.60.1:39318 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:12] INFO:     174.1.60.1:39328 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:13] INFO:     174.1.60.1:39336 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:14] INFO:     174.1.60.1:39344 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:15] INFO:     174.1.60.1:39346 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:16] INFO:     174.1.60.1:39352 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:17] INFO:     174.1.60.1:39364 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:18] INFO:     174.1.60.1:39378 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:19] INFO:     174.1.60.1:39390 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:20] INFO:     174.1.60.1:39402 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:25] INFO:     174.1.60.1:37792 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:26] INFO:     174.1.60.1:37804 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:27] INFO:     174.1.60.1:37812 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:28] INFO:     174.1.60.1:37826 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:29] INFO:     174.1.60.1:37830 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:30] INFO:     174.1.60.1:37844 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:31] INFO:     174.1.60.1:44208 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:32] INFO:     174.1.60.1:44220 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:33] INFO:     174.1.60.1:44226 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:34] INFO:     174.1.60.1:44228 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:35] INFO:     174.1.60.1:44238 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:37] INFO:     174.1.60.1:44248 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:38] INFO:     174.1.60.1:44258 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:39] INFO:     174.1.60.1:44272 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:40] INFO:     174.1.60.1:44276 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:41] INFO:     174.1.60.1:44194 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:42] INFO:     174.1.60.1:44198 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:43] INFO:     174.1.60.1:44212 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:44] INFO:     174.1.60.1:44216 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:45] INFO:     174.1.60.1:44222 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:46] INFO:     174.1.60.1:44236 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:47] INFO:     174.1.60.1:44246 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:48] INFO:     174.1.60.1:44252 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:49] INFO:     174.1.60.1:44264 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:50] INFO:     174.1.60.1:44278 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:51] INFO:     174.1.60.1:35728 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:52] INFO:     174.1.60.1:35736 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:53] INFO:     174.1.60.1:35752 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:54] INFO:     174.1.60.1:35754 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:55] INFO:     174.1.60.1:35758 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:04] INFO:     174.1.60.1:44514 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:05] INFO:     174.1.60.1:44518 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:06] INFO:     174.1.60.1:44520 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:07] INFO:     174.1.60.1:44526 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:08] INFO:     174.1.60.1:44530 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:09] INFO:     174.1.60.1:44546 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:10] INFO:     174.1.60.1:44550 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:11] INFO:     174.1.60.1:45098 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:12] INFO:     127.0.0.1:59722 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:12] INFO:     174.1.60.1:45106 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:13] INFO:     174.1.60.1:45122 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:14] INFO:     174.1.60.1:45124 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:15] INFO:     174.1.60.1:45136 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:16] INFO:     174.1.60.1:45146 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:17] INFO:     174.1.60.1:45152 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:19] INFO:     174.1.60.1:45158 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:20] INFO:     174.1.60.1:45170 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:21] INFO:     174.1.60.1:45182 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:22] INFO:     174.1.60.1:39946 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:23] INFO:     174.1.60.1:39958 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:24] INFO:     174.1.60.1:39962 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:25] INFO:     174.1.60.1:39978 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:26] INFO:     174.1.60.1:39990 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:27] INFO:     174.1.60.1:39994 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:28] INFO:     174.1.60.1:39996 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:29] INFO:     174.1.60.1:40002 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:30] INFO:     174.1.60.1:40012 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:31] INFO:     174.1.60.1:46978 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:32] INFO:     174.1.60.1:46988 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:33] INFO:     174.1.60.1:46996 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:34] INFO:     174.1.60.1:47004 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:39] INFO:     174.1.60.1:47020 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:40] INFO:     174.1.60.1:47022 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:41] INFO:     174.1.60.1:59158 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:42] INFO:     174.1.60.1:59170 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:43] INFO:     174.1.60.1:59172 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:44] INFO:     174.1.60.1:59180 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:45] INFO:     174.1.60.1:59184 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:46] INFO:     174.1.60.1:59186 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:47] INFO:     174.1.60.1:59200 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:48] INFO:     174.1.60.1:59202 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:49] INFO:     174.1.60.1:59214 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:50] INFO:     174.1.60.1:59220 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:52] INFO:     174.1.60.1:40566 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:53] INFO:     174.1.60.1:40576 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:54] INFO:     174.1.60.1:40588 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:55] INFO:     174.1.60.1:40596 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:56] INFO:     174.1.60.1:40598 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:57] INFO:     174.1.60.1:40604 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:58] INFO:     174.1.60.1:40608 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:59] INFO:     174.1.60.1:40610 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:00] INFO:     174.1.60.1:40622 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:01] INFO:     174.1.60.1:41102 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:02] INFO:     174.1.60.1:41108 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:03] INFO:     174.1.60.1:41122 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:04] INFO:     174.1.60.1:41124 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:05] INFO:     174.1.60.1:41134 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:06] INFO:     174.1.60.1:41140 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:07] INFO:     174.1.60.1:41142 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:08] INFO:     174.1.60.1:41158 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:09] INFO:     174.1.60.1:41160 - "GET /health HTTP/1.1" 503 Service Unavailable
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/p_starting_p_node1.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/p_starting_p_node1.log
new file mode 100644
index 0000000..19fb811
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/p_starting_p_node1.log
@@ -0,0 +1,741 @@
+/usr/local/lib/python3.12/dist-packages/torchao/quantization/quant_api.py:1731: SyntaxWarning: invalid escape sequence '\.'
+  """Configuration class for applying different quantization configs to modules or parameters based on their fully qualified names (FQNs).
+/sgl-workspace/sglang/python/sglang/launch_server.py:56: UserWarning: 'python -m sglang.launch_server' is still supported, but 'sglang serve' is the recommended entrypoint.
+  Example: sglang serve --model-path <model> [options]
+  warnings.warn(
+[2026-08-27 04:28:43] Breakable CUDA graph is incompatible with MLA attention; disabling prefill CUDA graph.
+[2026-08-27 04:28:43] Attention backend not specified. Use triton backend by default.
+[2026-08-27 04:28:43] Pipeline parallelism is incompatible with overlap schedule.
+[2026-08-27 04:28:46] server_args=ServerArgs(model_path='/data/hf_models/Kimi-K3', tokenizer_path='/data/hf_models/Kimi-K3', tokenizer_mode='auto', tokenizer_backend='huggingface', tokenizer_worker_num=1, detokenizer_worker_num=1, skip_tokenizer_init=False, load_format='auto', model_loader_extra_config='{}', trust_remote_code=True, context_length=None, is_embedding=False, enable_multimodal=None, revision=None, model_impl='auto', model_config_parser='auto', json_model_override_args='{}', dtype='auto', quantization=None, quantization_param_path=None, kv_cache_dtype='auto', enable_fp32_lm_head=False, modelopt_quant=None, modelopt_checkpoint_restore_path=None, modelopt_checkpoint_save_path=None, modelopt_export_path=None, quantize_and_serve=False, rl_quant_profile=None, enable_tf32_matmul=False, mem_fraction_static=0.88, max_running_requests=None, max_queued_requests=None, max_total_tokens=None, chunked_prefill_size=8192, enable_dynamic_chunking=False, max_prefill_tokens=16384, prefill_max_requests=None, schedule_policy='fcfs', enable_priority_scheduling=False, disable_priority_preemption=False, default_priority_value=None, abort_on_priority_when_disabled=False, schedule_low_priority_values_first=False, priority_scheduling_preemption_threshold=10, retraction_policy='length', schedule_conservativeness=1.0, page_size=64, swa_full_tokens_ratio=0.8, disable_hybrid_swa_memory=False, radix_eviction_policy='lru', prefill_only_disable_kv_cache=False, disable_radix_cache=True, enable_page_major_kv_layout=False, enable_unified_memory=False, disable_chunked_prefix_cache=False, disable_overlap_schedule=True, num_continuous_decode_steps=1, scheduler_recv_interval=1, enable_mixed_chunk=False, nccl_port=None, dist_timeout=3600, dist_init_addr='174.1.60.1:20000', nnodes=4, node_rank=1, tp_size=4, dcp_size=1, pp_size=8, pp_max_micro_batch_size=None, pp_async_batch_depth=0, dp_size=1, load_balance_method='follow_bootstrap_room', attn_cp_size=1, moe_dp_size=1, dcp_comm_backend='ag_rs', dcp_replicate_q_proj=None, dwdp_size=1, enable_prefill_cp=False, cp_strategy=None, enable_dsa_cache_layer_split=False, enable_dsa_prefill_context_parallel=False, dsa_prefill_cp_mode='round-robin-split', enable_prefill_context_parallel=False, prefill_cp_mode='in-seq-split', enable_cp_decode_attn_tp=False, enable_dp_attention=False, enable_dp_attention_local_control_broadcast=False, enable_dp_lm_head=False, enable_attn_tp_input_scattered=False, disable_attn_tp_gather=False, enable_p2p_check=False, device='cuda', base_gpu_id=0, gpu_id_step=1, random_seed=577755302, watchdog_timeout=300, soft_watchdog_timeout=None, sleep_on_idle=False, use_ray=False, custom_sigquit_handler=None, numa_node=None, gc_threshold=None, host='0.0.0.0', port=30000, fastapi_root_path='', smg_grpc_mode=False, grpc_mode=False, grpc_port=None, sidecar=None, sidecar_args=None, skip_server_warmup=False, warmups=None, enable_http2=False, ssl_keyfile=None, ssl_certfile=None, ssl_ca_certs=None, ssl_keyfile_password=None, enable_ssl_refresh=False, api_key=None, admin_api_key=None, served_model_name='kimi-k3', weight_version='default', chat_template=None, hf_chat_template_name=None, completion_template=None, file_storage_path='sglang_storage', enable_cache_report=False, reasoning_parser=None, default_chat_template_kwargs=None, strip_thinking_cache=False, enable_strict_thinking=False, tool_call_parser=None, tool_server=None, sampling_defaults='model', asr_max_buffer_seconds=60, asr_max_concurrent_sessions=32, preferred_sampling_params=None, allow_auto_truncate=False, stream_interval=1, batch_notify_size=16, stream_response_default_include_usage=False, incremental_streaming_output=False, enable_streaming_session=False, enable_session_radix_cache=False, log_level='info', log_level_http=None, log_requests=False, log_requests_level=2, log_requests_format='text', log_requests_target=None, uvicorn_access_log_exclude_prefixes=[], crash_dump_folder=None, show_time_cost=False, enable_metrics=False, smg_http_sidecar_port=None, enable_mfu_metrics=False, enable_metrics_for_all_schedulers=False, load_snapshot_publish_interval=15, tokenizer_metrics_custom_labels_header='x-custom-labels', tokenizer_metrics_allowed_custom_labels=None, extra_metric_labels=None, bucket_time_to_first_token=None, bucket_inter_token_latency=None, bucket_e2e_request_latency=None, prompt_tokens_buckets=None, generation_tokens_buckets=None, gc_warning_threshold_secs=0.0, decode_log_interval=40, enable_request_time_stats_logging=False, kv_events_config=None, enable_forward_pass_metrics=False, forward_pass_metrics_worker_id='', forward_pass_metrics_ipc_name=None, enable_trace=False, trace_modules='request', otlp_traces_endpoint='localhost:4317', export_metrics_to_file=False, export_metrics_to_file_dir=None, stat_loggers=None, constrained_json_whitespace_pattern=None, constrained_json_disable_any_whitespace=False, attention_backend='triton', decode_attention_backend=None, prefill_attention_backend=None, sampling_backend='flashinfer', grammar_backend='xgrammar', radix_cache_backend=None, mm_attention_backend=None, fp8_gemm_runner_backend='auto', fp4_gemm_runner_backend='auto', bf16_gemm_backend='auto', dsa_prefill_backend=None, dsa_decode_backend=None, dsa_paged_mqa_logits_backend='auto', dsa_topk_backend='sgl-kernel', disable_flashinfer_autotune=False, flashinfer_autotune_skip_ops=None, mamba_backend='triton', cuda_graph_config=CudaGraphConfig(decode=PhaseConfig(backend='disabled', max_bs=16, bs=[1, 2, 4, 8, 12, 16], tc_compiler='eager', full_prefill_max_req=None), prefill=PhaseConfig(backend='disabled', max_bs=2048, bs=[4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], tc_compiler='eager', full_prefill_max_req=None)), cuda_graph_backend_decode=None, cuda_graph_backend_prefill=None, cuda_graph_max_bs_decode=16, cuda_graph_max_bs_prefill=None, cuda_graph_bs_decode=None, cuda_graph_bs_prefill=None, cuda_graph_tc_compiler=None, disable_prefill_cuda_graph=False, disable_decode_cuda_graph=False, disable_cuda_graph=False, disable_cuda_graph_padding=False, enable_profile_cuda_graph=False, enable_cudagraph_gc=False, debug_cuda_graph=False, enable_layerwise_nvtx_marker=False, enable_nccl_nvls=False, enable_symm_mem=False, triton_attention_reduce_in_fp32=False, triton_attention_num_kv_splits=8, triton_attention_split_tile_size=None, flashinfer_mla_disable_ragged=False, enable_fused_qk_norm_rope=False, enable_precise_embedding_interpolation=False, enable_fused_moe_sum_all_reduce=False, enable_deepseek_v4_fp4_indexer=False, disable_custom_all_reduce=False, enable_mscclpp=False, enable_torch_symm_mem=False, enable_scattered_sconv=False, pre_warm_nccl=False, enable_quant_communications=False, enable_flashinfer_allreduce_fusion=False, enforce_disable_flashinfer_allreduce_fusion=False, flashinfer_allreduce_fusion_backend=None, enable_aiter_allreduce_fusion=False, enable_torch_compile=False, enable_torch_compile_debug_mode=False, torch_compile_max_bs=32, torchao_config='', speculative_algorithm=None, speculative_draft_model_path=None, speculative_draft_model_revision=None, speculative_draft_load_format=None, speculative_num_steps=None, speculative_eagle_topk=None, speculative_num_draft_tokens=None, speculative_dflash_block_size=None, speculative_dspark_block_size=None, speculative_dspark_sps_table_path=None, speculative_dspark_confidence_sts_path=None, speculative_dspark_align_verify_tokens_to_graph_tier=False, speculative_accept_threshold_single=1.0, speculative_accept_threshold_acc=1.0, speculative_use_rejection_sampling=False, speculative_token_map=None, speculative_attention_mode='prefill', speculative_draft_attention_backend=None, speculative_draft_window_size=None, speculative_moe_runner_backend='flashinfer_mxfp4', speculative_moe_a2a_backend=None, speculative_draft_model_quantization=None, speculative_skip_dp_mlp_sync=False, enable_multi_layer_eagle=False, speculative_adaptive=False, speculative_adaptive_config=None, decoupled_spec_bind_endpoint=None, decoupled_spec_connect_endpoints=None, decoupled_spec_rank=None, decoupled_spec_role='null', spec_trace_dir=None, speculative_ngram_min_bfs_breadth=1, speculative_ngram_max_bfs_breadth=10, speculative_ngram_match_type='BFS', speculative_ngram_max_trie_depth=18, speculative_ngram_capacity=10000000, speculative_ngram_external_corpus_path=None, speculative_ngram_external_sam_budget=0, speculative_ngram_external_corpus_max_tokens=10000000, ep_size=4, moe_a2a_backend='none', moe_runner_backend='flashinfer_mxfp4', flashinfer_mxfp4_moe_precision='default', deepep_mode='auto', fuseep_mode=2, deepep_dispatcher_output_dtype='auto', ep_num_redundant_experts=0, ep_dispatch_algorithm=None, init_expert_location='trivial', enable_eplb=False, eplb_algorithm='auto', eplb_rebalance_num_iterations=1000, eplb_rebalance_layers_per_chunk=None, eplb_min_rebalancing_utilization_threshold=1.0, expert_distribution_recorder_mode=None, expert_distribution_recorder_buffer_size=1000, enable_expert_distribution_metrics=False, deepep_config=None, moe_dense_tp_size=None, elastic_ep_backend=None, enable_elastic_expert_backup=False, mooncake_ib_device=None, enable_waterfill=False, ep_join_mode=None, ep_join_rank_offset=0, elastic_ep_initial_size=None, max_ep_size=None, elastic_ep_scale_timeout=600, elastic_ep_rejoin=False, disable_flashinfer_cutlass_moe_fp4_allgather=False, disable_shared_experts_fusion=False, enforce_shared_experts_fusion=False, max_mamba_cache_size=None, mamba_ssm_dtype=None, mamba_max_states_per_path=-1, enable_mamba_cache_stochastic_rounding=False, mamba_cache_philox_rounds=0, mamba_full_memory_ratio=0.36, mamba_radix_cache_strategy='extra_buffer_lazy', uses_mamba_radix_cache=False, mamba_track_interval=256, enable_int8_mamba_checkpoint=False, int8_mamba_ckpt_size=None, linear_attn_backend='triton', linear_attn_decode_backend=None, linear_attn_prefill_backend=None, linear_attn_verify_backend=None, enable_linear_replayssm=False, linear_replayssm_cache_len=16, enable_linear_replayssm_spec=False, enable_hierarchical_cache=False, hicache_ratio=2.0, hicache_size=0, hicache_write_policy='write_through', hicache_io_backend='kernel', hicache_mem_layout='page_first', hicache_storage_backend=None, hicache_storage_prefetch_policy='timeout', hicache_storage_backend_extra_config=None, enable_hisparse=False, hisparse_config=None, enable_broadcast_mm_inputs_process=False, enable_prefix_mm_cache=False, mm_enable_dp_encoder=False, mm_process_config={}, mm_processor_worker_num=0, mm_io_worker_num=0, limit_mm_data_per_request=None, enable_mm_global_cache=False, disable_fast_image_processor=False, mm_feature_transport='cpu', keep_mm_feature_on_device=False, enable_lora=None, enable_lora_overlap_loading=None, max_lora_rank=None, lora_target_modules=None, lora_paths=None, max_loaded_loras=None, max_loras_per_batch=8, lora_eviction_policy='lru', lora_backend='csgmv', max_lora_chunk_size=16, experts_shared_outer_loras=None, lora_use_virtual_experts=False, lora_strict_loading=False, lora_drain_wait_threshold=0.0, enable_two_batch_overlap=False, enable_single_batch_overlap=False, tbo_token_distribution_threshold=0.48, cpu_offload_gb=0, offload_group_size=-1, offload_num_in_group=1, offload_prefetch_step=1, offload_mode='cpu', enable_lmcache=False, lmcache_config_file=None, enable_flexkv=False, flexkv_config_file=None, kt_weight_path=None, kt_method='AMXINT4', kt_cpuinfer=None, kt_threadpool_count=2, kt_num_gpu_experts=None, kt_max_deferred_experts_per_token=None, dllm_algorithm=None, dllm_algorithm_config=None, dllm_fdfo=True, disaggregation_mode='prefill', disaggregation_transfer_backend='mooncake', disaggregation_bootstrap_port=28800, disaggregation_ib_device='mlx5_0,mlx5_1,mlx5_2,mlx5_3', disaggregation_decode_enable_radix_cache=False, disaggregation_decode_enable_offload_kvcache=False, num_reserved_decode_tokens=512, disaggregation_decode_extra_slots=None, disaggregation_decode_polling_interval=1, optimistic_prefill_attempts=0, encoder_only=False, language_only=False, encoder_transfer_backend='zmq_to_tokenizer', encoder_urls=[], encoder_bootstrap_port=8997, encoder_register_urls=[], enable_adaptive_dispatch_to_encoder=False, enable_pdmux=False, pdmux_config_path=None, sm_group_num=8, custom_weight_loader=[], weight_loader_disable_mmap=False, weight_loader_prefetch_checkpoints=False, weight_loader_prefetch_num_threads=4, weight_loader_drop_cache_after_load=False, remote_instance_weight_loader_seed_instance_ip=None, remote_instance_weight_loader_seed_instance_service_port=None, remote_instance_weight_loader_send_weights_group_ports=None, remote_instance_weight_loader_backend='nccl', remote_instance_weight_loader_start_seed_via_transfer_engine=False, engine_info_bootstrap_port=6789, modelexpress_config=None, download_dir=None, model_checksum=None, delete_ckpt_after_loading=False, decrypted_config_file=None, decrypted_draft_config_file=None, checkpoint_engine_wait_weights_before_ready=False, enable_prefill_delayer=False, prefill_delayer_max_delay_passes=30, prefill_delayer_token_usage_low_watermark=None, prefill_delayer_forward_passes_buckets=None, prefill_delayer_wait_seconds_buckets=None, prefill_delayer_queue_min_ratio=None, prefill_delayer_max_delay_ms=None, min_free_slots_delay=None, enable_deterministic_inference=False, rl_on_policy_target=None, kv_canary='none', kv_canary_real_data='none', kv_canary_sweep_interval=0, enable_dynamic_batch_tokenizer=False, dynamic_batch_tokenizer_batch_size=32, dynamic_batch_tokenizer_batch_timeout=0.002, enable_tokenizer_batch_encode=False, disable_tokenizer_batch_decode=False, debug_tensor_dump_output_folder=None, debug_tensor_dump_layers=None, debug_tensor_dump_input_file=None, enable_memory_saver=False, enable_weights_cpu_backup=False, enable_draft_weights_cpu_backup=False, enable_custom_logit_processor=False, enable_return_hidden_states=False, enable_return_routed_experts=False, enable_return_indexer_topk=False, disable_outlines_disk_cache=False, enable_mis=False, weight_cache_mode='off', weight_cache_socket=None, weight_cache_timeout=1800, forward_hooks=None, msprobe_dump_config=None)
+[2026-08-27 04:28:46] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:46] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:46] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:46] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:46] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:46] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:46] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:46] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:02 PP3 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:02 PP3 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:03 PP3 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:03 PP3 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:03 PP3 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:03 PP3 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:03 PP2 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:03 PP2 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:03 PP2 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:03 PP2 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:04 PP2 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:04 PP2 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:04 PP2 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:04 PP2 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:04 PP3 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:04 PP3 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:05 PP3 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:05 PP2 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:05 PP3 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:05 PP3 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:05 PP2 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:06.044592   241 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:06.044694   241 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:06.044710   241 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.2 port: 12001
+I0827 04:29:06.044759   241 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.2:16343
+I0827 04:29:06.044883   241 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:06.053642   242 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:06.053738   242 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:06.053754   242 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.2 port: 12001
+I0827 04:29:06.053797   242 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.2:15170
+I0827 04:29:06.053925   242 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:06.058039   237 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:06.058135   237 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:06.058149   237 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.2 port: 12001
+I0827 04:29:06.058192   237 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.2:15486
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:06.058215   236 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:06.058310   236 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:06.058316   237 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:06.058326   236 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.2 port: 12001
+I0827 04:29:06.058367   236 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.2:16140
+I0827 04:29:06.058492   236 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:06.059734   239 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:06.059828   239 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:06.059844   239 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.2 port: 12001
+I0827 04:29:06.059890   239 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.2:15755
+I0827 04:29:06.060027   239 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:06.064433   237 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:06.064513   236 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:06.065588   239 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:06.069945   237 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:06.070852   236 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:06.071113   239 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:06.071978   241 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:06.079348   236 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:06.079967   242 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:06.080070   241 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:06.087044   241 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:06.087074   242 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:06.094024   242 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:06.095021   241 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:06.095822   237 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:06.097810   241 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:06.097839   241 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:06.097889   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.099939   239 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:06.100984   242 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:06.103683   242 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:06.103708   242 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:06.103756   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.103785   236 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:06.103833   237 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:06.106014   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:06.106982   236 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:06.107007   236 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:06.107053   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.107446   241 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:02
+I0827 04:29:06.107475   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.107936   239 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:06.107971   237 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:06.107997   237 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:06.108048   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.110760   239 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:06.110790   239 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:06.110847   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.111824   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:06.113214   242 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:02
+I0827 04:29:06.113242   242 rdma_context.cpp:205] Using SIEVE endpoint store
+[2026-08-27 04:29:06 PP2 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+I0827 04:29:06.114821   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:06.115196   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+[2026-08-27 04:29:06 PP2 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+I0827 04:29:06.116189   241 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:02
+I0827 04:29:06.116214   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.116621   237 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:02
+I0827 04:29:06.116650   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.118325   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:06.119874   239 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:02
+I0827 04:29:06.119906   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.121601   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:06.122952   242 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:02
+I0827 04:29:06.122977   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.127393   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:06.128841   239 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:02
+I0827 04:29:06.128867   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.132040   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:06.133428   236 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:02
+I0827 04:29:06.133458   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.136312   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:06.137817   239 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:02
+I0827 04:29:06.137846   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.144285   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:06.144373   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:06.145742   237 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:02
+I0827 04:29:06.145776   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.145833   241 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:02
+I0827 04:29:06.145866   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.152084   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:06.153820   242 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:02
+I0827 04:29:06.153860   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.154352   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:06.155728   241 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:02
+I0827 04:29:06.155835   241 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:06 PP3 TP2 EP2] Init torch distributed begin.
+I0827 04:29:06.163965   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:06.164443   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:06.164605   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:06.165319   236 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:02
+I0827 04:29:06.165345   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.165951   239 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:02
+I0827 04:29:06.166070   239 transfer_engine_impl.cpp:388] installTransport, type=rdma
+I0827 04:29:06.166186   242 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:02
+I0827 04:29:06.166311   242 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:06 PP3 TP0 EP0] Init torch distributed begin.
+[2026-08-27 04:29:06 PP3 TP3 EP3] Init torch distributed begin.
+I0827 04:29:06.172065   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:06.173699   237 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:02
+I0827 04:29:06.173743   237 rdma_context.cpp:205] Using SIEVE endpoint store
+[2026-08-27 04:29:06 PP3 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+I0827 04:29:06.192251   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:06.193763   236 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:02
+I0827 04:29:06.193797   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.200259   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:06.201810   237 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:02
+I0827 04:29:06.201937   237 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:06 PP2 TP2 EP2] Init torch distributed begin.
+I0827 04:29:06.220070   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:06.221505   236 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:02
+I0827 04:29:06.221653   236 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:06 PP2 TP1 EP1] Init torch distributed begin.
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:06.613179   235 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:06.613286   235 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:06.613301   235 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.2 port: 12001
+I0827 04:29:06.613344   235 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.2:16382
+I0827 04:29:06.613492   235 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:06.618471   235 topology.cpp:127] Device mlx5_0 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:06.618719   238 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:06.618824   238 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:06.618839   238 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.2 port: 12001
+I0827 04:29:06.618882   238 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.2:16170
+I0827 04:29:06.619020   238 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:06.624203   238 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:06.629066   238 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:06.633975   238 topology.cpp:127] Device mlx5_2 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:06.636366   240 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:06.636454   240 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:06.636466   240 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.2 port: 12001
+I0827 04:29:06.636509   240 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.2:15840
+I0827 04:29:06.636675   240 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:06.638876   238 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:06.640780   238 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:06.640802   238 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:06.640842   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.641396   240 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:06.643196   235 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:06.645701   240 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:06.649896   235 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:06.650310   240 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:06.654819   240 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:06.656375   240 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:06.656396   240 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:06.656432   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.657156   235 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:06.659863   235 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:06.659895   235 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:06.659950   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.661671   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:06.662956   240 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:02
+I0827 04:29:06.662979   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.664074   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:06.665762   238 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:02
+I0827 04:29:06.665796   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.667351   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:06.668808   235 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:02
+I0827 04:29:06.668838   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.676294   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:06.677721   235 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:02
+I0827 04:29:06.677748   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.692129   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:06.693621   240 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:02
+I0827 04:29:06.693648   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.696089   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:06.697444   238 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:02
+I0827 04:29:06.697470   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.707720   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:06.709362   235 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:02
+I0827 04:29:06.709394   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.719987   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:06.721463   240 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:02
+I0827 04:29:06.721489   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.728108   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:06.728855   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:06.729467   238 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:02
+I0827 04:29:06.729491   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.730392   240 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:02
+I0827 04:29:06.730533   240 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:06 PP3 TP1 EP1] Init torch distributed begin.
+I0827 04:29:06.736888   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:06.738317   238 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:02
+I0827 04:29:06.738422   238 transfer_engine_impl.cpp:388] installTransport, type=rdma
+I0827 04:29:06.738466   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+[2026-08-27 04:29:06 PP2 TP3 EP3] Init torch distributed begin.
+I0827 04:29:06.739869   235 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:02
+I0827 04:29:06.739982   235 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:06 PP2 TP0 EP0] Init torch distributed begin.
+[2026-08-27 04:29:26 PP3 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:26 PP2 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:27 PP2 TP3 EP3] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP2 TP1 EP1] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP2 TP2 EP2] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP2 TP0 EP0] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP3 TP3 EP3] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP3 TP0 EP0] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP3 TP2 EP2] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP3 TP1 EP1] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP3 TP3 EP3] Init torch distributed ends. elapsed=21.61 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP3 TP2 EP2] Init torch distributed ends. elapsed=21.62 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP3 TP1 EP1] Init torch distributed ends. elapsed=21.04 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP3 TP0 EP0] Init torch distributed ends. elapsed=21.61 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP2 TP2 EP2] Init torch distributed ends. elapsed=21.57 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP2 TP1 EP1] Init torch distributed ends. elapsed=21.55 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP2 TP0 EP0] Init torch distributed ends. elapsed=21.03 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP2 TP3 EP3] Init torch distributed ends. elapsed=21.04 s, mem usage=0.54 GB
+[2026-08-27 04:29:28 PP3 TP3 EP3] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP3 TP0 EP0] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP2 TP0 EP0] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP3 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP2 TP1 EP1] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP3 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP2 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP3 TP2 EP2] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP2 TP3 EP3] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP2 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP2 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP3 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP2 TP2 EP2] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP3 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP3 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP2 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP3 TP1 EP1] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP2 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP3 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP2 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP3 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP3 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP2 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP3 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP2 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP3 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP2 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP2 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP2 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP2 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP3 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP3 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:30:31 PP2 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:31 PP2 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:31 PP2 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:31 PP2 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:31 PP2 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:31 PP2 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:31 PP2 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:31 PP2 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP2 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:32 PP2 TP0 EP0] Load weight end. elapsed=64.92 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=36.14 GB, mem usage=45.91 GB.
+[2026-08-27 04:30:33 PP3 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:33 PP3 TP1 EP1] Load weight end. elapsed=65.31 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:34 PP2 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:34 PP2 TP1 EP1] Load weight end. elapsed=66.07 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=36.14 GB, mem usage=45.91 GB.
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:35 PP3 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:35 PP3 TP0 EP0] Load weight end. elapsed=67.41 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:36 PP2 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:36 PP2 TP3 EP3] Load weight end. elapsed=68.04 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=36.14 GB, mem usage=45.91 GB.
+[2026-08-27 04:30:36 PP2 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:36 PP2 TP2 EP2] Load weight end. elapsed=68.23 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=36.14 GB, mem usage=45.91 GB.
+[2026-08-27 04:30:36 PP2 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:36 PP2 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:36 PP2 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:36 PP2 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:36 PP2 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:36 PP2 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:36 PP2 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:36 PP2 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:36 PP3 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:36 PP3 TP3 EP3] Load weight end. elapsed=68.86 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:37 PP3 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:37 PP3 TP2 EP2] Load weight end. elapsed=69.68 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:38 PP3 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:38 PP3 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:38 PP3 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:38 PP3 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:38 PP3 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:38 PP3 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:38 PP3 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:38 PP3 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:54 PP3 TP3 EP3] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP3 TP2 EP2] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP3 TP1 EP1] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP2 TP3 EP3] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP3 TP0 EP0] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP2 TP2 EP2] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP2 TP1 EP1] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP2 TP0 EP0] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP2 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:54 PP2 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:54 PP3 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP3 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP2 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:54 PP3 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP3 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP2 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:54 PP3 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP3 TP3 EP3] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP2 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP2 TP2 EP2] Memory pool end. avail mem=23.25 GB
+[2026-08-27 04:30:54 PP2 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP2 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP2 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP3 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP3 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP2 TP3 EP3] Memory pool end. avail mem=23.25 GB
+[2026-08-27 04:30:54 PP2 TP1 EP1] Memory pool end. avail mem=23.25 GB
+[2026-08-27 04:30:54 PP3 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP3 TP1 EP1] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP3 TP2 EP2] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP2 TP0 EP0] Memory pool end. avail mem=23.25 GB
+[2026-08-27 04:30:54 PP3 TP0 EP0] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP3 TP3 EP3] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp3_pp3_dp0.json
+[2026-08-27 04:30:54 PP2 TP1 EP1] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp1_pp2_dp0.json
+[2026-08-27 04:30:54 PP2 TP2 EP2] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp2_pp2_dp0.json
+[2026-08-27 04:30:54 PP3 TP1 EP1] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp1_pp3_dp0.json
+[2026-08-27 04:30:54 PP2 TP3 EP3] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp3_pp2_dp0.json
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO[TensorRT-LLM][INFO] Set logger level to INFO
+
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[2026-08-27 04:30:54 PP3 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:30:54 PP3 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:30:54 PP3 TP2 EP2] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp2_pp3_dp0.json
+[2026-08-27 04:30:54 PP2 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:30:54 PP2 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:30:54 PP3 TP0 EP0] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp0_pp3_dp0.json
+[2026-08-27 04:30:54 PP2 TP0 EP0] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp0_pp2_dp0.json
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.65profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.60profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.79profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.51profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.86profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.27profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.62profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.98profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.86profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 12.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 13.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.93profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.92profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.09profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.63profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  6.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  7.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.33profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.58profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  8.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.56profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.23profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.63profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.36profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.09profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.97profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  8.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.30profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 13.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 13.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  8.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  8.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 11.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 11.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.23profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.48profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.23profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.53profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.27profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.59profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.62profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.97profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.60profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.46profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.12profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.68profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.77profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.62profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  6.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.87profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.30profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.40profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.83profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.94profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.59profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.66profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.29profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.23profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:01,  3.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.68profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.75profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  7.01profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.22profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.65profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.62profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.46profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.22profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  7.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:01<00:06,  1.03s/profile]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:01,  3.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.50profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.93profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.48profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.11profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.12profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.30profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.68profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  7.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 11.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  9.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 12.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 11.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 11.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 11.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.11profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.31profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.65profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.31profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.35profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.49profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:01<00:01,  2.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.65profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.68profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  7.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.09profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 11.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.88profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.98profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.50profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.62profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.51profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.91profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:01,  3.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.31profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.86profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.27profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.59profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.17profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.76profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.73profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  9.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.37profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.15profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.98profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.65profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.97profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.23profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.46profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  9.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.03profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.57profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.40profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.40profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.57profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.13profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.56profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  6.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.73profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.23profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  3.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.86profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.77profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.92profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.37profile/s]
+[2026-08-27 04:31:26 PP2 TP3 EP3] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.39profile/s]
+[2026-08-27 04:31:26 PP2 TP2 EP2] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.75profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.46profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.35profile/s]
+[2026-08-27 04:31:26 PP2 TP1 EP1] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.12profile/s]
+[2026-08-27 04:31:26 PP2 TP0 EP0] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.66profile/s][2026-08-27 04:31:26 PP2 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:26 PP2 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:26 PP2 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:26 PP2 TP0 EP0] max_total_num_tokens=3801408, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=54, context_len=1048576, available_gpu_mem=20.47 GB
+[2026-08-27 04:31:26 PP2 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:26 PP2 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:26 PP2 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:26 PP2 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:26 PP2 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.93profile/s]
+[2026-08-27 04:31:27 PP3 TP1 EP1] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.16profile/s]
+[2026-08-27 04:31:27 PP3 TP0 EP0] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.43profile/s]
+[2026-08-27 04:31:27 PP3 TP2 EP2] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.05profile/s]
+[2026-08-27 04:31:27 PP3 TP3 EP3] FlashInfer autotune completed.
+[2026-08-27 04:31:27 PP3 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:27 PP3 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:27 PP3 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:27 PP3 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:27 PP3 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:27 PP3 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:27 PP3 TP0 EP0] max_total_num_tokens=3801408, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=54, context_len=1048576, available_gpu_mem=16.54 GB
+[2026-08-27 04:31:27 PP3 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:27 PP3 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+W0827 04:31:30.590376  2521 memory_location.cpp:73] Failed to get NUMA node, addr: 0x513c5ec0, len: 6912: Operation not permitted [1]
+W0827 04:31:30.590432  2522 memory_location.cpp:73] Failed to get NUMA node, addr: 0x514a0900, len: 6912: Operation not permitted [1]
+W0827 04:31:30.590495  2525 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51226dc0, len: 55296: Operation not permitted [1]
+W0827 04:31:30.590492  2523 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5128db40, len: 6912: Operation not permitted [1]
+W0827 04:31:30.590533  2524 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5156ff80, len: 6912: Operation not permitted [1]
+W0827 04:31:30.590600  2526 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51320ac0, len: 55296: Operation not permitted [1]
+W0827 04:31:30.590636  2527 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51768540, len: 6912: Operation not permitted [1]
+W0827 04:31:30.590826  2528 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51483dc0, len: 13824: Operation not permitted [1]
+W0827 04:31:30.590914  2529 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51119380, len: 6912: Operation not permitted [1]
+W0827 04:31:30.590979  2530 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50fe1e00, len: 6912: Operation not permitted [1]
+W0827 04:31:30.846120  2547 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50ceebc0, len: 6912: Operation not permitted [1]
+W0827 04:31:30.846271  2548 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50c57640, len: 6912: Operation not permitted [1]
+W0827 04:31:30.846328  2549 memory_location.cpp:73] Failed to get NUMA node, addr: 0x502ed680, len: 6912: Operation not permitted [1]
+W0827 04:31:30.846354  2550 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50d1c800, len: 6912: Operation not permitted [1]
+W0827 04:31:30.846448  2551 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50510a40, len: 55296: Operation not permitted [1]
+W0827 04:31:30.846473  2552 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5051e280, len: 55296: Operation not permitted [1]
+W0827 04:31:30.846571  2553 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50d01340, len: 6912: Operation not permitted [1]
+W0827 04:31:30.846689  2554 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50ce3440, len: 13824: Operation not permitted [1]
+W0827 04:31:30.846858  2555 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5030f440, len: 6912: Operation not permitted [1]
+W0827 04:31:30.846926  2556 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50cae980, len: 6912: Operation not permitted [1]
+W0827 04:31:31.198441  2575 memory_location.cpp:73] Failed to get NUMA node, addr: 0x514f46c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.198501  2576 memory_location.cpp:73] Failed to get NUMA node, addr: 0x509c6400, len: 6912: Operation not permitted [1]
+W0827 04:31:31.198501  2577 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51513300, len: 6912: Operation not permitted [1]
+W0827 04:31:31.198679  2578 memory_location.cpp:73] Failed to get NUMA node, addr: 0x506eef40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.198787  2579 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50dc9180, len: 55296: Operation not permitted [1]
+W0827 04:31:31.198915  2580 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50904700, len: 55296: Operation not permitted [1]
+W0827 04:31:31.198982  2581 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50aa9440, len: 6912: Operation not permitted [1]
+W0827 04:31:31.199081  2582 memory_location.cpp:73] Failed to get NUMA node, addr: 0x515012c0, len: 13824: Operation not permitted [1]
+W0827 04:31:31.199148  2583 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5161ac80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.199185  2584 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50dd69c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.763885  2603 memory_location.cpp:73] Failed to get NUMA node, addr: 0x519f1500, len: 6912: Operation not permitted [1]
+W0827 04:31:31.763998  2605 memory_location.cpp:73] Failed to get NUMA node, addr: 0x517df940, len: 6912: Operation not permitted [1]
+W0827 04:31:31.763923  2604 memory_location.cpp:73] Failed to get NUMA node, addr: 0x514ab440, len: 6912: Operation not permitted [1]
+W0827 04:31:31.764139  2606 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51556180, len: 6912: Operation not permitted [1]
+W0827 04:31:31.764169  2608 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50ed8140, len: 55296: Operation not permitted [1]
+W0827 04:31:31.764184  2607 memory_location.cpp:73] Failed to get NUMA node, addr: 0x515df040, len: 55296: Operation not permitted [1]
+W0827 04:31:31.764356  2609 memory_location.cpp:73] Failed to get NUMA node, addr: 0x517e3580, len: 6912: Operation not permitted [1]
+W0827 04:31:31.764564  2610 memory_location.cpp:73] Failed to get NUMA node, addr: 0x519cdb80, len: 13824: Operation not permitted [1]
+W0827 04:31:31.764624  2611 memory_location.cpp:73] Failed to get NUMA node, addr: 0x519dbf80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.764714  2612 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50f08d80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.868036  2629 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50a7a680, len: 6912: Operation not permitted [1]
+W0827 04:31:31.868046  2630 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50b6e400, len: 6912: Operation not permitted [1]
+W0827 04:31:31.868139  2631 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50e5f580, len: 6912: Operation not permitted [1]
+W0827 04:31:31.868261  2632 memory_location.cpp:73] Failed to get NUMA node, addr: 0x503c7100, len: 6912: Operation not permitted [1]
+W0827 04:31:31.868330  2633 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50c94ac0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.868432  2635 memory_location.cpp:73] Failed to get NUMA node, addr: 0x509b2880, len: 6912: Operation not permitted [1]
+W0827 04:31:31.868502  2634 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50b14c80, len: 55296: Operation not permitted [1]
+W0827 04:31:31.868564  2636 memory_location.cpp:73] Failed to get NUMA node, addr: 0x503ee840, len: 13824: Operation not permitted [1]
+W0827 04:31:31.868708  2637 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51057d80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.868835  2638 memory_location.cpp:73] Failed to get NUMA node, addr: 0x509cd940, len: 6912: Operation not permitted [1]
+W0827 04:31:32.003983  2656 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50715b00, len: 6912: Operation not permitted [1]
+W0827 04:31:32.004096  2657 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51346e80, len: 6912: Operation not permitted [1]
+W0827 04:31:32.004005  2655 memory_location.cpp:73] Failed to get NUMA node, addr: 0x507176c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.004179  2658 memory_location.cpp:73] Failed to get NUMA node, addr: 0x507e3a40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.004302  2659 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5078aa80, len: 55296: Operation not permitted [1]
+W0827 04:31:32.004357  2660 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5135c980, len: 55296: Operation not permitted [1]
+W0827 04:31:32.004410  2661 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5099f180, len: 6912: Operation not permitted [1]
+W0827 04:31:32.004585  2662 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50b287c0, len: 13824: Operation not permitted [1]
+W0827 04:31:32.004739  2663 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5070b6c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.004812  2664 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50753540, len: 6912: Operation not permitted [1]
+W0827 04:31:32.144867  2688 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50aed700, len: 6912: Operation not permitted [1]
+W0827 04:31:32.144946  2689 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50792740, len: 6912: Operation not permitted [1]
+W0827 04:31:32.145030  2690 memory_location.cpp:73] Failed to get NUMA node, addr: 0x508fca40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.145135  2691 memory_location.cpp:73] Failed to get NUMA node, addr: 0x505d7c40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.145318  2693 memory_location.cpp:73] Failed to get NUMA node, addr: 0x505ec340, len: 55296: Operation not permitted [1]
+W0827 04:31:32.145387  2692 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50602cc0, len: 55296: Operation not permitted [1]
+W0827 04:31:32.145479  2694 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50748c00, len: 6912: Operation not permitted [1]
+W0827 04:31:32.145632  2695 memory_location.cpp:73] Failed to get NUMA node, addr: 0x507e3480, len: 13824: Operation not permitted [1]
+W0827 04:31:32.145767  2696 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50316400, len: 6912: Operation not permitted [1]
+W0827 04:31:32.145864  2697 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50788600, len: 6912: Operation not permitted [1]
+W0827 04:31:32.177534  2726 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fff3c80, len: 6912: Operation not permitted [1]
+W0827 04:31:32.177623  2728 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fad6780, len: 6912: Operation not permitted [1]
+W0827 04:31:32.177625  2727 memory_location.cpp:73] Failed to get NUMA node, addr: 0x501dc2c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.178279  2729 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50596480, len: 6912: Operation not permitted [1]
+W0827 04:31:32.178463  2730 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50222dc0, len: 55296: Operation not permitted [1]
+W0827 04:31:32.178498  2731 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5038f6c0, len: 55296: Operation not permitted [1]
+W0827 04:31:32.178581  2732 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4ffd7ec0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.178695  2733 memory_location.cpp:73] Failed to get NUMA node, addr: 0x506d84c0, len: 13824: Operation not permitted [1]
+W0827 04:31:32.178884  2734 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5078efc0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.178922  2735 memory_location.cpp:73] Failed to get NUMA node, addr: 0x502c7b00, len: 6912: Operation not permitted [1]
+[2026-08-27 04:31:32] Dummy health check server started in background thread at 0.0.0.0:30000
+[rank10]:[W827 04:32:32.605391940 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank11]:[W827 04:32:32.605815131 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank9]:[W827 04:32:32.606041691 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank8]:[W827 04:32:32.608313727 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank11]:[W827 04:33:19.698792619 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank8]:[W827 04:33:19.699392665 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank9]:[W827 04:33:19.699501937 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank10]:[W827 04:33:19.699589651 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank14]:[W827 04:33:19.728965278 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank12]:[W827 04:33:19.729026815 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank13]:[W827 04:33:19.729092243 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank15]:[W827 04:33:19.730463091 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank14]:[W827 04:33:25.763484084 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank13]:[W827 04:33:25.764518868 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank15]:[W827 04:33:25.764776298 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank12]:[W827 04:33:25.765271843 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[2026-08-27 04:35:10 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.29
+[2026-08-27 04:35:10 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.29
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/p_starting_p_node2.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/p_starting_p_node2.log
new file mode 100644
index 0000000..a8a444b
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/p_starting_p_node2.log
@@ -0,0 +1,750 @@
+/usr/local/lib/python3.12/dist-packages/torchao/quantization/quant_api.py:1731: SyntaxWarning: invalid escape sequence '\.'
+  """Configuration class for applying different quantization configs to modules or parameters based on their fully qualified names (FQNs).
+/sgl-workspace/sglang/python/sglang/launch_server.py:56: UserWarning: 'python -m sglang.launch_server' is still supported, but 'sglang serve' is the recommended entrypoint.
+  Example: sglang serve --model-path <model> [options]
+  warnings.warn(
+[2026-08-27 04:28:43] Breakable CUDA graph is incompatible with MLA attention; disabling prefill CUDA graph.
+[2026-08-27 04:28:43] Attention backend not specified. Use triton backend by default.
+[2026-08-27 04:28:43] Pipeline parallelism is incompatible with overlap schedule.
+[2026-08-27 04:28:45] server_args=ServerArgs(model_path='/data/hf_models/Kimi-K3', tokenizer_path='/data/hf_models/Kimi-K3', tokenizer_mode='auto', tokenizer_backend='huggingface', tokenizer_worker_num=1, detokenizer_worker_num=1, skip_tokenizer_init=False, load_format='auto', model_loader_extra_config='{}', trust_remote_code=True, context_length=None, is_embedding=False, enable_multimodal=None, revision=None, model_impl='auto', model_config_parser='auto', json_model_override_args='{}', dtype='auto', quantization=None, quantization_param_path=None, kv_cache_dtype='auto', enable_fp32_lm_head=False, modelopt_quant=None, modelopt_checkpoint_restore_path=None, modelopt_checkpoint_save_path=None, modelopt_export_path=None, quantize_and_serve=False, rl_quant_profile=None, enable_tf32_matmul=False, mem_fraction_static=0.88, max_running_requests=None, max_queued_requests=None, max_total_tokens=None, chunked_prefill_size=8192, enable_dynamic_chunking=False, max_prefill_tokens=16384, prefill_max_requests=None, schedule_policy='fcfs', enable_priority_scheduling=False, disable_priority_preemption=False, default_priority_value=None, abort_on_priority_when_disabled=False, schedule_low_priority_values_first=False, priority_scheduling_preemption_threshold=10, retraction_policy='length', schedule_conservativeness=1.0, page_size=64, swa_full_tokens_ratio=0.8, disable_hybrid_swa_memory=False, radix_eviction_policy='lru', prefill_only_disable_kv_cache=False, disable_radix_cache=True, enable_page_major_kv_layout=False, enable_unified_memory=False, disable_chunked_prefix_cache=False, disable_overlap_schedule=True, num_continuous_decode_steps=1, scheduler_recv_interval=1, enable_mixed_chunk=False, nccl_port=None, dist_timeout=3600, dist_init_addr='174.1.60.1:20000', nnodes=4, node_rank=2, tp_size=4, dcp_size=1, pp_size=8, pp_max_micro_batch_size=None, pp_async_batch_depth=0, dp_size=1, load_balance_method='follow_bootstrap_room', attn_cp_size=1, moe_dp_size=1, dcp_comm_backend='ag_rs', dcp_replicate_q_proj=None, dwdp_size=1, enable_prefill_cp=False, cp_strategy=None, enable_dsa_cache_layer_split=False, enable_dsa_prefill_context_parallel=False, dsa_prefill_cp_mode='round-robin-split', enable_prefill_context_parallel=False, prefill_cp_mode='in-seq-split', enable_cp_decode_attn_tp=False, enable_dp_attention=False, enable_dp_attention_local_control_broadcast=False, enable_dp_lm_head=False, enable_attn_tp_input_scattered=False, disable_attn_tp_gather=False, enable_p2p_check=False, device='cuda', base_gpu_id=0, gpu_id_step=1, random_seed=336932344, watchdog_timeout=300, soft_watchdog_timeout=None, sleep_on_idle=False, use_ray=False, custom_sigquit_handler=None, numa_node=None, gc_threshold=None, host='0.0.0.0', port=30000, fastapi_root_path='', smg_grpc_mode=False, grpc_mode=False, grpc_port=None, sidecar=None, sidecar_args=None, skip_server_warmup=False, warmups=None, enable_http2=False, ssl_keyfile=None, ssl_certfile=None, ssl_ca_certs=None, ssl_keyfile_password=None, enable_ssl_refresh=False, api_key=None, admin_api_key=None, served_model_name='kimi-k3', weight_version='default', chat_template=None, hf_chat_template_name=None, completion_template=None, file_storage_path='sglang_storage', enable_cache_report=False, reasoning_parser=None, default_chat_template_kwargs=None, strip_thinking_cache=False, enable_strict_thinking=False, tool_call_parser=None, tool_server=None, sampling_defaults='model', asr_max_buffer_seconds=60, asr_max_concurrent_sessions=32, preferred_sampling_params=None, allow_auto_truncate=False, stream_interval=1, batch_notify_size=16, stream_response_default_include_usage=False, incremental_streaming_output=False, enable_streaming_session=False, enable_session_radix_cache=False, log_level='info', log_level_http=None, log_requests=False, log_requests_level=2, log_requests_format='text', log_requests_target=None, uvicorn_access_log_exclude_prefixes=[], crash_dump_folder=None, show_time_cost=False, enable_metrics=False, smg_http_sidecar_port=None, enable_mfu_metrics=False, enable_metrics_for_all_schedulers=False, load_snapshot_publish_interval=15, tokenizer_metrics_custom_labels_header='x-custom-labels', tokenizer_metrics_allowed_custom_labels=None, extra_metric_labels=None, bucket_time_to_first_token=None, bucket_inter_token_latency=None, bucket_e2e_request_latency=None, prompt_tokens_buckets=None, generation_tokens_buckets=None, gc_warning_threshold_secs=0.0, decode_log_interval=40, enable_request_time_stats_logging=False, kv_events_config=None, enable_forward_pass_metrics=False, forward_pass_metrics_worker_id='', forward_pass_metrics_ipc_name=None, enable_trace=False, trace_modules='request', otlp_traces_endpoint='localhost:4317', export_metrics_to_file=False, export_metrics_to_file_dir=None, stat_loggers=None, constrained_json_whitespace_pattern=None, constrained_json_disable_any_whitespace=False, attention_backend='triton', decode_attention_backend=None, prefill_attention_backend=None, sampling_backend='flashinfer', grammar_backend='xgrammar', radix_cache_backend=None, mm_attention_backend=None, fp8_gemm_runner_backend='auto', fp4_gemm_runner_backend='auto', bf16_gemm_backend='auto', dsa_prefill_backend=None, dsa_decode_backend=None, dsa_paged_mqa_logits_backend='auto', dsa_topk_backend='sgl-kernel', disable_flashinfer_autotune=False, flashinfer_autotune_skip_ops=None, mamba_backend='triton', cuda_graph_config=CudaGraphConfig(decode=PhaseConfig(backend='disabled', max_bs=16, bs=[1, 2, 4, 8, 12, 16], tc_compiler='eager', full_prefill_max_req=None), prefill=PhaseConfig(backend='disabled', max_bs=2048, bs=[4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], tc_compiler='eager', full_prefill_max_req=None)), cuda_graph_backend_decode=None, cuda_graph_backend_prefill=None, cuda_graph_max_bs_decode=16, cuda_graph_max_bs_prefill=None, cuda_graph_bs_decode=None, cuda_graph_bs_prefill=None, cuda_graph_tc_compiler=None, disable_prefill_cuda_graph=False, disable_decode_cuda_graph=False, disable_cuda_graph=False, disable_cuda_graph_padding=False, enable_profile_cuda_graph=False, enable_cudagraph_gc=False, debug_cuda_graph=False, enable_layerwise_nvtx_marker=False, enable_nccl_nvls=False, enable_symm_mem=False, triton_attention_reduce_in_fp32=False, triton_attention_num_kv_splits=8, triton_attention_split_tile_size=None, flashinfer_mla_disable_ragged=False, enable_fused_qk_norm_rope=False, enable_precise_embedding_interpolation=False, enable_fused_moe_sum_all_reduce=False, enable_deepseek_v4_fp4_indexer=False, disable_custom_all_reduce=False, enable_mscclpp=False, enable_torch_symm_mem=False, enable_scattered_sconv=False, pre_warm_nccl=False, enable_quant_communications=False, enable_flashinfer_allreduce_fusion=False, enforce_disable_flashinfer_allreduce_fusion=False, flashinfer_allreduce_fusion_backend=None, enable_aiter_allreduce_fusion=False, enable_torch_compile=False, enable_torch_compile_debug_mode=False, torch_compile_max_bs=32, torchao_config='', speculative_algorithm=None, speculative_draft_model_path=None, speculative_draft_model_revision=None, speculative_draft_load_format=None, speculative_num_steps=None, speculative_eagle_topk=None, speculative_num_draft_tokens=None, speculative_dflash_block_size=None, speculative_dspark_block_size=None, speculative_dspark_sps_table_path=None, speculative_dspark_confidence_sts_path=None, speculative_dspark_align_verify_tokens_to_graph_tier=False, speculative_accept_threshold_single=1.0, speculative_accept_threshold_acc=1.0, speculative_use_rejection_sampling=False, speculative_token_map=None, speculative_attention_mode='prefill', speculative_draft_attention_backend=None, speculative_draft_window_size=None, speculative_moe_runner_backend='flashinfer_mxfp4', speculative_moe_a2a_backend=None, speculative_draft_model_quantization=None, speculative_skip_dp_mlp_sync=False, enable_multi_layer_eagle=False, speculative_adaptive=False, speculative_adaptive_config=None, decoupled_spec_bind_endpoint=None, decoupled_spec_connect_endpoints=None, decoupled_spec_rank=None, decoupled_spec_role='null', spec_trace_dir=None, speculative_ngram_min_bfs_breadth=1, speculative_ngram_max_bfs_breadth=10, speculative_ngram_match_type='BFS', speculative_ngram_max_trie_depth=18, speculative_ngram_capacity=10000000, speculative_ngram_external_corpus_path=None, speculative_ngram_external_sam_budget=0, speculative_ngram_external_corpus_max_tokens=10000000, ep_size=4, moe_a2a_backend='none', moe_runner_backend='flashinfer_mxfp4', flashinfer_mxfp4_moe_precision='default', deepep_mode='auto', fuseep_mode=2, deepep_dispatcher_output_dtype='auto', ep_num_redundant_experts=0, ep_dispatch_algorithm=None, init_expert_location='trivial', enable_eplb=False, eplb_algorithm='auto', eplb_rebalance_num_iterations=1000, eplb_rebalance_layers_per_chunk=None, eplb_min_rebalancing_utilization_threshold=1.0, expert_distribution_recorder_mode=None, expert_distribution_recorder_buffer_size=1000, enable_expert_distribution_metrics=False, deepep_config=None, moe_dense_tp_size=None, elastic_ep_backend=None, enable_elastic_expert_backup=False, mooncake_ib_device=None, enable_waterfill=False, ep_join_mode=None, ep_join_rank_offset=0, elastic_ep_initial_size=None, max_ep_size=None, elastic_ep_scale_timeout=600, elastic_ep_rejoin=False, disable_flashinfer_cutlass_moe_fp4_allgather=False, disable_shared_experts_fusion=False, enforce_shared_experts_fusion=False, max_mamba_cache_size=None, mamba_ssm_dtype=None, mamba_max_states_per_path=-1, enable_mamba_cache_stochastic_rounding=False, mamba_cache_philox_rounds=0, mamba_full_memory_ratio=0.36, mamba_radix_cache_strategy='extra_buffer_lazy', uses_mamba_radix_cache=False, mamba_track_interval=256, enable_int8_mamba_checkpoint=False, int8_mamba_ckpt_size=None, linear_attn_backend='triton', linear_attn_decode_backend=None, linear_attn_prefill_backend=None, linear_attn_verify_backend=None, enable_linear_replayssm=False, linear_replayssm_cache_len=16, enable_linear_replayssm_spec=False, enable_hierarchical_cache=False, hicache_ratio=2.0, hicache_size=0, hicache_write_policy='write_through', hicache_io_backend='kernel', hicache_mem_layout='page_first', hicache_storage_backend=None, hicache_storage_prefetch_policy='timeout', hicache_storage_backend_extra_config=None, enable_hisparse=False, hisparse_config=None, enable_broadcast_mm_inputs_process=False, enable_prefix_mm_cache=False, mm_enable_dp_encoder=False, mm_process_config={}, mm_processor_worker_num=0, mm_io_worker_num=0, limit_mm_data_per_request=None, enable_mm_global_cache=False, disable_fast_image_processor=False, mm_feature_transport='cpu', keep_mm_feature_on_device=False, enable_lora=None, enable_lora_overlap_loading=None, max_lora_rank=None, lora_target_modules=None, lora_paths=None, max_loaded_loras=None, max_loras_per_batch=8, lora_eviction_policy='lru', lora_backend='csgmv', max_lora_chunk_size=16, experts_shared_outer_loras=None, lora_use_virtual_experts=False, lora_strict_loading=False, lora_drain_wait_threshold=0.0, enable_two_batch_overlap=False, enable_single_batch_overlap=False, tbo_token_distribution_threshold=0.48, cpu_offload_gb=0, offload_group_size=-1, offload_num_in_group=1, offload_prefetch_step=1, offload_mode='cpu', enable_lmcache=False, lmcache_config_file=None, enable_flexkv=False, flexkv_config_file=None, kt_weight_path=None, kt_method='AMXINT4', kt_cpuinfer=None, kt_threadpool_count=2, kt_num_gpu_experts=None, kt_max_deferred_experts_per_token=None, dllm_algorithm=None, dllm_algorithm_config=None, dllm_fdfo=True, disaggregation_mode='prefill', disaggregation_transfer_backend='mooncake', disaggregation_bootstrap_port=28800, disaggregation_ib_device='mlx5_0,mlx5_1,mlx5_2,mlx5_3', disaggregation_decode_enable_radix_cache=False, disaggregation_decode_enable_offload_kvcache=False, num_reserved_decode_tokens=512, disaggregation_decode_extra_slots=None, disaggregation_decode_polling_interval=1, optimistic_prefill_attempts=0, encoder_only=False, language_only=False, encoder_transfer_backend='zmq_to_tokenizer', encoder_urls=[], encoder_bootstrap_port=8997, encoder_register_urls=[], enable_adaptive_dispatch_to_encoder=False, enable_pdmux=False, pdmux_config_path=None, sm_group_num=8, custom_weight_loader=[], weight_loader_disable_mmap=False, weight_loader_prefetch_checkpoints=False, weight_loader_prefetch_num_threads=4, weight_loader_drop_cache_after_load=False, remote_instance_weight_loader_seed_instance_ip=None, remote_instance_weight_loader_seed_instance_service_port=None, remote_instance_weight_loader_send_weights_group_ports=None, remote_instance_weight_loader_backend='nccl', remote_instance_weight_loader_start_seed_via_transfer_engine=False, engine_info_bootstrap_port=6789, modelexpress_config=None, download_dir=None, model_checksum=None, delete_ckpt_after_loading=False, decrypted_config_file=None, decrypted_draft_config_file=None, checkpoint_engine_wait_weights_before_ready=False, enable_prefill_delayer=False, prefill_delayer_max_delay_passes=30, prefill_delayer_token_usage_low_watermark=None, prefill_delayer_forward_passes_buckets=None, prefill_delayer_wait_seconds_buckets=None, prefill_delayer_queue_min_ratio=None, prefill_delayer_max_delay_ms=None, min_free_slots_delay=None, enable_deterministic_inference=False, rl_on_policy_target=None, kv_canary='none', kv_canary_real_data='none', kv_canary_sweep_interval=0, enable_dynamic_batch_tokenizer=False, dynamic_batch_tokenizer_batch_size=32, dynamic_batch_tokenizer_batch_timeout=0.002, enable_tokenizer_batch_encode=False, disable_tokenizer_batch_decode=False, debug_tensor_dump_output_folder=None, debug_tensor_dump_layers=None, debug_tensor_dump_input_file=None, enable_memory_saver=False, enable_weights_cpu_backup=False, enable_draft_weights_cpu_backup=False, enable_custom_logit_processor=False, enable_return_hidden_states=False, enable_return_routed_experts=False, enable_return_indexer_topk=False, disable_outlines_disk_cache=False, enable_mis=False, weight_cache_mode='off', weight_cache_socket=None, weight_cache_timeout=1800, forward_hooks=None, msprobe_dump_config=None)
+[2026-08-27 04:28:45] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:45] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:45] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:45] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:45] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:45] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:45] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:45] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:01 PP4 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:01 PP4 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:01 PP5 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:01 PP5 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:01 PP4 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:01 PP4 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:01 PP5 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:01 PP5 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:01 PP4 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:01 PP4 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:01 PP4 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:01 PP4 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:02 PP5 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:02 PP5 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:02 PP5 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:02 PP5 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:03 PP5 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:03 PP5 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:03 PP4 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:03 PP4 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:03 PP5 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:03 PP5 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:03 PP4 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:03 PP4 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:05.195727   240 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:05.195828   240 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:05.195843   240 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.3 port: 12001
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:05.195847   236 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:05.195899   240 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.3:15652
+I0827 04:29:05.195935   236 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:05.195948   236 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.3 port: 12001
+I0827 04:29:05.195989   236 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.3:15032
+I0827 04:29:05.196039   240 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:05.196110   236 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:05.206050   241 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:05.206164   241 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:05.206182   241 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.3 port: 12001
+I0827 04:29:05.206233   241 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.3:16487
+I0827 04:29:05.206414   241 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:05.206894   238 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:05.207010   238 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:05.207029   238 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.3 port: 12001
+I0827 04:29:05.207082   238 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.3:15213
+I0827 04:29:05.207248   238 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:05.213657   238 topology.cpp:127] Device mlx5_0 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:05.215921   237 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:05.216045   237 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:05.216063   237 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.3 port: 12001
+I0827 04:29:05.216125   237 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.3:16450
+I0827 04:29:05.216308   237 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:05.219630   235 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:05.219745   235 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:05.219763   235 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.3 port: 12001
+I0827 04:29:05.219815   235 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.3:15658
+I0827 04:29:05.219810   236 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:05.220000   235 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:05.222787   237 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:05.223301   240 topology.cpp:127] Device mlx5_0 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:05.223796   239 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+W0827 04:29:05.223798   242 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:05.223948   242 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:05.223948   239 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:05.223969   242 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.3 port: 12001
+I0827 04:29:05.223973   239 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.3 port: 12001
+I0827 04:29:05.224030   242 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.3:16182
+I0827 04:29:05.224030   239 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.3:15581
+I0827 04:29:05.224237   242 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:05.224242   239 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:05.226227   235 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:05.227162   236 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:05.228888   237 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:05.229261   240 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:05.232141   235 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:05.232787   239 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:05.232861   242 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:05.233434   241 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:05.233938   236 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:05.236400   237 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:05.236641   240 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:05.238986   235 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:05.241036   236 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:05.241355   238 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:05.241638   241 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:05.242061   239 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:05.242151   242 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:05.244179   236 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:05.244223   236 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:05.244287   237 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:05.244297   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.244899   240 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:05.245494   235 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:05.249099   237 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:05.249130   237 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:05.249187   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.251125   235 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:05.251137   240 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:05.251168   235 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:05.251168   240 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:05.251241   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.251241   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.251624   239 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:05.251729   242 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:05.251823   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:05.253651   236 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:03
+I0827 04:29:05.253687   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.256857   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:05.257697   239 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:05.258617   237 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:03
+I0827 04:29:05.258652   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.258757   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:05.258822   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:05.260108   239 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:05.260136   239 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:05.260191   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.260488   235 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:03
+I0827 04:29:05.260502   240 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:03
+I0827 04:29:05.260530   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.260533   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.262187   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:05.264094   236 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:03
+I0827 04:29:05.264137   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.266639   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:05.268234   237 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:03
+I0827 04:29:05.268261   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.268549   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:05.269187   241 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:05.269232   238 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:05.270207   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:05.270285   240 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:03
+I0827 04:29:05.270316   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.271827   235 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:03
+I0827 04:29:05.271884   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.275274   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:05.276836   238 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:05.276907   242 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:05.276996   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:05.277058   236 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:03
+I0827 04:29:05.277101   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.278084   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:05.278632   237 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:03
+I0827 04:29:05.278667   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.279789   240 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:03
+I0827 04:29:05.279817   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.282152   238 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:05.282193   238 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:05.282260   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.283128   242 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:05.283200   242 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:05.283263   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.286593   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:05.287469   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:05.288283   239 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:03
+I0827 04:29:05.288322   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.289064   240 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:03
+I0827 04:29:05.289194   240 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:05 PP5 TP1 EP1] Init torch distributed begin.
+I0827 04:29:05.290841   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:05.292457   242 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:03
+I0827 04:29:05.292488   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.292999   241 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:05.296203   241 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:05.296272   241 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:05.296341   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.300206   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:05.301280   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:05.301775   242 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:03
+I0827 04:29:05.301801   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.303004   235 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:03
+I0827 04:29:05.303042   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.303999   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:05.305142   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:05.305728   241 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:03
+I0827 04:29:05.305768   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.305974   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:05.306758   237 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:03
+I0827 04:29:05.306921   237 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:05 PP4 TP2 EP2] Init torch distributed begin.
+I0827 04:29:05.307497   236 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:03
+I0827 04:29:05.307627   236 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:05 PP4 TP1 EP1] Init torch distributed begin.
+I0827 04:29:05.309624   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:05.309773   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:05.311504   238 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:03
+I0827 04:29:05.311542   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.311590   242 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:03
+I0827 04:29:05.311633   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.313076   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:05.314762   239 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:03
+I0827 04:29:05.314813   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.319522   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:05.319566   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:05.321249   242 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:03
+I0827 04:29:05.321398   242 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:05 PP5 TP3 EP3] Init torch distributed begin.
+I0827 04:29:05.326727   238 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:03
+I0827 04:29:05.326759   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.329339   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:05.331007   235 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:03
+I0827 04:29:05.331142   235 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:05 PP4 TP0 EP0] Init torch distributed begin.
+I0827 04:29:05.334666   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:05.336285   238 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:03
+I0827 04:29:05.336313   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.337296   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:05.338828   241 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:03
+I0827 04:29:05.338877   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.344906   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:05.346731   239 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:03
+I0827 04:29:05.346776   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.365299   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:05.366928   238 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:03
+I0827 04:29:05.367062   238 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:05 PP4 TP3 EP3] Init torch distributed begin.
+I0827 04:29:05.368464   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:05.369644   241 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:03
+I0827 04:29:05.369663   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.377357   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:05.379019   239 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:03
+I0827 04:29:05.379146   239 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:05 PP5 TP0 EP0] Init torch distributed begin.
+I0827 04:29:05.396831   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:05.398531   241 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:03
+I0827 04:29:05.398662   241 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:05 PP5 TP2 EP2] Init torch distributed begin.
+[2026-08-27 04:29:26 PP5 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:26 PP4 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:27 PP4 TP1 EP1] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP4 TP3 EP3] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP4 TP2 EP2] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP4 TP0 EP0] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP5 TP2 EP2] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP5 TP3 EP3] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP5 TP0 EP0] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP5 TP1 EP1] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP5 TP0 EP0] Init torch distributed ends. elapsed=22.37 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP5 TP3 EP3] Init torch distributed ends. elapsed=22.42 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP5 TP2 EP2] Init torch distributed ends. elapsed=22.35 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP4 TP1 EP1] Init torch distributed ends. elapsed=22.44 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP5 TP1 EP1] Init torch distributed ends. elapsed=22.46 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP4 TP3 EP3] Init torch distributed ends. elapsed=22.38 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP4 TP2 EP2] Init torch distributed ends. elapsed=22.44 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP4 TP0 EP0] Init torch distributed ends. elapsed=22.41 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP5 TP0 EP0] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP4 TP1 EP1] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP4 TP3 EP3] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP4 TP2 EP2] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP5 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP4 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP4 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP5 TP2 EP2] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP4 TP0 EP0] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP4 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP5 TP3 EP3] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP5 TP1 EP1] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP4 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP5 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP5 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP5 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP4 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP5 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP4 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP5 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP5 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP4 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP4 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP5 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP5 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP4 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP4 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP4 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP5 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP5 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP4 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP5 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:30:33 PP4 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:33 PP4 TP2 EP2] Load weight end. elapsed=65.50 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:34 PP4 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:34 PP4 TP3 EP3] Load weight end. elapsed=66.61 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP4 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:36 PP4 TP1 EP1] Load weight end. elapsed=68.45 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:36 PP5 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:36 PP5 TP0 EP0] Load weight end. elapsed=69.00 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:37 PP5 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:37 PP5 TP3 EP3] Load weight end. elapsed=69.63 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:37 PP5 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:37 PP5 TP1 EP1] Load weight end. elapsed=69.72 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:38 PP5 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:38 PP5 TP2 EP2] Load weight end. elapsed=70.08 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:38 PP5 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:38 PP5 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:38 PP5 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:38 PP5 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:38 PP5 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:38 PP5 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:38 PP5 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:38 PP5 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:47 PP4 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:47 PP4 TP0 EP0] Load weight end. elapsed=79.31 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:47 PP4 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:47 PP4 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:47 PP4 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:47 PP4 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:47 PP4 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:47 PP4 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:47 PP4 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:47 PP4 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:54 PP5 TP3 EP3] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP5 TP1 EP1] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP5 TP2 EP2] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP5 TP0 EP0] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP4 TP3 EP3] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP4 TP2 EP2] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP4 TP1 EP1] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP4 TP0 EP0] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP5 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP5 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP4 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP4 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP4 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP5 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP4 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP5 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP5 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP5 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP4 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP5 TP3 EP3] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP4 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP4 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP4 TP1 EP1] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP5 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP5 TP2 EP2] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP4 TP3 EP3] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP4 TP0 EP0] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP5 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP5 TP1 EP1] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP5 TP0 EP0] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP4 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP4 TP2 EP2] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP4 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:30:54 PP4 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:30:54 PP4 TP0 EP0] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp0_pp4_dp0.json
+[2026-08-27 04:30:54 PP4 TP1 EP1] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp1_pp4_dp0.json
+[2026-08-27 04:30:54 PP5 TP2 EP2] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp2_pp5_dp0.json
+[2026-08-27 04:30:54 PP5 TP3 EP3] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp3_pp5_dp0.json
+[2026-08-27 04:30:54 PP5 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:30:54 PP5 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:30:54 PP5 TP0 EP0] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp0_pp5_dp0.json
+[2026-08-27 04:30:54 PP4 TP2 EP2] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp2_pp4_dp0.json
+[2026-08-27 04:30:54 PP4 TP3 EP3] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp3_pp4_dp0.json
+[2026-08-27 04:30:54 PP5 TP1 EP1] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp1_pp5_dp0.json
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO[TensorRT-LLM][INFO] Set logger level to INFO
+
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.37profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.12profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.87profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.32profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.91profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.77profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.94profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.88profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.20profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.87profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  7.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 10.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.46profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.67profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.91profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.29profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.49profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  7.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.30profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.94profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.47profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.12profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.06profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.35profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  9.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:01<00:02,  2.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.89profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.58profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.32profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.70profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.20profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.27profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  6.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  7.03profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.56profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.73profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.88profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.37profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:05,  1.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:01<00:02,  2.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.05profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.19profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.77profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.18profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.25profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  3.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.18profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.20profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.37profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  9.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.27profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  8.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.43profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.89profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.35profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.96profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.90profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.30profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.72profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.18profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.33profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.51profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  2.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.03profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.68profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.16profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.01profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.24profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 10.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:02,  2.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.06profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.83profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.35profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.87profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.19profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  3.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.06profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.86profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 13.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 13.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.13profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:01,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:01,  3.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.47profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  6.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.47profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.57profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.22profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.17profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.50profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.71profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.19profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.55profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.90profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:05,  1.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:01<00:06,  1.00s/profile]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.33profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.19profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.03profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.76profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.28profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.32profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.15profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.79profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.31profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.18profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 11.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:01,  3.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.06profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:01,  3.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  6.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  3.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.05profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.60profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.50profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.92profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.00profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.29profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 11.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.09profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.43profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.11profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.91profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  7.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:02,  2.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.18profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.78profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.75profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.89profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.64profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.05profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.47profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.47profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.65profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:05,  1.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.85profile/s]
+[2026-08-27 04:31:28 PP5 TP1 EP1] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.21profile/s]
+[2026-08-27 04:31:28 PP5 TP0 EP0] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.74profile/s]
+[2026-08-27 04:31:28 PP5 TP2 EP2] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.77profile/s]
+[2026-08-27 04:31:28 PP5 TP3 EP3] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.19profile/s][2026-08-27 04:31:28 PP5 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:28 PP5 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:28 PP5 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:28 PP5 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:28 PP5 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:28 PP5 TP0 EP0] max_total_num_tokens=3801408, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=54, context_len=1048576, available_gpu_mem=16.30 GB
+[2026-08-27 04:31:28 PP5 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:28 PP5 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:28 PP5 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.87profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.90profile/s]
+[2026-08-27 04:31:29 PP4 TP3 EP3] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.87profile/s]
+[2026-08-27 04:31:29 PP4 TP2 EP2] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.93profile/s]
+[2026-08-27 04:31:29 PP4 TP0 EP0] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.43profile/s]
+[2026-08-27 04:31:29 PP4 TP1 EP1] FlashInfer autotune completed.
+[2026-08-27 04:31:29 PP4 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:29 PP4 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:29 PP4 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:29 PP4 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:29 PP4 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:29 PP4 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:29 PP4 TP0 EP0] max_total_num_tokens=3801408, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=54, context_len=1048576, available_gpu_mem=16.30 GB
+[2026-08-27 04:31:29 PP4 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:29 PP4 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+W0827 04:31:30.513926  2527 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50a93b80, len: 6912: Operation not permitted [1]
+W0827 04:31:30.514082  2528 memory_location.cpp:73] Failed to get NUMA node, addr: 0x507df700, len: 6912: Operation not permitted [1]
+W0827 04:31:30.514106  2529 memory_location.cpp:73] Failed to get NUMA node, addr: 0x508aa940, len: 6912: Operation not permitted [1]
+W0827 04:31:30.514214  2530 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50f67bc0, len: 6912: Operation not permitted [1]
+W0827 04:31:30.514272  2531 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50a2f680, len: 55296: Operation not permitted [1]
+W0827 04:31:30.514393  2532 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50afecc0, len: 55296: Operation not permitted [1]
+W0827 04:31:30.514506  2533 memory_location.cpp:73] Failed to get NUMA node, addr: 0x502be580, len: 6912: Operation not permitted [1]
+W0827 04:31:30.514607  2534 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5054d3c0, len: 13824: Operation not permitted [1]
+W0827 04:31:30.514652  2535 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50ad4580, len: 6912: Operation not permitted [1]
+W0827 04:31:30.514743  2536 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50eb9880, len: 6912: Operation not permitted [1]
+W0827 04:31:31.680995  2556 memory_location.cpp:73] Failed to get NUMA node, addr: 0x501c3300, len: 6912: Operation not permitted [1]
+W0827 04:31:31.680985  2557 memory_location.cpp:73] Failed to get NUMA node, addr: 0x504ecec0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.680985  2555 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd4bac0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681169  2558 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50706900, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681272  2559 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5006d700, len: 55296: Operation not permitted [1]
+W0827 04:31:31.681360  2560 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50041cc0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.681468  2561 memory_location.cpp:73] Failed to get NUMA node, addr: 0x506f5a40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681562  2562 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fdfdf40, len: 13824: Operation not permitted [1]
+W0827 04:31:31.681607  2563 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd16080, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681685  2564 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fcfa300, len: 6912: Operation not permitted [1]
+W0827 04:31:31.733497  2583 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51554040, len: 6912: Operation not permitted [1]
+W0827 04:31:31.733552  2584 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5110a8c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.733618  2585 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51ca8e80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.733795  2586 memory_location.cpp:73] Failed to get NUMA node, addr: 0x52017680, len: 6912: Operation not permitted [1]
+W0827 04:31:31.733909  2587 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5174af80, len: 55296: Operation not permitted [1]
+W0827 04:31:31.734004  2588 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51770440, len: 55296: Operation not permitted [1]
+W0827 04:31:31.734061  2589 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51118000, len: 6912: Operation not permitted [1]
+W0827 04:31:31.734179  2590 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51564f40, len: 13824: Operation not permitted [1]
+W0827 04:31:31.734277  2591 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51f69480, len: 6912: Operation not permitted [1]
+W0827 04:31:31.734400  2592 memory_location.cpp:73] Failed to get NUMA node, addr: 0x517c2940, len: 6912: Operation not permitted [1]
+W0827 04:31:31.741586  2611 memory_location.cpp:73] Failed to get NUMA node, addr: 0x519520c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.741609  2612 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50ff8c40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.741679  2613 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51a2ad80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.741792  2614 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51b5b300, len: 6912: Operation not permitted [1]
+W0827 04:31:31.741938  2615 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50fe5b80, len: 55296: Operation not permitted [1]
+W0827 04:31:31.742043  2616 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51625140, len: 55296: Operation not permitted [1]
+W0827 04:31:31.742168  2617 memory_location.cpp:73] Failed to get NUMA node, addr: 0x516bf640, len: 6912: Operation not permitted [1]
+W0827 04:31:31.742278  2618 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51b73e80, len: 13824: Operation not permitted [1]
+W0827 04:31:31.742414  2619 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51ef60c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.742429  2620 memory_location.cpp:73] Failed to get NUMA node, addr: 0x513e95c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.900663  2641 memory_location.cpp:73] Failed to get NUMA node, addr: 0x517e8c80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.900679  2640 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5170b840, len: 6912: Operation not permitted [1]
+W0827 04:31:31.900738  2642 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5131d0c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.900669  2639 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51372400, len: 6912: Operation not permitted [1]
+W0827 04:31:31.900923  2643 memory_location.cpp:73] Failed to get NUMA node, addr: 0x512b7e00, len: 55296: Operation not permitted [1]
+W0827 04:31:31.900976  2644 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51380200, len: 55296: Operation not permitted [1]
+W0827 04:31:31.901000  2646 memory_location.cpp:73] Failed to get NUMA node, addr: 0x513151c0, len: 13824: Operation not permitted [1]
+W0827 04:31:31.901011  2645 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51194f40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.901038  2647 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51520740, len: 6912: Operation not permitted [1]
+W0827 04:31:31.901073  2648 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51724980, len: 6912: Operation not permitted [1]
+W0827 04:31:31.931138  2668 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51061000, len: 6912: Operation not permitted [1]
+W0827 04:31:31.931178  2669 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50d74c00, len: 6912: Operation not permitted [1]
+W0827 04:31:31.931098  2667 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50b2fb40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.931272  2670 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50de9000, len: 6912: Operation not permitted [1]
+W0827 04:31:31.931349  2671 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5097dc80, len: 55296: Operation not permitted [1]
+W0827 04:31:31.931519  2672 memory_location.cpp:73] Failed to get NUMA node, addr: 0x515aa000, len: 55296: Operation not permitted [1]
+W0827 04:31:31.931557  2675 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51269700, len: 6912: Operation not permitted [1]
+W0827 04:31:31.931556  2673 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5097bd00, len: 6912: Operation not permitted [1]
+W0827 04:31:31.931576  2674 memory_location.cpp:73] Failed to get NUMA node, addr: 0x514ee800, len: 13824: Operation not permitted [1]
+W0827 04:31:31.931759  2676 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50deac80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.983201  2696 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50819380, len: 6912: Operation not permitted [1]
+W0827 04:31:31.983201  2695 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50600a80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.983378  2697 memory_location.cpp:73] Failed to get NUMA node, addr: 0x506ddf40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.983438  2698 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50716800, len: 6912: Operation not permitted [1]
+W0827 04:31:31.983490  2699 memory_location.cpp:73] Failed to get NUMA node, addr: 0x501575c0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.983624  2700 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50863340, len: 55296: Operation not permitted [1]
+W0827 04:31:31.983737  2701 memory_location.cpp:73] Failed to get NUMA node, addr: 0x508d7a40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.983930  2702 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50668c40, len: 13824: Operation not permitted [1]
+W0827 04:31:31.984035  2703 memory_location.cpp:73] Failed to get NUMA node, addr: 0x506841c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.984107  2704 memory_location.cpp:73] Failed to get NUMA node, addr: 0x508f16c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.001853  2725 memory_location.cpp:73] Failed to get NUMA node, addr: 0x507c8d40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.001852  2724 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50b39240, len: 6912: Operation not permitted [1]
+W0827 04:31:32.002048  2727 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50a8ecc0, len: 55296: Operation not permitted [1]
+W0827 04:31:32.001888  2726 memory_location.cpp:73] Failed to get NUMA node, addr: 0x512a8140, len: 6912: Operation not permitted [1]
+W0827 04:31:32.002044  2728 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50a68880, len: 55296: Operation not permitted [1]
+W0827 04:31:32.002257  2729 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5087b880, len: 6912: Operation not permitted [1]
+W0827 04:31:32.002264  2731 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50fc6980, len: 6912: Operation not permitted [1]
+W0827 04:31:32.002266  2730 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51331100, len: 13824: Operation not permitted [1]
+W0827 04:31:32.002470  2732 memory_location.cpp:73] Failed to get NUMA node, addr: 0x513384c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.002780  2723 memory_location.cpp:73] Failed to get NUMA node, addr: 0x512c6940, len: 6912: Operation not permitted [1]
+[2026-08-27 04:31:32] Dummy health check server started in background thread at 0.0.0.0:30000
+[rank16]:[W827 04:33:25.972222579 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank17]:[W827 04:33:25.980800100 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank19]:[W827 04:33:25.981413562 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank18]:[W827 04:33:26.644294178 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank16]:[W827 04:34:10.467297472 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank18]:[W827 04:34:10.467837454 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank17]:[W827 04:34:10.468151853 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank19]:[W827 04:34:10.468427857 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank22]:[W827 04:34:10.495396864 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank23]:[W827 04:34:10.495456053 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank21]:[W827 04:34:10.495795969 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank20]:[W827 04:34:10.495814123 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank22]:[W827 04:34:16.703081720 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank20]:[W827 04:34:16.703915664 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank21]:[W827 04:34:16.704033167 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank23]:[W827 04:34:16.704314444 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[2026-08-27 04:35:10 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.29
+[2026-08-27 04:35:10 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.29
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/p_starting_p_node3.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/p_starting_p_node3.log
new file mode 100644
index 0000000..61b4d1e
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/p_starting_p_node3.log
@@ -0,0 +1,753 @@
+/usr/local/lib/python3.12/dist-packages/torchao/quantization/quant_api.py:1731: SyntaxWarning: invalid escape sequence '\.'
+  """Configuration class for applying different quantization configs to modules or parameters based on their fully qualified names (FQNs).
+/sgl-workspace/sglang/python/sglang/launch_server.py:56: UserWarning: 'python -m sglang.launch_server' is still supported, but 'sglang serve' is the recommended entrypoint.
+  Example: sglang serve --model-path <model> [options]
+  warnings.warn(
+[2026-08-27 04:28:46] Breakable CUDA graph is incompatible with MLA attention; disabling prefill CUDA graph.
+[2026-08-27 04:28:47] Attention backend not specified. Use triton backend by default.
+[2026-08-27 04:28:47] Pipeline parallelism is incompatible with overlap schedule.
+[2026-08-27 04:28:49] server_args=ServerArgs(model_path='/data/hf_models/Kimi-K3', tokenizer_path='/data/hf_models/Kimi-K3', tokenizer_mode='auto', tokenizer_backend='huggingface', tokenizer_worker_num=1, detokenizer_worker_num=1, skip_tokenizer_init=False, load_format='auto', model_loader_extra_config='{}', trust_remote_code=True, context_length=None, is_embedding=False, enable_multimodal=None, revision=None, model_impl='auto', model_config_parser='auto', json_model_override_args='{}', dtype='auto', quantization=None, quantization_param_path=None, kv_cache_dtype='auto', enable_fp32_lm_head=False, modelopt_quant=None, modelopt_checkpoint_restore_path=None, modelopt_checkpoint_save_path=None, modelopt_export_path=None, quantize_and_serve=False, rl_quant_profile=None, enable_tf32_matmul=False, mem_fraction_static=0.88, max_running_requests=None, max_queued_requests=None, max_total_tokens=None, chunked_prefill_size=8192, enable_dynamic_chunking=False, max_prefill_tokens=16384, prefill_max_requests=None, schedule_policy='fcfs', enable_priority_scheduling=False, disable_priority_preemption=False, default_priority_value=None, abort_on_priority_when_disabled=False, schedule_low_priority_values_first=False, priority_scheduling_preemption_threshold=10, retraction_policy='length', schedule_conservativeness=1.0, page_size=64, swa_full_tokens_ratio=0.8, disable_hybrid_swa_memory=False, radix_eviction_policy='lru', prefill_only_disable_kv_cache=False, disable_radix_cache=True, enable_page_major_kv_layout=False, enable_unified_memory=False, disable_chunked_prefix_cache=False, disable_overlap_schedule=True, num_continuous_decode_steps=1, scheduler_recv_interval=1, enable_mixed_chunk=False, nccl_port=None, dist_timeout=3600, dist_init_addr='174.1.60.1:20000', nnodes=4, node_rank=3, tp_size=4, dcp_size=1, pp_size=8, pp_max_micro_batch_size=None, pp_async_batch_depth=0, dp_size=1, load_balance_method='follow_bootstrap_room', attn_cp_size=1, moe_dp_size=1, dcp_comm_backend='ag_rs', dcp_replicate_q_proj=None, dwdp_size=1, enable_prefill_cp=False, cp_strategy=None, enable_dsa_cache_layer_split=False, enable_dsa_prefill_context_parallel=False, dsa_prefill_cp_mode='round-robin-split', enable_prefill_context_parallel=False, prefill_cp_mode='in-seq-split', enable_cp_decode_attn_tp=False, enable_dp_attention=False, enable_dp_attention_local_control_broadcast=False, enable_dp_lm_head=False, enable_attn_tp_input_scattered=False, disable_attn_tp_gather=False, enable_p2p_check=False, device='cuda', base_gpu_id=0, gpu_id_step=1, random_seed=927276655, watchdog_timeout=300, soft_watchdog_timeout=None, sleep_on_idle=False, use_ray=False, custom_sigquit_handler=None, numa_node=None, gc_threshold=None, host='0.0.0.0', port=30000, fastapi_root_path='', smg_grpc_mode=False, grpc_mode=False, grpc_port=None, sidecar=None, sidecar_args=None, skip_server_warmup=False, warmups=None, enable_http2=False, ssl_keyfile=None, ssl_certfile=None, ssl_ca_certs=None, ssl_keyfile_password=None, enable_ssl_refresh=False, api_key=None, admin_api_key=None, served_model_name='kimi-k3', weight_version='default', chat_template=None, hf_chat_template_name=None, completion_template=None, file_storage_path='sglang_storage', enable_cache_report=False, reasoning_parser=None, default_chat_template_kwargs=None, strip_thinking_cache=False, enable_strict_thinking=False, tool_call_parser=None, tool_server=None, sampling_defaults='model', asr_max_buffer_seconds=60, asr_max_concurrent_sessions=32, preferred_sampling_params=None, allow_auto_truncate=False, stream_interval=1, batch_notify_size=16, stream_response_default_include_usage=False, incremental_streaming_output=False, enable_streaming_session=False, enable_session_radix_cache=False, log_level='info', log_level_http=None, log_requests=False, log_requests_level=2, log_requests_format='text', log_requests_target=None, uvicorn_access_log_exclude_prefixes=[], crash_dump_folder=None, show_time_cost=False, enable_metrics=False, smg_http_sidecar_port=None, enable_mfu_metrics=False, enable_metrics_for_all_schedulers=False, load_snapshot_publish_interval=15, tokenizer_metrics_custom_labels_header='x-custom-labels', tokenizer_metrics_allowed_custom_labels=None, extra_metric_labels=None, bucket_time_to_first_token=None, bucket_inter_token_latency=None, bucket_e2e_request_latency=None, prompt_tokens_buckets=None, generation_tokens_buckets=None, gc_warning_threshold_secs=0.0, decode_log_interval=40, enable_request_time_stats_logging=False, kv_events_config=None, enable_forward_pass_metrics=False, forward_pass_metrics_worker_id='', forward_pass_metrics_ipc_name=None, enable_trace=False, trace_modules='request', otlp_traces_endpoint='localhost:4317', export_metrics_to_file=False, export_metrics_to_file_dir=None, stat_loggers=None, constrained_json_whitespace_pattern=None, constrained_json_disable_any_whitespace=False, attention_backend='triton', decode_attention_backend=None, prefill_attention_backend=None, sampling_backend='flashinfer', grammar_backend='xgrammar', radix_cache_backend=None, mm_attention_backend=None, fp8_gemm_runner_backend='auto', fp4_gemm_runner_backend='auto', bf16_gemm_backend='auto', dsa_prefill_backend=None, dsa_decode_backend=None, dsa_paged_mqa_logits_backend='auto', dsa_topk_backend='sgl-kernel', disable_flashinfer_autotune=False, flashinfer_autotune_skip_ops=None, mamba_backend='triton', cuda_graph_config=CudaGraphConfig(decode=PhaseConfig(backend='disabled', max_bs=16, bs=[1, 2, 4, 8, 12, 16], tc_compiler='eager', full_prefill_max_req=None), prefill=PhaseConfig(backend='disabled', max_bs=2048, bs=[4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], tc_compiler='eager', full_prefill_max_req=None)), cuda_graph_backend_decode=None, cuda_graph_backend_prefill=None, cuda_graph_max_bs_decode=16, cuda_graph_max_bs_prefill=None, cuda_graph_bs_decode=None, cuda_graph_bs_prefill=None, cuda_graph_tc_compiler=None, disable_prefill_cuda_graph=False, disable_decode_cuda_graph=False, disable_cuda_graph=False, disable_cuda_graph_padding=False, enable_profile_cuda_graph=False, enable_cudagraph_gc=False, debug_cuda_graph=False, enable_layerwise_nvtx_marker=False, enable_nccl_nvls=False, enable_symm_mem=False, triton_attention_reduce_in_fp32=False, triton_attention_num_kv_splits=8, triton_attention_split_tile_size=None, flashinfer_mla_disable_ragged=False, enable_fused_qk_norm_rope=False, enable_precise_embedding_interpolation=False, enable_fused_moe_sum_all_reduce=False, enable_deepseek_v4_fp4_indexer=False, disable_custom_all_reduce=False, enable_mscclpp=False, enable_torch_symm_mem=False, enable_scattered_sconv=False, pre_warm_nccl=False, enable_quant_communications=False, enable_flashinfer_allreduce_fusion=False, enforce_disable_flashinfer_allreduce_fusion=False, flashinfer_allreduce_fusion_backend=None, enable_aiter_allreduce_fusion=False, enable_torch_compile=False, enable_torch_compile_debug_mode=False, torch_compile_max_bs=32, torchao_config='', speculative_algorithm=None, speculative_draft_model_path=None, speculative_draft_model_revision=None, speculative_draft_load_format=None, speculative_num_steps=None, speculative_eagle_topk=None, speculative_num_draft_tokens=None, speculative_dflash_block_size=None, speculative_dspark_block_size=None, speculative_dspark_sps_table_path=None, speculative_dspark_confidence_sts_path=None, speculative_dspark_align_verify_tokens_to_graph_tier=False, speculative_accept_threshold_single=1.0, speculative_accept_threshold_acc=1.0, speculative_use_rejection_sampling=False, speculative_token_map=None, speculative_attention_mode='prefill', speculative_draft_attention_backend=None, speculative_draft_window_size=None, speculative_moe_runner_backend='flashinfer_mxfp4', speculative_moe_a2a_backend=None, speculative_draft_model_quantization=None, speculative_skip_dp_mlp_sync=False, enable_multi_layer_eagle=False, speculative_adaptive=False, speculative_adaptive_config=None, decoupled_spec_bind_endpoint=None, decoupled_spec_connect_endpoints=None, decoupled_spec_rank=None, decoupled_spec_role='null', spec_trace_dir=None, speculative_ngram_min_bfs_breadth=1, speculative_ngram_max_bfs_breadth=10, speculative_ngram_match_type='BFS', speculative_ngram_max_trie_depth=18, speculative_ngram_capacity=10000000, speculative_ngram_external_corpus_path=None, speculative_ngram_external_sam_budget=0, speculative_ngram_external_corpus_max_tokens=10000000, ep_size=4, moe_a2a_backend='none', moe_runner_backend='flashinfer_mxfp4', flashinfer_mxfp4_moe_precision='default', deepep_mode='auto', fuseep_mode=2, deepep_dispatcher_output_dtype='auto', ep_num_redundant_experts=0, ep_dispatch_algorithm=None, init_expert_location='trivial', enable_eplb=False, eplb_algorithm='auto', eplb_rebalance_num_iterations=1000, eplb_rebalance_layers_per_chunk=None, eplb_min_rebalancing_utilization_threshold=1.0, expert_distribution_recorder_mode=None, expert_distribution_recorder_buffer_size=1000, enable_expert_distribution_metrics=False, deepep_config=None, moe_dense_tp_size=None, elastic_ep_backend=None, enable_elastic_expert_backup=False, mooncake_ib_device=None, enable_waterfill=False, ep_join_mode=None, ep_join_rank_offset=0, elastic_ep_initial_size=None, max_ep_size=None, elastic_ep_scale_timeout=600, elastic_ep_rejoin=False, disable_flashinfer_cutlass_moe_fp4_allgather=False, disable_shared_experts_fusion=False, enforce_shared_experts_fusion=False, max_mamba_cache_size=None, mamba_ssm_dtype=None, mamba_max_states_per_path=-1, enable_mamba_cache_stochastic_rounding=False, mamba_cache_philox_rounds=0, mamba_full_memory_ratio=0.36, mamba_radix_cache_strategy='extra_buffer_lazy', uses_mamba_radix_cache=False, mamba_track_interval=256, enable_int8_mamba_checkpoint=False, int8_mamba_ckpt_size=None, linear_attn_backend='triton', linear_attn_decode_backend=None, linear_attn_prefill_backend=None, linear_attn_verify_backend=None, enable_linear_replayssm=False, linear_replayssm_cache_len=16, enable_linear_replayssm_spec=False, enable_hierarchical_cache=False, hicache_ratio=2.0, hicache_size=0, hicache_write_policy='write_through', hicache_io_backend='kernel', hicache_mem_layout='page_first', hicache_storage_backend=None, hicache_storage_prefetch_policy='timeout', hicache_storage_backend_extra_config=None, enable_hisparse=False, hisparse_config=None, enable_broadcast_mm_inputs_process=False, enable_prefix_mm_cache=False, mm_enable_dp_encoder=False, mm_process_config={}, mm_processor_worker_num=0, mm_io_worker_num=0, limit_mm_data_per_request=None, enable_mm_global_cache=False, disable_fast_image_processor=False, mm_feature_transport='cpu', keep_mm_feature_on_device=False, enable_lora=None, enable_lora_overlap_loading=None, max_lora_rank=None, lora_target_modules=None, lora_paths=None, max_loaded_loras=None, max_loras_per_batch=8, lora_eviction_policy='lru', lora_backend='csgmv', max_lora_chunk_size=16, experts_shared_outer_loras=None, lora_use_virtual_experts=False, lora_strict_loading=False, lora_drain_wait_threshold=0.0, enable_two_batch_overlap=False, enable_single_batch_overlap=False, tbo_token_distribution_threshold=0.48, cpu_offload_gb=0, offload_group_size=-1, offload_num_in_group=1, offload_prefetch_step=1, offload_mode='cpu', enable_lmcache=False, lmcache_config_file=None, enable_flexkv=False, flexkv_config_file=None, kt_weight_path=None, kt_method='AMXINT4', kt_cpuinfer=None, kt_threadpool_count=2, kt_num_gpu_experts=None, kt_max_deferred_experts_per_token=None, dllm_algorithm=None, dllm_algorithm_config=None, dllm_fdfo=True, disaggregation_mode='prefill', disaggregation_transfer_backend='mooncake', disaggregation_bootstrap_port=28800, disaggregation_ib_device='mlx5_0,mlx5_1,mlx5_2,mlx5_3', disaggregation_decode_enable_radix_cache=False, disaggregation_decode_enable_offload_kvcache=False, num_reserved_decode_tokens=512, disaggregation_decode_extra_slots=None, disaggregation_decode_polling_interval=1, optimistic_prefill_attempts=0, encoder_only=False, language_only=False, encoder_transfer_backend='zmq_to_tokenizer', encoder_urls=[], encoder_bootstrap_port=8997, encoder_register_urls=[], enable_adaptive_dispatch_to_encoder=False, enable_pdmux=False, pdmux_config_path=None, sm_group_num=8, custom_weight_loader=[], weight_loader_disable_mmap=False, weight_loader_prefetch_checkpoints=False, weight_loader_prefetch_num_threads=4, weight_loader_drop_cache_after_load=False, remote_instance_weight_loader_seed_instance_ip=None, remote_instance_weight_loader_seed_instance_service_port=None, remote_instance_weight_loader_send_weights_group_ports=None, remote_instance_weight_loader_backend='nccl', remote_instance_weight_loader_start_seed_via_transfer_engine=False, engine_info_bootstrap_port=6789, modelexpress_config=None, download_dir=None, model_checksum=None, delete_ckpt_after_loading=False, decrypted_config_file=None, decrypted_draft_config_file=None, checkpoint_engine_wait_weights_before_ready=False, enable_prefill_delayer=False, prefill_delayer_max_delay_passes=30, prefill_delayer_token_usage_low_watermark=None, prefill_delayer_forward_passes_buckets=None, prefill_delayer_wait_seconds_buckets=None, prefill_delayer_queue_min_ratio=None, prefill_delayer_max_delay_ms=None, min_free_slots_delay=None, enable_deterministic_inference=False, rl_on_policy_target=None, kv_canary='none', kv_canary_real_data='none', kv_canary_sweep_interval=0, enable_dynamic_batch_tokenizer=False, dynamic_batch_tokenizer_batch_size=32, dynamic_batch_tokenizer_batch_timeout=0.002, enable_tokenizer_batch_encode=False, disable_tokenizer_batch_decode=False, debug_tensor_dump_output_folder=None, debug_tensor_dump_layers=None, debug_tensor_dump_input_file=None, enable_memory_saver=False, enable_weights_cpu_backup=False, enable_draft_weights_cpu_backup=False, enable_custom_logit_processor=False, enable_return_hidden_states=False, enable_return_routed_experts=False, enable_return_indexer_topk=False, disable_outlines_disk_cache=False, enable_mis=False, weight_cache_mode='off', weight_cache_socket=None, weight_cache_timeout=1800, forward_hooks=None, msprobe_dump_config=None)
+[2026-08-27 04:28:49] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:49] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:49] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:49] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:49] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:49] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:49] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:49] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:05 PP6 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:05 PP6 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:07 PP6 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:07 PP6 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:07 PP6 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:07 PP7 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:07 PP7 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:07 PP7 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:07 PP7 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:07 PP6 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:07 PP6 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:07.662256   238 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:07.662379   238 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:07.662392   238 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.4 port: 12001
+I0827 04:29:07.662429   238 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.4:15914
+I0827 04:29:07.662514   238 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:07.685189   238 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:07.689950   238 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:07.695539   238 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:07.701294   238 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:07.703691   238 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:07.703715   238 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:07.703760   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:07.710492   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:07.711953   238 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:04
+I0827 04:29:07.711989   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:07.735716   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:07.737195   238 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:04
+I0827 04:29:07.737223   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:07.763653   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:07.765146   238 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:04
+I0827 04:29:07.765172   238 rdma_context.cpp:205] Using SIEVE endpoint store
+[2026-08-27 04:29:07 PP7 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:07 PP7 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:07 PP7 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:07 PP7 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+I0827 04:29:07.791589   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:07.793025   238 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:04
+I0827 04:29:07.793139   238 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:07 PP6 TP3 EP3] Init torch distributed begin.
+[2026-08-27 04:29:07 PP6 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:07 PP6 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:08 PP6 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:08.916008   235 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:08.916072   235 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:08.916082   235 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.4 port: 12001
+I0827 04:29:08.916121   235 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.4:15737
+I0827 04:29:08.916215   235 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:08.941840   235 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:08.950320   235 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:08.979278   235 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:08.988204   235 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:08.991855   235 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:08.991879   235 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:08.991928   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:08.999985   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:09.001416   235 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:04
+I0827 04:29:09.001446   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:09.027477   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:09.028954   235 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:04
+I0827 04:29:09.028980   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:09.055524   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:09.056981   235 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:04
+I0827 04:29:09.057020   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:09.083544   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:09.084911   235 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:04
+I0827 04:29:09.085031   235 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:09 PP6 TP0 EP0] Init torch distributed begin.
+[2026-08-27 04:29:09 PP7 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:09 PP6 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:09 PP7 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:09 PP6 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:09 PP7 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:09 PP7 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:10.000486   242 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:10.000609   242 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:10.000628   242 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.4 port: 12001
+I0827 04:29:10.000689   242 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.4:16890
+I0827 04:29:10.000851   242 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:10.006592   236 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:10.006711   236 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:10.006728   236 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.4 port: 12001
+I0827 04:29:10.006786   236 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.4:15893
+I0827 04:29:10.006963   236 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:10.022238   241 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:10.022354   241 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:10.022370   241 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.4 port: 12001
+I0827 04:29:10.022423   241 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.4:16277
+I0827 04:29:10.022593   241 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:10.032052   242 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:10.032101   236 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:10.039772   236 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:10.039858   242 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:10.047292   236 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:10.047341   242 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:10.047785   241 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:10.054806   242 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:10.054838   236 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:10.055280   241 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:10.059140   236 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:10.059171   236 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:10.059175   242 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:10.059201   242 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:10.059227   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.059260   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.062610   241 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:10.066787   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:10.066826   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:10.068500   242 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:04
+I0827 04:29:10.068532   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.068558   236 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:04
+I0827 04:29:10.068590   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.069866   241 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:10.072566   241 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:10.072592   241 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:10.072649   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.080149   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:10.081809   241 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:04
+I0827 04:29:10.081843   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.095701   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:10.097368   236 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:04
+I0827 04:29:10.097400   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.099974   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:10.101672   242 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:04
+I0827 04:29:10.101717   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.107705   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:10.109316   241 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:04
+I0827 04:29:10.109347   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.123868   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:10.125504   236 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:04
+I0827 04:29:10.125536   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.127957   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:10.129673   242 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:04
+I0827 04:29:10.129703   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.139813   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:10.141404   241 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:04
+I0827 04:29:10.141431   241 rdma_context.cpp:205] Using SIEVE endpoint store
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:10.150050   237 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:10.150159   237 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:10.150173   237 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.4 port: 12001
+I0827 04:29:10.150219   237 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.4:15413
+I0827 04:29:10.150358   237 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:10.151880   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:10.153475   236 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:04
+I0827 04:29:10.153601   236 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:10 PP6 TP1 EP1] Init torch distributed begin.
+I0827 04:29:10.155849   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:10.155864   237 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:10.157462   242 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:04
+I0827 04:29:10.157585   242 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:10 PP7 TP3 EP3] Init torch distributed begin.
+I0827 04:29:10.171923   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:10.172516   240 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:10.172667   240 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:10.172688   240 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.4 port: 12001
+I0827 04:29:10.172762   240 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.4:15705
+I0827 04:29:10.172977   240 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:10.173573   241 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:04
+I0827 04:29:10.173692   241 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:10 PP7 TP2 EP2] Init torch distributed begin.
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:10.177717   239 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:10.177865   239 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:10.177886   239 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.4 port: 12001
+I0827 04:29:10.177956   239 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.4:16192
+I0827 04:29:10.178146   239 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:10.183638   237 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:10.187090   239 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:10.191001   237 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:10.195258   239 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:10.198271   237 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:10.200984   237 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:10.201025   237 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:10.201079   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.203014   239 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:10.203156   240 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:10.209589   240 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:10.210305   239 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:10.212793   239 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:10.212821   239 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:10.212878   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.216092   240 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:10.219946   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:10.221297   239 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:04
+I0827 04:29:10.221323   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.222187   240 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:10.224284   240 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:10.224304   240 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:10.224344   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.230640   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:10.231778   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:10.231937   240 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:04
+I0827 04:29:10.231962   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.233362   237 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:04
+I0827 04:29:10.233397   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.236478   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:10.237651   240 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:04
+I0827 04:29:10.237670   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.250197   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:10.251711   239 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:04
+I0827 04:29:10.251737   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.263743   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:10.265311   237 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:04
+I0827 04:29:10.265339   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.267139   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:10.268607   240 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:04
+I0827 04:29:10.268632   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.273075   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:10.274682   237 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:04
+I0827 04:29:10.274710   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.276602   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:10.278275   239 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:04
+I0827 04:29:10.278314   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.295709   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:10.297385   240 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:04
+I0827 04:29:10.297510   240 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:10 PP7 TP1 EP1] Init torch distributed begin.
+I0827 04:29:10.299893   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:10.301563   237 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:04
+I0827 04:29:10.301689   237 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:10 PP6 TP2 EP2] Init torch distributed begin.
+I0827 04:29:10.303737   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:10.305307   239 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:04
+I0827 04:29:10.305433   239 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:10 PP7 TP0 EP0] Init torch distributed begin.
+[2026-08-27 04:29:26 PP7 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:26 PP6 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:27 PP6 TP0 EP0] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP6 TP3 EP3] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP6 TP2 EP2] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP6 TP1 EP1] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP7 TP1 EP1] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP7 TP0 EP0] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP7 TP3 EP3] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP7 TP2 EP2] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP7 TP1 EP1] Init torch distributed ends. elapsed=17.45 s, mem usage=0.46 GB
+[2026-08-27 04:29:27 PP7 TP3 EP3] Init torch distributed ends. elapsed=17.59 s, mem usage=0.46 GB
+[2026-08-27 04:29:27 PP7 TP2 EP2] Init torch distributed ends. elapsed=17.57 s, mem usage=0.46 GB
+[2026-08-27 04:29:27 PP7 TP0 EP0] Init torch distributed ends. elapsed=17.44 s, mem usage=0.46 GB
+[2026-08-27 04:29:27 PP6 TP1 EP1] Init torch distributed ends. elapsed=17.59 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP6 TP3 EP3] Init torch distributed ends. elapsed=19.95 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP6 TP0 EP0] Init torch distributed ends. elapsed=18.66 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP6 TP2 EP2] Init torch distributed ends. elapsed=17.44 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP6 TP1 EP1] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP7 TP1 EP1] Load weight begin. avail mem=82.12 GB
+[2026-08-27 04:29:27 PP6 TP0 EP0] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP6 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP7 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP7 TP3 EP3] Load weight begin. avail mem=82.12 GB
+[2026-08-27 04:29:27 PP6 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP7 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP6 TP2 EP2] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP6 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP7 TP2 EP2] Load weight begin. avail mem=82.12 GB
+[2026-08-27 04:29:28 PP7 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP6 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP7 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP6 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP7 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP7 TP0 EP0] Load weight begin. avail mem=82.12 GB
+[2026-08-27 04:29:28 PP6 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP7 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP7 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP6 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP7 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP6 TP3 EP3] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP6 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP6 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP6 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP7 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP7 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP7 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP6 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP7 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP6 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:30:40 PP7 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:40 PP7 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:40 PP7 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:40 PP7 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:40 PP7 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:40 PP7 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:40 PP7 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:40 PP7 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:41 PP7 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:41 PP7 TP1 EP1] Load weight end. elapsed=73.17 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=31.91 GB, mem usage=50.21 GB.
+[2026-08-27 04:30:41 PP7 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:41 PP7 TP0 EP0] Load weight end. elapsed=73.80 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=31.91 GB, mem usage=50.21 GB.
+[2026-08-27 04:30:42 PP6 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:42 PP6 TP1 EP1] Load weight end. elapsed=74.47 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:45 PP6 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:45 PP6 TP2 EP2] Load weight end. elapsed=77.20 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:50 PP7 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:50 PP7 TP3 EP3] Load weight end. elapsed=82.79 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=31.91 GB, mem usage=50.21 GB.
+[2026-08-27 04:30:51 PP6 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:51 PP6 TP3 EP3] Load weight end. elapsed=83.19 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:51 PP7 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:51 PP7 TP2 EP2] Load weight end. elapsed=83.68 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=31.91 GB, mem usage=50.21 GB.
+[2026-08-27 04:30:51 PP7 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:51 PP7 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:52 PP7 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:52 PP7 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:52 PP7 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:52 PP7 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:52 PP7 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:52 PP7 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:53 PP6 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:53 PP6 TP0 EP0] Load weight end. elapsed=85.47 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:53 PP6 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:53 PP6 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:53 PP6 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:53 PP6 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:53 PP6 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:53 PP6 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:53 PP6 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:53 PP6 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:54 PP7 TP3 EP3] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP7 TP2 EP2] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP7 TP1 EP1] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP7 TP0 EP0] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP6 TP3 EP3] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP6 TP2 EP2] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP6 TP1 EP1] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP6 TP0 EP0] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP6 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP6 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP7 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:54 PP6 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP7 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:54 PP7 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:54 PP7 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:54 PP6 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP6 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP6 TP3 EP3] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP6 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP6 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP6 TP2 EP2] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP6 TP1 EP1] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP7 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 16.31 GB
+[2026-08-27 04:30:54 PP7 TP2 EP2] Memory pool end. avail mem=14.95 GB
+[2026-08-27 04:30:54 PP6 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP6 TP0 EP0] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP7 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 16.31 GB
+[2026-08-27 04:30:54 PP7 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 16.31 GB
+[2026-08-27 04:30:54 PP7 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 16.31 GB
+[2026-08-27 04:30:54 PP7 TP3 EP3] Memory pool end. avail mem=14.95 GB
+[2026-08-27 04:30:54 PP7 TP0 EP0] Memory pool end. avail mem=14.95 GB
+[2026-08-27 04:30:54 PP7 TP1 EP1] Memory pool end. avail mem=14.95 GB
+[2026-08-27 04:30:54 PP7 TP2 EP2] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp2_pp7_dp0.json
+[2026-08-27 04:30:54 PP6 TP3 EP3] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp3_pp6_dp0.json
+[2026-08-27 04:30:54 PP6 TP1 EP1] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp1_pp6_dp0.json
+[2026-08-27 04:30:54 PP7 TP3 EP3] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp3_pp7_dp0.json
+[2026-08-27 04:30:54 PP6 TP2 EP2] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp2_pp6_dp0.json
+[2026-08-27 04:30:54 PP6 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:30:54 PP6 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:30:54 PP6 TP0 EP0] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp0_pp6_dp0.json
+[2026-08-27 04:30:54 PP7 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:30:54 PP7 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:30:54 PP7 TP0 EP0] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp0_pp7_dp0.json
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[2026-08-27 04:30:54 PP7 TP1 EP1] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp1_pp7_dp0.json
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 11.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 13.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.76profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.62profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.66profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.72profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.68profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.55profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.12profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  6.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  7.16profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.92profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.15profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.59profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.56profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 19.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  8.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.67profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.40profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.88profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  8.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  6.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.83profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.50profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.32profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.92profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.77profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.94profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.57profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 12.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 12.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 13.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.36profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.66profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  6.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.72profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  6.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.22profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.65profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.13profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.03profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.46profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.59profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.16profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  6.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.96profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.76profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.67profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  8.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.27profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.88profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.01profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.57profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.12profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 11.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  2.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.68profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.90profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.43profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.92profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  8.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.66profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  6.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  7.13profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.32profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.71profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:05,  1.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.16profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.31profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.77profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.28profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.48profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.87profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.40profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 13.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.35profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.98profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.88profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.73profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  3.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.88profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.09profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.20profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.97profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.83profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.47profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.40profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.60profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  8.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  7.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.97profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.05profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.05profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  6.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.49profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.72profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.19profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 10.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.91profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.64profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.58profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  3.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.35profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  9.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 12.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.09profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:02,  2.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 12.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.15profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.76profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.49profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.03profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.79profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.96profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.47profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.65profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.55profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.03profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  9.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.63profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.98profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:05,  1.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:01,  3.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.50profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  6.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.94profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.13profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.78profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.53profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.35profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.56profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.36profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:01<00:06,  1.06s/profile]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  8.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.91profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.66profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  8.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  7.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:02,  2.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.83profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.01profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.66profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.13profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.59profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  8.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  8.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.71profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.29profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.97profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.88profile/s]
+[2026-08-27 04:31:27 PP6 TP3 EP3] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.86profile/s]
+[2026-08-27 04:31:27 PP6 TP1 EP1] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.24profile/s]
+[2026-08-27 04:31:27 PP6 TP2 EP2] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.33profile/s]
+[2026-08-27 04:31:27 PP6 TP0 EP0] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.56profile/s][2026-08-27 04:31:27 PP6 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:27 PP6 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:27 PP6 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:27 PP6 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:27 PP6 TP0 EP0] max_total_num_tokens=3801408, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=54, context_len=1048576, available_gpu_mem=16.52 GB
+[2026-08-27 04:31:27 PP6 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:27 PP6 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:27 PP6 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:27 PP6 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
+[2026-08-27 04:31:28 PP7 TP2 EP2] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.32profile/s]
+[2026-08-27 04:31:28 PP7 TP3 EP3] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.60profile/s]
+[2026-08-27 04:31:28 PP7 TP0 EP0] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.60profile/s]
+[2026-08-27 04:31:28 PP7 TP1 EP1] FlashInfer autotune completed.
+[2026-08-27 04:31:28 PP7 TP2 EP2] multimem all-gather disabled (no multicast for world_size=4)
+[2026-08-27 04:31:28 PP7 TP3 EP3] multimem all-gather disabled (no multicast for world_size=4)
+[2026-08-27 04:31:28 PP7 TP1 EP1] multimem all-gather disabled (no multicast for world_size=4)
+[2026-08-27 04:31:28 PP7 TP0 EP0] multimem all-gather disabled (no multicast for world_size=4)
+[2026-08-27 04:31:28 PP7 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:28 PP7 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:28 PP7 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:28 PP7 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:28 PP7 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:28 PP7 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:28 PP7 TP0 EP0] max_total_num_tokens=3801408, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=54, context_len=1048576, available_gpu_mem=12.14 GB
+[2026-08-27 04:31:28 PP7 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:28 PP7 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+W0827 04:31:31.959518  2596 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5efc8ec0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.959637  2598 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ee654c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.959640  2597 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5eda3c80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.959542  2595 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5efc6f80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.959741  2599 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5eeaf500, len: 55296: Operation not permitted [1]
+W0827 04:31:31.959794  2600 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ef4e380, len: 55296: Operation not permitted [1]
+W0827 04:31:31.959827  2601 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5f548500, len: 6912: Operation not permitted [1]
+W0827 04:31:31.959918  2602 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5eec3bc0, len: 13824: Operation not permitted [1]
+W0827 04:31:31.960085  2603 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5f01bcc0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.960222  2604 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5f4c8ac0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.980162  2623 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60ba35c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.980152  2624 memory_location.cpp:73] Failed to get NUMA node, addr: 0x609ccf40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.980242  2625 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60d38600, len: 6912: Operation not permitted [1]
+W0827 04:31:31.980361  2626 memory_location.cpp:73] Failed to get NUMA node, addr: 0x609546c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.980516  2628 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60c49a00, len: 55296: Operation not permitted [1]
+W0827 04:31:31.980537  2627 memory_location.cpp:73] Failed to get NUMA node, addr: 0x607c6800, len: 55296: Operation not permitted [1]
+W0827 04:31:31.980547  2629 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60d11800, len: 6912: Operation not permitted [1]
+W0827 04:31:31.980834  2630 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60a1b740, len: 13824: Operation not permitted [1]
+W0827 04:31:31.981053  2632 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60ce5000, len: 6912: Operation not permitted [1]
+W0827 04:31:31.981062  2631 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60aef300, len: 6912: Operation not permitted [1]
+W0827 04:31:32.117022  2651 memory_location.cpp:73] Failed to get NUMA node, addr: 0x74c79d40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.117132  2652 memory_location.cpp:73] Failed to get NUMA node, addr: 0x74f99700, len: 6912: Operation not permitted [1]
+W0827 04:31:32.117317  2653 memory_location.cpp:73] Failed to get NUMA node, addr: 0x7546c740, len: 6912: Operation not permitted [1]
+W0827 04:31:32.117331  2654 memory_location.cpp:73] Failed to get NUMA node, addr: 0x74f4f140, len: 6912: Operation not permitted [1]
+W0827 04:31:32.117414  2656 memory_location.cpp:73] Failed to get NUMA node, addr: 0x750037c0, len: 55296: Operation not permitted [1]
+W0827 04:31:32.117445  2655 memory_location.cpp:73] Failed to get NUMA node, addr: 0x74bf1200, len: 55296: Operation not permitted [1]
+W0827 04:31:32.117705  2657 memory_location.cpp:73] Failed to get NUMA node, addr: 0x74f80080, len: 6912: Operation not permitted [1]
+W0827 04:31:32.117738  2658 memory_location.cpp:73] Failed to get NUMA node, addr: 0x74fbd2c0, len: 13824: Operation not permitted [1]
+W0827 04:31:32.117790  2659 memory_location.cpp:73] Failed to get NUMA node, addr: 0x74fa5e40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.118012  2660 memory_location.cpp:73] Failed to get NUMA node, addr: 0x74c24880, len: 6912: Operation not permitted [1]
+W0827 04:31:32.137427  2677 memory_location.cpp:73] Failed to get NUMA node, addr: 0x685bea40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.137485  2678 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6865c240, len: 6912: Operation not permitted [1]
+W0827 04:31:32.137612  2680 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6832f280, len: 6912: Operation not permitted [1]
+W0827 04:31:32.137683  2679 memory_location.cpp:73] Failed to get NUMA node, addr: 0x68511ec0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.137791  2681 memory_location.cpp:73] Failed to get NUMA node, addr: 0x687ae280, len: 55296: Operation not permitted [1]
+W0827 04:31:32.137800  2682 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6863aec0, len: 55296: Operation not permitted [1]
+W0827 04:31:32.137928  2683 memory_location.cpp:73] Failed to get NUMA node, addr: 0x685f4540, len: 6912: Operation not permitted [1]
+W0827 04:31:32.138145  2685 memory_location.cpp:73] Failed to get NUMA node, addr: 0x68b70f80, len: 6912: Operation not permitted [1]
+W0827 04:31:32.138144  2684 memory_location.cpp:73] Failed to get NUMA node, addr: 0x686e87c0, len: 13824: Operation not permitted [1]
+W0827 04:31:32.138219  2686 memory_location.cpp:73] Failed to get NUMA node, addr: 0x685491c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.187127  2706 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8eaeff40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.187187  2707 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8eaa1740, len: 6912: Operation not permitted [1]
+W0827 04:31:32.187136  2705 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8e5f2600, len: 6912: Operation not permitted [1]
+W0827 04:31:32.187413  2709 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ec4eb00, len: 55296: Operation not permitted [1]
+W0827 04:31:32.187405  2708 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ea3b800, len: 6912: Operation not permitted [1]
+W0827 04:31:32.187651  2710 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ed92ec0, len: 55296: Operation not permitted [1]
+W0827 04:31:32.187799  2711 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8eb3fbc0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.188133  2712 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ed0e1c0, len: 13824: Operation not permitted [1]
+W0827 04:31:32.188231  2713 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8e9d20c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.188272  2714 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8f2c7a80, len: 6912: Operation not permitted [1]
+W0827 04:31:32.252096  2732 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72b6b180, len: 6912: Operation not permitted [1]
+W0827 04:31:32.252050  2731 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72be0600, len: 6912: Operation not permitted [1]
+W0827 04:31:32.252147  2733 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72716c00, len: 6912: Operation not permitted [1]
+W0827 04:31:32.252276  2734 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72739940, len: 6912: Operation not permitted [1]
+W0827 04:31:32.252379  2735 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72da8a80, len: 55296: Operation not permitted [1]
+W0827 04:31:32.252439  2736 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72d15800, len: 55296: Operation not permitted [1]
+W0827 04:31:32.252678  2737 memory_location.cpp:73] Failed to get NUMA node, addr: 0x7315db40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.252701  2738 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72e0e700, len: 13824: Operation not permitted [1]
+W0827 04:31:32.252745  2739 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72f5f480, len: 6912: Operation not permitted [1]
+W0827 04:31:32.252832  2740 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72b295c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.268508  2760 memory_location.cpp:73] Failed to get NUMA node, addr: 0x69eefc00, len: 6912: Operation not permitted [1]
+W0827 04:31:32.268532  2759 memory_location.cpp:73] Failed to get NUMA node, addr: 0x69ec8680, len: 6912: Operation not permitted [1]
+W0827 04:31:32.268577  2761 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6a02e280, len: 6912: Operation not permitted [1]
+W0827 04:31:32.268708  2762 memory_location.cpp:73] Failed to get NUMA node, addr: 0x69d75dc0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.268766  2763 memory_location.cpp:73] Failed to get NUMA node, addr: 0x69fc4880, len: 55296: Operation not permitted [1]
+W0827 04:31:32.268886  2764 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6a0c8540, len: 55296: Operation not permitted [1]
+W0827 04:31:32.268975  2765 memory_location.cpp:73] Failed to get NUMA node, addr: 0x69c75940, len: 6912: Operation not permitted [1]
+W0827 04:31:32.269125  2766 memory_location.cpp:73] Failed to get NUMA node, addr: 0x698ae1c0, len: 13824: Operation not permitted [1]
+W0827 04:31:32.269232  2767 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6a5457c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.269354  2768 memory_location.cpp:73] Failed to get NUMA node, addr: 0x69c8fe80, len: 6912: Operation not permitted [1]
+W0827 04:31:32.281724  2785 memory_location.cpp:73] Failed to get NUMA node, addr: 0x678e47c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.281751  2786 memory_location.cpp:73] Failed to get NUMA node, addr: 0x67a39b80, len: 6912: Operation not permitted [1]
+W0827 04:31:32.281852  2787 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6798d9c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.281994  2788 memory_location.cpp:73] Failed to get NUMA node, addr: 0x67981dc0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.282085  2789 memory_location.cpp:73] Failed to get NUMA node, addr: 0x67823300, len: 55296: Operation not permitted [1]
+W0827 04:31:32.282146  2790 memory_location.cpp:73] Failed to get NUMA node, addr: 0x67a3bb00, len: 55296: Operation not permitted [1]
+W0827 04:31:32.282214  2791 memory_location.cpp:73] Failed to get NUMA node, addr: 0x67cdad40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.282285  2792 memory_location.cpp:73] Failed to get NUMA node, addr: 0x68028e40, len: 13824: Operation not permitted [1]
+W0827 04:31:32.282544  2794 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6796cf40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.282545  2793 memory_location.cpp:73] Failed to get NUMA node, addr: 0x67fa6600, len: 6912: Operation not permitted [1]
+[2026-08-27 04:31:32] Dummy health check server started in background thread at 0.0.0.0:30000
+[rank27]:[W827 04:34:16.127713860 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank25]:[W827 04:34:16.128748517 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank24]:[W827 04:34:16.129831915 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank26]:[W827 04:34:16.136239904 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank26]:[W827 04:35:04.964905528 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank25]:[W827 04:35:04.966225881 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank27]:[W827 04:35:04.966239242 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank24]:[W827 04:35:04.966301664 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank31]:[W827 04:35:04.990701389 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank28]:[W827 04:35:04.991914270 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank29]:[W827 04:35:04.994546241 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank30]:[W827 04:35:04.999020079 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank31]:[W827 04:35:10.873647776 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank30]:[W827 04:35:10.873901815 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank28]:[W827 04:35:10.874227564 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank29]:[W827 04:35:10.885318244 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[2026-08-27 04:35:10 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.29
+[2026-08-27 04:35:10 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.29
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/p_starting_router.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/p_starting_router.log
new file mode 100644
index 0000000..1190be5
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/p_starting_router.log
@@ -0,0 +1 @@
+Error response from daemon: No such container: kimi3_pro6000_pd_pp8_standard_router
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/ready_d_node0.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/ready_d_node0.log
new file mode 100644
index 0000000..91d61b8
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/ready_d_node0.log
@@ -0,0 +1,600 @@
+/usr/local/lib/python3.12/dist-packages/torchao/quantization/quant_api.py:1731: SyntaxWarning: invalid escape sequence '\.'
+  """Configuration class for applying different quantization configs to modules or parameters based on their fully qualified names (FQNs).
+/sgl-workspace/sglang/python/sglang/launch_server.py:56: UserWarning: 'python -m sglang.launch_server' is still supported, but 'sglang serve' is the recommended entrypoint.
+  Example: sglang serve --model-path <model> [options]
+  warnings.warn(
+[2026-08-27 04:36:08] KV cache is forced as chunk cache for decode server
+[2026-08-27 04:36:11] Breakable CUDA graph is incompatible with MLA attention; disabling prefill CUDA graph.
+[2026-08-27 04:36:12] Attention backend not specified. Use triton backend by default.
+[2026-08-27 04:36:14] server_args=ServerArgs(model_path='/data/hf_models/Kimi-K3', tokenizer_path='/data/hf_models/Kimi-K3', tokenizer_mode='auto', tokenizer_backend='huggingface', tokenizer_worker_num=1, detokenizer_worker_num=1, skip_tokenizer_init=False, load_format='auto', model_loader_extra_config='{}', trust_remote_code=True, context_length=None, is_embedding=False, enable_multimodal=None, revision=None, model_impl='auto', model_config_parser='auto', json_model_override_args='{}', dtype='auto', quantization=None, quantization_param_path=None, kv_cache_dtype='auto', enable_fp32_lm_head=False, modelopt_quant=None, modelopt_checkpoint_restore_path=None, modelopt_checkpoint_save_path=None, modelopt_export_path=None, quantize_and_serve=False, rl_quant_profile=None, enable_tf32_matmul=False, mem_fraction_static=0.88, max_running_requests=None, max_queued_requests=None, max_total_tokens=None, chunked_prefill_size=8192, enable_dynamic_chunking=False, max_prefill_tokens=16384, prefill_max_requests=None, schedule_policy='fcfs', enable_priority_scheduling=False, disable_priority_preemption=False, default_priority_value=None, abort_on_priority_when_disabled=False, schedule_low_priority_values_first=False, priority_scheduling_preemption_threshold=10, retraction_policy='length', schedule_conservativeness=1.0, page_size=64, swa_full_tokens_ratio=0.8, disable_hybrid_swa_memory=False, radix_eviction_policy='lru', prefill_only_disable_kv_cache=False, disable_radix_cache=True, enable_page_major_kv_layout=False, enable_unified_memory=False, disable_chunked_prefix_cache=False, disable_overlap_schedule=False, num_continuous_decode_steps=1, scheduler_recv_interval=1, enable_mixed_chunk=False, nccl_port=None, dist_timeout=3600, dist_init_addr='174.1.60.5:20000', nnodes=4, node_rank=0, tp_size=32, dcp_size=1, pp_size=1, pp_max_micro_batch_size=None, pp_async_batch_depth=0, dp_size=1, load_balance_method='round_robin', attn_cp_size=1, moe_dp_size=1, dcp_comm_backend='ag_rs', dcp_replicate_q_proj=None, dwdp_size=1, enable_prefill_cp=False, cp_strategy=None, enable_dsa_cache_layer_split=False, enable_dsa_prefill_context_parallel=False, dsa_prefill_cp_mode='round-robin-split', enable_prefill_context_parallel=False, prefill_cp_mode='in-seq-split', enable_cp_decode_attn_tp=False, enable_dp_attention=False, enable_dp_attention_local_control_broadcast=False, enable_dp_lm_head=False, enable_attn_tp_input_scattered=False, disable_attn_tp_gather=False, enable_p2p_check=False, device='cuda', base_gpu_id=0, gpu_id_step=1, random_seed=369235288, watchdog_timeout=300, soft_watchdog_timeout=None, sleep_on_idle=False, use_ray=False, custom_sigquit_handler=None, numa_node=None, gc_threshold=None, host='0.0.0.0', port=30000, fastapi_root_path='', smg_grpc_mode=False, grpc_mode=False, grpc_port=None, sidecar=None, sidecar_args=None, skip_server_warmup=False, warmups=None, enable_http2=False, ssl_keyfile=None, ssl_certfile=None, ssl_ca_certs=None, ssl_keyfile_password=None, enable_ssl_refresh=False, api_key=None, admin_api_key=None, served_model_name='kimi-k3', weight_version='default', chat_template=None, hf_chat_template_name=None, completion_template=None, file_storage_path='sglang_storage', enable_cache_report=False, reasoning_parser=None, default_chat_template_kwargs=None, strip_thinking_cache=False, enable_strict_thinking=False, tool_call_parser=None, tool_server=None, sampling_defaults='model', asr_max_buffer_seconds=60, asr_max_concurrent_sessions=32, preferred_sampling_params=None, allow_auto_truncate=False, stream_interval=1, batch_notify_size=16, stream_response_default_include_usage=False, incremental_streaming_output=False, enable_streaming_session=False, enable_session_radix_cache=False, log_level='info', log_level_http=None, log_requests=False, log_requests_level=2, log_requests_format='text', log_requests_target=None, uvicorn_access_log_exclude_prefixes=[], crash_dump_folder=None, show_time_cost=False, enable_metrics=False, smg_http_sidecar_port=None, enable_mfu_metrics=False, enable_metrics_for_all_schedulers=False, load_snapshot_publish_interval=15, tokenizer_metrics_custom_labels_header='x-custom-labels', tokenizer_metrics_allowed_custom_labels=None, extra_metric_labels=None, bucket_time_to_first_token=None, bucket_inter_token_latency=None, bucket_e2e_request_latency=None, prompt_tokens_buckets=None, generation_tokens_buckets=None, gc_warning_threshold_secs=0.0, decode_log_interval=40, enable_request_time_stats_logging=False, kv_events_config=None, enable_forward_pass_metrics=False, forward_pass_metrics_worker_id='', forward_pass_metrics_ipc_name=None, enable_trace=False, trace_modules='request', otlp_traces_endpoint='localhost:4317', export_metrics_to_file=False, export_metrics_to_file_dir=None, stat_loggers=None, constrained_json_whitespace_pattern=None, constrained_json_disable_any_whitespace=False, attention_backend='triton', decode_attention_backend=None, prefill_attention_backend=None, sampling_backend='flashinfer', grammar_backend='xgrammar', radix_cache_backend=None, mm_attention_backend=None, fp8_gemm_runner_backend='auto', fp4_gemm_runner_backend='auto', bf16_gemm_backend='auto', dsa_prefill_backend=None, dsa_decode_backend=None, dsa_paged_mqa_logits_backend='auto', dsa_topk_backend='sgl-kernel', disable_flashinfer_autotune=False, flashinfer_autotune_skip_ops=None, mamba_backend='triton', cuda_graph_config=CudaGraphConfig(decode=PhaseConfig(backend='full', max_bs=16, bs=[1, 2, 4, 8, 12, 16], tc_compiler='eager', full_prefill_max_req=None), prefill=PhaseConfig(backend='disabled', max_bs=2048, bs=[4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], tc_compiler='eager', full_prefill_max_req=None)), cuda_graph_backend_decode=None, cuda_graph_backend_prefill=None, cuda_graph_max_bs_decode=16, cuda_graph_max_bs_prefill=None, cuda_graph_bs_decode=None, cuda_graph_bs_prefill=None, cuda_graph_tc_compiler=None, disable_prefill_cuda_graph=False, disable_decode_cuda_graph=False, disable_cuda_graph=False, disable_cuda_graph_padding=False, enable_profile_cuda_graph=False, enable_cudagraph_gc=False, debug_cuda_graph=False, enable_layerwise_nvtx_marker=False, enable_nccl_nvls=False, enable_symm_mem=False, triton_attention_reduce_in_fp32=False, triton_attention_num_kv_splits=8, triton_attention_split_tile_size=None, flashinfer_mla_disable_ragged=False, enable_fused_qk_norm_rope=False, enable_precise_embedding_interpolation=False, enable_fused_moe_sum_all_reduce=False, enable_deepseek_v4_fp4_indexer=False, disable_custom_all_reduce=False, enable_mscclpp=False, enable_torch_symm_mem=False, enable_scattered_sconv=False, pre_warm_nccl=False, enable_quant_communications=False, enable_flashinfer_allreduce_fusion=False, enforce_disable_flashinfer_allreduce_fusion=False, flashinfer_allreduce_fusion_backend=None, enable_aiter_allreduce_fusion=False, enable_torch_compile=False, enable_torch_compile_debug_mode=False, torch_compile_max_bs=32, torchao_config='', speculative_algorithm=None, speculative_draft_model_path=None, speculative_draft_model_revision=None, speculative_draft_load_format=None, speculative_num_steps=None, speculative_eagle_topk=None, speculative_num_draft_tokens=None, speculative_dflash_block_size=None, speculative_dspark_block_size=None, speculative_dspark_sps_table_path=None, speculative_dspark_confidence_sts_path=None, speculative_dspark_align_verify_tokens_to_graph_tier=False, speculative_accept_threshold_single=1.0, speculative_accept_threshold_acc=1.0, speculative_use_rejection_sampling=False, speculative_token_map=None, speculative_attention_mode='prefill', speculative_draft_attention_backend=None, speculative_draft_window_size=None, speculative_moe_runner_backend='marlin', speculative_moe_a2a_backend=None, speculative_draft_model_quantization=None, speculative_skip_dp_mlp_sync=False, enable_multi_layer_eagle=False, speculative_adaptive=False, speculative_adaptive_config=None, decoupled_spec_bind_endpoint=None, decoupled_spec_connect_endpoints=None, decoupled_spec_rank=None, decoupled_spec_role='null', spec_trace_dir=None, speculative_ngram_min_bfs_breadth=1, speculative_ngram_max_bfs_breadth=10, speculative_ngram_match_type='BFS', speculative_ngram_max_trie_depth=18, speculative_ngram_capacity=10000000, speculative_ngram_external_corpus_path=None, speculative_ngram_external_sam_budget=0, speculative_ngram_external_corpus_max_tokens=10000000, ep_size=32, moe_a2a_backend='none', moe_runner_backend='marlin', flashinfer_mxfp4_moe_precision='default', deepep_mode='auto', fuseep_mode=2, deepep_dispatcher_output_dtype='auto', ep_num_redundant_experts=0, ep_dispatch_algorithm=None, init_expert_location='trivial', enable_eplb=False, eplb_algorithm='auto', eplb_rebalance_num_iterations=1000, eplb_rebalance_layers_per_chunk=None, eplb_min_rebalancing_utilization_threshold=1.0, expert_distribution_recorder_mode=None, expert_distribution_recorder_buffer_size=1000, enable_expert_distribution_metrics=False, deepep_config=None, moe_dense_tp_size=None, elastic_ep_backend=None, enable_elastic_expert_backup=False, mooncake_ib_device=None, enable_waterfill=False, ep_join_mode=None, ep_join_rank_offset=0, elastic_ep_initial_size=None, max_ep_size=None, elastic_ep_scale_timeout=600, elastic_ep_rejoin=False, disable_flashinfer_cutlass_moe_fp4_allgather=False, disable_shared_experts_fusion=False, enforce_shared_experts_fusion=False, max_mamba_cache_size=None, mamba_ssm_dtype=None, mamba_max_states_per_path=-1, enable_mamba_cache_stochastic_rounding=False, mamba_cache_philox_rounds=0, mamba_full_memory_ratio=0.21, mamba_radix_cache_strategy='extra_buffer_lazy', uses_mamba_radix_cache=False, mamba_track_interval=256, enable_int8_mamba_checkpoint=False, int8_mamba_ckpt_size=None, linear_attn_backend='triton', linear_attn_decode_backend=None, linear_attn_prefill_backend=None, linear_attn_verify_backend=None, enable_linear_replayssm=False, linear_replayssm_cache_len=16, enable_linear_replayssm_spec=False, enable_hierarchical_cache=False, hicache_ratio=2.0, hicache_size=0, hicache_write_policy='write_through', hicache_io_backend='kernel', hicache_mem_layout='page_first', hicache_storage_backend=None, hicache_storage_prefetch_policy='timeout', hicache_storage_backend_extra_config=None, enable_hisparse=False, hisparse_config=None, enable_broadcast_mm_inputs_process=False, enable_prefix_mm_cache=False, mm_enable_dp_encoder=False, mm_process_config={}, mm_processor_worker_num=0, mm_io_worker_num=0, limit_mm_data_per_request=None, enable_mm_global_cache=False, disable_fast_image_processor=False, mm_feature_transport='cpu', keep_mm_feature_on_device=False, enable_lora=None, enable_lora_overlap_loading=None, max_lora_rank=None, lora_target_modules=None, lora_paths=None, max_loaded_loras=None, max_loras_per_batch=8, lora_eviction_policy='lru', lora_backend='csgmv', max_lora_chunk_size=16, experts_shared_outer_loras=None, lora_use_virtual_experts=False, lora_strict_loading=False, lora_drain_wait_threshold=0.0, enable_two_batch_overlap=False, enable_single_batch_overlap=False, tbo_token_distribution_threshold=0.48, cpu_offload_gb=0, offload_group_size=-1, offload_num_in_group=1, offload_prefetch_step=1, offload_mode='cpu', enable_lmcache=False, lmcache_config_file=None, enable_flexkv=False, flexkv_config_file=None, kt_weight_path=None, kt_method='AMXINT4', kt_cpuinfer=None, kt_threadpool_count=2, kt_num_gpu_experts=None, kt_max_deferred_experts_per_token=None, dllm_algorithm=None, dllm_algorithm_config=None, dllm_fdfo=True, disaggregation_mode='decode', disaggregation_transfer_backend='mooncake', disaggregation_bootstrap_port=28800, disaggregation_ib_device='mlx5_0,mlx5_1,mlx5_2,mlx5_3', disaggregation_decode_enable_radix_cache=False, disaggregation_decode_enable_offload_kvcache=False, num_reserved_decode_tokens=512, disaggregation_decode_extra_slots=0, disaggregation_decode_polling_interval=1, optimistic_prefill_attempts=0, encoder_only=False, language_only=False, encoder_transfer_backend='zmq_to_tokenizer', encoder_urls=[], encoder_bootstrap_port=8997, encoder_register_urls=[], enable_adaptive_dispatch_to_encoder=False, enable_pdmux=False, pdmux_config_path=None, sm_group_num=8, custom_weight_loader=[], weight_loader_disable_mmap=False, weight_loader_prefetch_checkpoints=False, weight_loader_prefetch_num_threads=4, weight_loader_drop_cache_after_load=False, remote_instance_weight_loader_seed_instance_ip=None, remote_instance_weight_loader_seed_instance_service_port=None, remote_instance_weight_loader_send_weights_group_ports=None, remote_instance_weight_loader_backend='nccl', remote_instance_weight_loader_start_seed_via_transfer_engine=False, engine_info_bootstrap_port=6789, modelexpress_config=None, download_dir=None, model_checksum=None, delete_ckpt_after_loading=False, decrypted_config_file=None, decrypted_draft_config_file=None, checkpoint_engine_wait_weights_before_ready=False, enable_prefill_delayer=False, prefill_delayer_max_delay_passes=30, prefill_delayer_token_usage_low_watermark=None, prefill_delayer_forward_passes_buckets=None, prefill_delayer_wait_seconds_buckets=None, prefill_delayer_queue_min_ratio=None, prefill_delayer_max_delay_ms=None, min_free_slots_delay=None, enable_deterministic_inference=False, rl_on_policy_target=None, kv_canary='none', kv_canary_real_data='none', kv_canary_sweep_interval=0, enable_dynamic_batch_tokenizer=False, dynamic_batch_tokenizer_batch_size=32, dynamic_batch_tokenizer_batch_timeout=0.002, enable_tokenizer_batch_encode=False, disable_tokenizer_batch_decode=False, debug_tensor_dump_output_folder=None, debug_tensor_dump_layers=None, debug_tensor_dump_input_file=None, enable_memory_saver=False, enable_weights_cpu_backup=False, enable_draft_weights_cpu_backup=False, enable_custom_logit_processor=False, enable_return_hidden_states=False, enable_return_routed_experts=False, enable_return_indexer_topk=False, disable_outlines_disk_cache=False, enable_mis=False, weight_cache_mode='off', weight_cache_socket=None, weight_cache_timeout=1800, forward_hooks=None, msprobe_dump_config=None)
+[2026-08-27 04:36:14] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:14] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:14] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:14] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:14] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:14] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:14] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:14] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:17] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:17] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:18] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:19] Multimodal processor concurrency enabled with 2 isolated worker threads (auto).
+[2026-08-27 04:36:19] No HuggingFace chat template found
+[2026-08-27 04:36:19] No chat template found, defaulting to 'string' content format
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:34 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:34 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:34 TP4 EP4] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:34 TP4 EP4] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:34 TP6 EP6] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:34 TP6 EP6] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:34] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:34] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:34 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:34 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+/sgl-workspace/sglang/python/sglang/srt/utils/hf_transformers/tokenizer.py:423: UserWarning: Using a slow tokenizer. This might cause a significant slowdown. Consider using a fast tokenizer instead.
+  warnings.warn(
+[2026-08-27 04:36:34] Applying special tokens cache patch for Kimi tokenizer: <class 'transformers_modules.Kimi_hyphen_K3.caf40b2773f9029e.tokenization_kimi.TikTokenTokenizer'>
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:34 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:34 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:34 TP7 EP7] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:34 TP7 EP7] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:34 TP5 EP5] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:34 TP5 EP5] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:35 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:35 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:36 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:36 TP6 EP6] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:36 TP4 EP4] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:36 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:36 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:36 TP7 EP7] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:36 TP5 EP5] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:37 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:37.190651   236 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:37.190778   236 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:37.190796   236 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.5 port: 12001
+I0827 04:36:37.190856   236 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.5:16500
+I0827 04:36:37.191038   236 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:37.194572   241 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:37.194692   241 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:37.194712   241 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.5 port: 12001
+I0827 04:36:37.194767   241 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.5:15678
+I0827 04:36:37.194965   241 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:37.219420   236 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:37.223201   241 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:37.227941   236 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:37.231628   241 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:37.236233   236 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:37.239850   241 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:37.244647   236 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:37.248145   241 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:37.248418   236 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:37.248464   236 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:37.248525   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.250921   241 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:37.250947   241 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:37.251004   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.255973   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:37.257668   236 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:05
+I0827 04:36:37.257705   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.283210   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:37.283238   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:37.284873   236 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:05
+I0827 04:36:37.284902   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.284904   241 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:05
+I0827 04:36:37.284934   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.293713   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:37.295406   241 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:05
+I0827 04:36:37.295433   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.311285   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:37.312896   236 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:05
+I0827 04:36:37.312925   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.327399   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:37.328944   241 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:05
+I0827 04:36:37.328972   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.339579   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:37.341301   236 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:05
+I0827 04:36:37.341434   236 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:37 TP1 EP1] Init torch distributed begin.
+I0827 04:36:37.359290   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:37.360814   241 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:05
+I0827 04:36:37.360936   241 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:37 TP6 EP6] Init torch distributed begin.
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:37.366210   239 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:37.366365   239 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:37.366390   239 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.5 port: 12001
+I0827 04:36:37.366467   239 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.5:15787
+I0827 04:36:37.366704   239 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:37.395485   239 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:37.402652   239 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:37.409775   239 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:37.434130   239 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:37.436257   239 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:37.436282   239 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:37.436331   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.442425   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:37.443729   239 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:05
+I0827 04:36:37.443749   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.448421   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:37.449750   239 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:05
+I0827 04:36:37.449769   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.454622   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:37.455952   239 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:05
+I0827 04:36:37.455972   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.463356   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:37.464902   239 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:05
+I0827 04:36:37.465021   239 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:37 TP4 EP4] Init torch distributed begin.
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:37.510618   238 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:37.510744   238 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:37.510763   238 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.5 port: 12001
+I0827 04:36:37.510824   238 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.5:16205
+I0827 04:36:37.511003   238 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:37.524268   242 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:37.524391   242 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:37.524411   242 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.5 port: 12001
+I0827 04:36:37.524524   242 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.5:16605
+I0827 04:36:37.524709   242 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:37.528693   235 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:37.528817   235 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:37.528836   235 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.5 port: 12001
+I0827 04:36:37.528893   235 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.5:16672
+I0827 04:36:37.529069   235 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:37.535192   238 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:37.535425   235 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:37.541376   235 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:37.542625   238 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:37.547586   235 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:37.549854   238 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:37.551366   242 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:37.553905   235 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:37.556406   235 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:37.556468   235 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:37.556523   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.558794   242 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:37.563313   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:37.566102   242 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:37.573271   242 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:37.574895   238 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:37.576385   242 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:37.576412   242 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:37.576478   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.577947   238 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:37.577976   238 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:37.578033   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.583841   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:37.585472   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:37.585623   242 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:05
+I0827 04:36:37.585656   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.587072   238 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:05
+I0827 04:36:37.587105   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.594069   235 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:05
+I0827 04:36:37.594128   235 rdma_context.cpp:205] Using SIEVE endpoint store
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:37.598918   240 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:37.599033   240 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:37.599050   240 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.5 port: 12001
+I0827 04:36:37.599107   240 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.5:15428
+I0827 04:36:37.599270   240 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:37.605341   240 topology.cpp:127] Device mlx5_0 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:37.606741   237 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:37.606841   237 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:37.606855   237 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.5 port: 12001
+I0827 04:36:37.606901   237 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.5:16531
+I0827 04:36:37.607028   237 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:37.611215   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:37.611989   237 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:37.612797   242 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:05
+I0827 04:36:37.612828   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.615110   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:37.616685   238 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:05
+I0827 04:36:37.616714   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.619258   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:37.620560   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:37.620851   235 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:05
+I0827 04:36:37.620885   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.622239   242 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:05
+I0827 04:36:37.622268   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.624459   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:37.625998   238 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:05
+I0827 04:36:37.626025   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.630085   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:37.630686   240 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:37.631680   242 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:05
+I0827 04:36:37.631806   242 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:37 TP7 EP7] Init torch distributed begin.
+I0827 04:36:37.633723   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:37.635268   238 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:05
+I0827 04:36:37.635390   238 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:37 TP3 EP3] Init torch distributed begin.
+I0827 04:36:37.642561   237 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:37.649834   237 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:37.651238   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:37.652786   235 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:05
+I0827 04:36:37.652813   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.657148   237 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:37.658787   240 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:37.659991   237 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:37.660017   237 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:37.660071   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.667523   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:37.669195   240 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:37.669226   237 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:05
+I0827 04:36:37.669257   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.672511   240 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:37.672547   240 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:37.672606   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.677067   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:37.678661   237 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:05
+I0827 04:36:37.678689   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.682178   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:37.683274   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:37.683740   240 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:05
+I0827 04:36:37.683773   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.684834   235 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:05
+I0827 04:36:37.684954   235 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:37 TP0 EP0] Init torch distributed begin.
+I0827 04:36:37.703369   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:37.704975   237 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:05
+I0827 04:36:37.705003   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.711591   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:37.713003   240 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:05
+I0827 04:36:37.713030   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.735574   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:37.737313   237 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:05
+I0827 04:36:37.737439   237 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:37 TP2 EP2] Init torch distributed begin.
+I0827 04:36:37.740485   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:37.741588   240 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:05
+I0827 04:36:37.741605   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.766494   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:37.767982   240 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:05
+I0827 04:36:37.768137   240 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:37 TP5 EP5] Init torch distributed begin.
+[2026-08-27 04:36:46 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:36:48 TP0 EP0] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP1 EP1] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP2 EP2] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP5 EP5] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP4 EP4] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP3 EP3] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP6 EP6] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP7 EP7] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP0 EP0] Init torch distributed ends. elapsed=10.55 s, mem usage=0.40 GB
+[2026-08-27 04:36:48 TP7 EP7] Init torch distributed ends. elapsed=10.61 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP6 EP6] Init torch distributed ends. elapsed=10.88 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP4 EP4] Init torch distributed ends. elapsed=10.77 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP3 EP3] Init torch distributed ends. elapsed=10.60 s, mem usage=0.40 GB
+[2026-08-27 04:36:48 TP1 EP1] Init torch distributed ends. elapsed=10.90 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP5 EP5] Init torch distributed ends. elapsed=10.47 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP2 EP2] Init torch distributed ends. elapsed=10.50 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP7 EP7] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP4 EP4] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP1 EP1] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP6 EP6] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP7 EP7] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP4 EP4] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP5 EP5] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP6 EP6] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP3 EP3] Load weight begin. avail mem=82.19 GB
+[2026-08-27 04:36:48 TP0 EP0] Load weight begin. avail mem=82.19 GB
+[2026-08-27 04:36:48 TP5 EP5] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP2 EP2] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP4 EP4] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP7 EP7] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP6 EP6] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP4 EP4] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP7 EP7] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP6 EP6] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP5 EP5] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP5 EP5] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:50 TP1 EP1] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP0 EP0] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP2 EP2] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP3 EP3] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP5 EP5] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP4 EP4] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP6 EP6] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP7 EP7] multimem all-gather disabled because the TP group spans across nodes.
+
Multi-thread loading shards:   0% Completed | 0/96 [00:00<?, ?it/s]
Multi-thread loading shards:   1% Completed | 1/96 [00:00<00:24,  3.85it/s]
Multi-thread loading shards:   2% Completed | 2/96 [00:07<07:13,  4.61s/it]
Multi-thread loading shards:   3% Completed | 3/96 [00:09<04:44,  3.06s/it]
Multi-thread loading shards:   4% Completed | 4/96 [00:10<03:33,  2.32s/it]
Multi-thread loading shards:   5% Completed | 5/96 [00:11<02:54,  1.92s/it]
Multi-thread loading shards:   6% Completed | 6/96 [00:12<02:29,  1.66s/it]
Multi-thread loading shards:   7% Completed | 7/96 [00:13<02:13,  1.50s/it]
Multi-thread loading shards:   8% Completed | 8/96 [00:15<02:04,  1.41s/it]
Multi-thread loading shards:   9% Completed | 9/96 [00:16<01:56,  1.34s/it]
Multi-thread loading shards:  10% Completed | 10/96 [00:17<01:47,  1.25s/it]
Multi-thread loading shards:  11% Completed | 11/96 [00:18<01:44,  1.23s/it]
Multi-thread loading shards:  12% Completed | 12/96 [00:19<01:43,  1.23s/it]
Multi-thread loading shards:  14% Completed | 13/96 [00:21<01:59,  1.44s/it]
Multi-thread loading shards:  15% Completed | 14/96 [00:22<01:51,  1.36s/it]
Multi-thread loading shards:  16% Completed | 15/96 [00:24<01:45,  1.30s/it]
Multi-thread loading shards:  17% Completed | 16/96 [00:25<01:42,  1.28s/it]
Multi-thread loading shards:  18% Completed | 17/96 [00:26<01:39,  1.26s/it]
Multi-thread loading shards:  19% Completed | 18/96 [00:27<01:36,  1.24s/it]
Multi-thread loading shards:  20% Completed | 19/96 [00:28<01:32,  1.20s/it]
Multi-thread loading shards:  21% Completed | 20/96 [00:29<01:30,  1.19s/it]
Multi-thread loading shards:  22% Completed | 21/96 [00:31<01:29,  1.19s/it]
Multi-thread loading shards:  23% Completed | 22/96 [00:32<01:27,  1.18s/it]
Multi-thread loading shards:  24% Completed | 23/96 [00:33<01:26,  1.18s/it]
Multi-thread loading shards:  25% Completed | 24/96 [00:34<01:23,  1.16s/it]
Multi-thread loading shards:  26% Completed | 25/96 [00:35<01:20,  1.14s/it]
Multi-thread loading shards:  27% Completed | 26/96 [00:36<01:18,  1.13s/it]
Multi-thread loading shards:  28% Completed | 27/96 [00:37<01:19,  1.15s/it]
Multi-thread loading shards:  29% Completed | 28/96 [00:39<01:18,  1.16s/it]
Multi-thread loading shards:  30% Completed | 29/96 [00:40<01:17,  1.16s/it]
Multi-thread loading shards:  31% Completed | 30/96 [00:41<01:17,  1.18s/it]
Multi-thread loading shards:  32% Completed | 31/96 [00:42<01:16,  1.18s/it]
Multi-thread loading shards:  33% Completed | 32/96 [00:44<01:30,  1.42s/it]
Multi-thread loading shards:  34% Completed | 33/96 [00:45<01:23,  1.33s/it]
Multi-thread loading shards:  35% Completed | 34/96 [00:46<01:17,  1.26s/it]
Multi-thread loading shards:  36% Completed | 35/96 [00:48<01:14,  1.23s/it]
Multi-thread loading shards:  38% Completed | 36/96 [00:49<01:13,  1.22s/it]
Multi-thread loading shards:  39% Completed | 37/96 [00:50<01:11,  1.21s/it]
Multi-thread loading shards:  40% Completed | 38/96 [00:51<01:09,  1.20s/it]
Multi-thread loading shards:  41% Completed | 39/96 [00:52<01:07,  1.18s/it]
Multi-thread loading shards:  42% Completed | 40/96 [00:53<01:05,  1.18s/it]
Multi-thread loading shards:  43% Completed | 41/96 [00:55<01:04,  1.17s/it]
Multi-thread loading shards:  44% Completed | 42/96 [00:56<01:02,  1.17s/it]
Multi-thread loading shards:  45% Completed | 43/96 [00:57<01:03,  1.19s/it]
Multi-thread loading shards:  46% Completed | 44/96 [00:58<01:01,  1.18s/it]
Multi-thread loading shards:  47% Completed | 45/96 [00:59<00:57,  1.13s/it]
Multi-thread loading shards:  48% Completed | 46/96 [01:00<00:57,  1.15s/it]
Multi-thread loading shards:  49% Completed | 47/96 [01:02<00:57,  1.17s/it]
Multi-thread loading shards:  50% Completed | 48/96 [01:03<00:55,  1.15s/it]
Multi-thread loading shards:  51% Completed | 49/96 [01:04<00:53,  1.13s/it]
Multi-thread loading shards:  52% Completed | 50/96 [01:06<01:01,  1.34s/it]
Multi-thread loading shards:  53% Completed | 51/96 [01:07<00:58,  1.29s/it]
Multi-thread loading shards:  54% Completed | 52/96 [01:08<00:55,  1.27s/it]
Multi-thread loading shards:  55% Completed | 53/96 [01:09<00:53,  1.24s/it]
Multi-thread loading shards:  56% Completed | 54/96 [01:10<00:50,  1.21s/it]
Multi-thread loading shards:  57% Completed | 55/96 [01:12<00:49,  1.22s/it]
Multi-thread loading shards:  58% Completed | 56/96 [01:13<00:48,  1.22s/it]
Multi-thread loading shards:  59% Completed | 57/96 [01:14<00:47,  1.21s/it]
Multi-thread loading shards:  60% Completed | 58/96 [01:15<00:45,  1.20s/it]
Multi-thread loading shards:  61% Completed | 59/96 [01:16<00:44,  1.19s/it]
Multi-thread loading shards:  62% Completed | 60/96 [01:17<00:43,  1.20s/it]
Multi-thread loading shards:  64% Completed | 61/96 [01:19<00:42,  1.21s/it]
Multi-thread loading shards:  65% Completed | 62/96 [01:20<00:40,  1.19s/it]
Multi-thread loading shards:  66% Completed | 63/96 [01:21<00:40,  1.21s/it]
Multi-thread loading shards:  67% Completed | 64/96 [01:22<00:38,  1.20s/it]
Multi-thread loading shards:  68% Completed | 65/96 [01:24<00:37,  1.20s/it]
Multi-thread loading shards:  69% Completed | 66/96 [01:25<00:36,  1.20s/it]
Multi-thread loading shards:  70% Completed | 67/96 [01:26<00:35,  1.22s/it]
Multi-thread loading shards:  71% Completed | 68/96 [01:28<00:41,  1.48s/it]
Multi-thread loading shards:  72% Completed | 69/96 [01:29<00:37,  1.39s/it]
Multi-thread loading shards:  73% Completed | 70/96 [01:30<00:34,  1.34s/it]
Multi-thread loading shards:  74% Completed | 71/96 [01:32<00:32,  1.29s/it]
Multi-thread loading shards:  75% Completed | 72/96 [01:33<00:31,  1.30s/it]
Multi-thread loading shards:  76% Completed | 73/96 [01:34<00:29,  1.29s/it]
Multi-thread loading shards:  77% Completed | 74/96 [01:35<00:27,  1.27s/it]
Multi-thread loading shards:  78% Completed | 75/96 [01:37<00:26,  1.26s/it]
Multi-thread loading shards:  79% Completed | 76/96 [01:38<00:24,  1.24s/it]
Multi-thread loading shards:  80% Completed | 77/96 [01:39<00:23,  1.23s/it]
Multi-thread loading shards:  81% Completed | 78/96 [01:40<00:21,  1.19s/it]
Multi-thread loading shards:  82% Completed | 79/96 [01:41<00:20,  1.20s/it]
Multi-thread loading shards:  83% Completed | 80/96 [01:43<00:19,  1.20s/it]
Multi-thread loading shards:  84% Completed | 81/96 [01:44<00:18,  1.21s/it]
Multi-thread loading shards:  85% Completed | 82/96 [01:45<00:16,  1.20s/it]
Multi-thread loading shards:  86% Completed | 83/96 [01:46<00:15,  1.17s/it]
Multi-thread loading shards:  88% Completed | 84/96 [01:47<00:14,  1.17s/it]
Multi-thread loading shards:  89% Completed | 85/96 [01:48<00:12,  1.10s/it]
Multi-thread loading shards:  90% Completed | 86/96 [01:49<00:10,  1.05s/it]
Multi-thread loading shards:  91% Completed | 87/96 [01:50<00:09,  1.02s/it]
Multi-thread loading shards:  92% Completed | 88/96 [01:51<00:07,  1.01it/s]
Multi-thread loading shards:  93% Completed | 89/96 [01:52<00:06,  1.03it/s]
Multi-thread loading shards:  94% Completed | 90/96 [01:53<00:05,  1.04it/s]
Multi-thread loading shards:  95% Completed | 91/96 [01:54<00:04,  1.06it/s]
Multi-thread loading shards:  96% Completed | 92/96 [01:55<00:03,  1.06it/s]
Multi-thread loading shards:  97% Completed | 93/96 [01:56<00:02,  1.08it/s]
Multi-thread loading shards:  98% Completed | 94/96 [01:56<00:01,  1.46it/s]
Multi-thread loading shards: 100% Completed | 96/96 [01:56<00:00,  2.28it/s]
Multi-thread loading shards: 100% Completed | 96/96 [01:56<00:00,  1.21s/it]
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:58 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP6 EP6] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP0 EP0] Load weight end. elapsed=130.28 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP6 EP6] Load weight end. elapsed=130.29 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP3 EP3] Load weight end. elapsed=130.28 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP2 EP2] Load weight end. elapsed=130.32 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP5 EP5] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP5 EP5] Load weight end. elapsed=130.38 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP4 EP4] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP4 EP4] Load weight end. elapsed=130.48 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP7 EP7] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP7 EP7] Load weight end. elapsed=130.48 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:39:01 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:39:01 TP1 EP1] Load weight end. elapsed=132.81 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:39:01 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP6 EP6] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP6 EP6] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP5 EP5] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP5 EP5] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP7 EP7] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP7 EP7] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP4 EP4] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP4 EP4] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:02 TP0 EP0] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP7 EP7] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP6 EP6] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP5 EP5] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP1 EP1] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP4 EP4] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP3 EP3] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP2 EP2] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP7 EP7] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP4 EP4] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP6 EP6] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP5 EP5] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP0 EP0] Memory pool end. avail mem=9.18 GB
+[2026-08-27 04:39:02 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP1 EP1] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP6 EP6] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP5 EP5] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP7 EP7] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP6 EP6] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP5 EP5] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP7 EP7] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP4 EP4] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP4 EP4] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP2 EP2] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP3 EP3] Memory pool end. avail mem=9.18 GB
+[2026-08-27 04:39:02 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:39:02 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:39:02 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP0 EP0] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.17 GB
+[2026-08-27 04:39:02 TP6 EP6] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP6 EP6] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP1 EP1] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP5 EP5] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP7 EP7] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP5 EP5] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP7 EP7] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:39:02 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP3 EP3] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.17 GB
+[2026-08-27 04:39:02 TP4 EP4] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP4 EP4] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:39:02 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP2 EP2] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+
  0%|          | 0/6 [00:00<?, ?it/s]
Capturing batches (bs=16 avail_mem=9.00 GB):   0%|          | 0/6 [00:00<?, ?it/s]
Capturing batches (bs=16 avail_mem=9.00 GB):  17%|█▋        | 1/6 [03:43<18:38, 223.66s/it]
Capturing batches (bs=12 avail_mem=8.84 GB):  17%|█▋        | 1/6 [03:43<18:38, 223.66s/it]
Capturing batches (bs=12 avail_mem=8.84 GB):  33%|███▎      | 2/6 [03:44<06:11, 92.84s/it] 
Capturing batches (bs=8 avail_mem=8.78 GB):  33%|███▎      | 2/6 [03:44<06:11, 92.84s/it] 
Capturing batches (bs=8 avail_mem=8.78 GB):  50%|█████     | 3/6 [03:49<02:36, 52.32s/it]
Capturing batches (bs=4 avail_mem=8.73 GB):  50%|█████     | 3/6 [03:49<02:36, 52.32s/it]
Capturing batches (bs=4 avail_mem=8.73 GB):  67%|██████▋   | 4/6 [03:49<01:03, 32.00s/it]
Capturing batches (bs=2 avail_mem=8.68 GB):  67%|██████▋   | 4/6 [03:49<01:03, 32.00s/it]
Capturing batches (bs=2 avail_mem=8.68 GB):  83%|████████▎ | 5/6 [03:50<00:20, 20.77s/it]
Capturing batches (bs=1 avail_mem=8.62 GB):  83%|████████▎ | 5/6 [03:50<00:20, 20.77s/it]
Capturing batches (bs=1 avail_mem=8.62 GB): 100%|██████████| 6/6 [03:52<00:00, 14.36s/it]
Capturing batches (bs=1 avail_mem=8.62 GB): 100%|██████████| 6/6 [03:52<00:00, 38.78s/it]
+[2026-08-27 04:42:56 TP4 EP4] Capture target decode CUDA graph end. elapsed=233.75 s, mem usage=0.60 GB, avail mem=8.52 GB.
+[2026-08-27 04:42:56 TP4 EP4] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP3 EP3] Capture target decode CUDA graph end. elapsed=233.75 s, mem usage=0.62 GB, avail mem=8.55 GB.
+[2026-08-27 04:42:56 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP0 EP0] Capture target decode CUDA graph end. elapsed=233.77 s, mem usage=0.60 GB, avail mem=8.56 GB.
+[2026-08-27 04:42:56 TP0 EP0] max_total_num_tokens=440256, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=181, context_len=1048576, available_gpu_mem=8.56 GB
+[2026-08-27 04:42:56 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP7 EP7] Capture target decode CUDA graph end. elapsed=233.77 s, mem usage=0.62 GB, avail mem=8.51 GB.
+[2026-08-27 04:42:56 TP7 EP7] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP6 EP6] Capture target decode CUDA graph end. elapsed=233.86 s, mem usage=0.56 GB, avail mem=8.59 GB.
+[2026-08-27 04:42:56 TP6 EP6] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP1 EP1] Capture target decode CUDA graph end. elapsed=233.86 s, mem usage=0.58 GB, avail mem=8.58 GB.
+[2026-08-27 04:42:56 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP2 EP2] Capture target decode CUDA graph end. elapsed=233.86 s, mem usage=0.56 GB, avail mem=8.59 GB.
+[2026-08-27 04:42:56 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP5 EP5] Capture target decode CUDA graph end. elapsed=233.90 s, mem usage=0.58 GB, avail mem=8.58 GB.
+[2026-08-27 04:42:56 TP5 EP5] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+W0827 04:42:57.550421  2569 memory_location.cpp:73] Failed to get NUMA node, addr: 0x599a0800, len: 23168: Operation not permitted [1]
+W0827 04:42:57.550521  2570 memory_location.cpp:73] Failed to get NUMA node, addr: 0x599cb380, len: 23168: Operation not permitted [1]
+W0827 04:42:57.550542  2571 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59e1b540, len: 23168: Operation not permitted [1]
+W0827 04:42:57.550628  2572 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59c8a2c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.550725  2573 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5be7a240, len: 185344: Operation not permitted [1]
+W0827 04:42:57.550769  2574 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5bea7680, len: 185344: Operation not permitted [1]
+W0827 04:42:57.550900  2575 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59cc48c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.551011  2576 memory_location.cpp:73] Failed to get NUMA node, addr: 0x45f39fc0, len: 46336: Operation not permitted [1]
+W0827 04:42:57.551070  2577 memory_location.cpp:73] Failed to get NUMA node, addr: 0x599bfbc0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.551105  2578 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59c69c80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.580384  2676 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59c988c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.580493  2678 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59a30c40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.580636  2677 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59d06340, len: 23168: Operation not permitted [1]
+W0827 04:42:57.580636  2679 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5bd91640, len: 185344: Operation not permitted [1]
+W0827 04:42:57.580685  2675 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59e30d80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.580811  2681 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59f53440, len: 23168: Operation not permitted [1]
+W0827 04:42:57.580950  2680 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5bdbea80, len: 185344: Operation not permitted [1]
+W0827 04:42:57.581368  2686 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59dca000, len: 23168: Operation not permitted [1]
+W0827 04:42:57.581421  2682 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5987fc00, len: 46336: Operation not permitted [1]
+W0827 04:42:57.581741  2684 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59e10480, len: 23168: Operation not permitted [1]
+W0827 04:42:57.597819  2845 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59224900, len: 23168: Operation not permitted [1]
+W0827 04:42:57.598558  2839 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5917e740, len: 23168: Operation not permitted [1]
+W0827 04:42:57.598562  2849 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58e6ce00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.598762  2838 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58e982c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.598834  2841 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5b32ec00, len: 185344: Operation not permitted [1]
+W0827 04:42:57.598897  2834 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5927b680, len: 23168: Operation not permitted [1]
+W0827 04:42:57.598922  2843 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5b35c040, len: 185344: Operation not permitted [1]
+W0827 04:42:57.598953  2836 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5939fc40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.599097  2850 memory_location.cpp:73] Failed to get NUMA node, addr: 0x592d3fc0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.599287  2847 memory_location.cpp:73] Failed to get NUMA node, addr: 0x453f4c40, len: 46336: Operation not permitted [1]
+W0827 04:42:57.612057  2918 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59f4a140, len: 23168: Operation not permitted [1]
+W0827 04:42:57.612175  2919 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59f098c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.612243  2920 memory_location.cpp:73] Failed to get NUMA node, addr: 0x46045dc0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.612710  2925 memory_location.cpp:73] Failed to get NUMA node, addr: 0x46106480, len: 46336: Operation not permitted [1]
+W0827 04:42:57.612715  2924 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59a7ab40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.612880  2922 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5bf25300, len: 185344: Operation not permitted [1]
+W0827 04:42:57.612898  2927 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59e8a9c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.612931  2921 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59fec840, len: 23168: Operation not permitted [1]
+W0827 04:42:57.612948  2923 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5bf52740, len: 185344: Operation not permitted [1]
+W0827 04:42:57.612996  2926 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59edf040, len: 23168: Operation not permitted [1]
+W0827 04:42:57.618498  2975 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a4cda80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.618582  2974 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a3878c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.618582  2973 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a3abc80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.618669  2976 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5c40da00, len: 185344: Operation not permitted [1]
+W0827 04:42:57.618669  2978 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59f66a40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.618741  2977 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5c43ae40, len: 185344: Operation not permitted [1]
+W0827 04:42:57.618854  2979 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59ce6c00, len: 46336: Operation not permitted [1]
+W0827 04:42:57.618898  2980 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a2ebe80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.619125  2982 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a37d0c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.632592  2972 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a371340, len: 23168: Operation not permitted [1]
+W0827 04:42:57.776260  3310 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58ee7080, len: 23168: Operation not permitted [1]
+W0827 04:42:57.776290  3311 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5854f480, len: 23168: Operation not permitted [1]
+W0827 04:42:57.776294  3312 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58edd900, len: 23168: Operation not permitted [1]
+W0827 04:42:57.776240  3309 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58f70e40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.776324  3313 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a63db40, len: 185344: Operation not permitted [1]
+W0827 04:42:57.776337  3314 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a66af80, len: 185344: Operation not permitted [1]
+W0827 04:42:57.776417  3315 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58e5ec40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.776444  3316 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58109800, len: 46336: Operation not permitted [1]
+W0827 04:42:57.776451  3317 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58ed2cc0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.776649  3318 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58316800, len: 23168: Operation not permitted [1]
+W0827 04:42:57.812515  3457 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58919a80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.812533  3458 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5892c840, len: 23168: Operation not permitted [1]
+W0827 04:42:57.812597  3460 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58063380, len: 23168: Operation not permitted [1]
+W0827 04:42:57.812603  3459 memory_location.cpp:73] Failed to get NUMA node, addr: 0x588ed280, len: 23168: Operation not permitted [1]
+W0827 04:42:57.812654  3461 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59fc0080, len: 185344: Operation not permitted [1]
+W0827 04:42:57.812734  3462 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59fed4c0, len: 185344: Operation not permitted [1]
+W0827 04:42:57.812816  3463 memory_location.cpp:73] Failed to get NUMA node, addr: 0x57ebf300, len: 23168: Operation not permitted [1]
+W0827 04:42:57.812877  3464 memory_location.cpp:73] Failed to get NUMA node, addr: 0x44a36bc0, len: 46336: Operation not permitted [1]
+W0827 04:42:57.812991  3466 memory_location.cpp:73] Failed to get NUMA node, addr: 0x44982040, len: 23168: Operation not permitted [1]
+W0827 04:42:57.812996  3465 memory_location.cpp:73] Failed to get NUMA node, addr: 0x588ba900, len: 23168: Operation not permitted [1]
+W0827 04:42:57.851508  3605 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5888adc0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.851550  3606 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58775700, len: 23168: Operation not permitted [1]
+W0827 04:42:57.851619  3607 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58aca800, len: 23168: Operation not permitted [1]
+W0827 04:42:57.851691  3608 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5886ea80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.851799  3609 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5aaacb40, len: 185344: Operation not permitted [1]
+W0827 04:42:57.851871  3611 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58ad7b00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.851959  3610 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5aad9f80, len: 185344: Operation not permitted [1]
+W0827 04:42:57.851996  3612 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58aa8840, len: 46336: Operation not permitted [1]
+W0827 04:42:57.852005  3614 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58ca01c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.852105  3613 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5879a9c0, len: 23168: Operation not permitted [1]
+[2026-08-27 04:43:00] INFO:     Started server process [1]
+[2026-08-27 04:43:00] INFO:     Waiting for application startup.
+[2026-08-27 04:43:00] Calling super().encode with {'add_special_tokens': False}
+[2026-08-27 04:43:00] Calling super().encode with {'add_special_tokens': False}
+[2026-08-27 04:43:00] INFO:     Application startup complete.
+[2026-08-27 04:43:00] INFO:     Uvicorn running on http://0.0.0.0:30000 (Press CTRL+C to quit)
+[2026-08-27 04:43:01] INFO:     174.1.60.1:55082 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:43:01] INFO:     127.0.0.1:38530 - "GET /model_info HTTP/1.1" 200 OK
+[2026-08-27 04:43:02] INFO:     174.1.60.1:55098 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:43:03] INFO:     127.0.0.1:38536 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 04:43:04] INFO:     174.1.60.1:55110 - "GET /health HTTP/1.1" 200 OK
+[2026-08-27 04:43:01] Start of pd disaggregation warmup ...
+[2026-08-27 04:43:03] Disaggregation warmup requests completed for all 1 DP ranks
+[2026-08-27 04:43:03] End of disaggregation warmup
+[2026-08-27 04:43:03] The server is fired up and ready to roll!
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/ready_d_node1.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/ready_d_node1.log
new file mode 100644
index 0000000..2e1c104
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/ready_d_node1.log
@@ -0,0 +1,498 @@
+/usr/local/lib/python3.12/dist-packages/torchao/quantization/quant_api.py:1731: SyntaxWarning: invalid escape sequence '\.'
+  """Configuration class for applying different quantization configs to modules or parameters based on their fully qualified names (FQNs).
+/sgl-workspace/sglang/python/sglang/launch_server.py:56: UserWarning: 'python -m sglang.launch_server' is still supported, but 'sglang serve' is the recommended entrypoint.
+  Example: sglang serve --model-path <model> [options]
+  warnings.warn(
+[2026-08-27 04:35:58] KV cache is forced as chunk cache for decode server
+[2026-08-27 04:36:02] Breakable CUDA graph is incompatible with MLA attention; disabling prefill CUDA graph.
+[2026-08-27 04:36:02] Attention backend not specified. Use triton backend by default.
+[2026-08-27 04:36:05] server_args=ServerArgs(model_path='/data/hf_models/Kimi-K3', tokenizer_path='/data/hf_models/Kimi-K3', tokenizer_mode='auto', tokenizer_backend='huggingface', tokenizer_worker_num=1, detokenizer_worker_num=1, skip_tokenizer_init=False, load_format='auto', model_loader_extra_config='{}', trust_remote_code=True, context_length=None, is_embedding=False, enable_multimodal=None, revision=None, model_impl='auto', model_config_parser='auto', json_model_override_args='{}', dtype='auto', quantization=None, quantization_param_path=None, kv_cache_dtype='auto', enable_fp32_lm_head=False, modelopt_quant=None, modelopt_checkpoint_restore_path=None, modelopt_checkpoint_save_path=None, modelopt_export_path=None, quantize_and_serve=False, rl_quant_profile=None, enable_tf32_matmul=False, mem_fraction_static=0.88, max_running_requests=None, max_queued_requests=None, max_total_tokens=None, chunked_prefill_size=8192, enable_dynamic_chunking=False, max_prefill_tokens=16384, prefill_max_requests=None, schedule_policy='fcfs', enable_priority_scheduling=False, disable_priority_preemption=False, default_priority_value=None, abort_on_priority_when_disabled=False, schedule_low_priority_values_first=False, priority_scheduling_preemption_threshold=10, retraction_policy='length', schedule_conservativeness=1.0, page_size=64, swa_full_tokens_ratio=0.8, disable_hybrid_swa_memory=False, radix_eviction_policy='lru', prefill_only_disable_kv_cache=False, disable_radix_cache=True, enable_page_major_kv_layout=False, enable_unified_memory=False, disable_chunked_prefix_cache=False, disable_overlap_schedule=False, num_continuous_decode_steps=1, scheduler_recv_interval=1, enable_mixed_chunk=False, nccl_port=None, dist_timeout=3600, dist_init_addr='174.1.60.5:20000', nnodes=4, node_rank=1, tp_size=32, dcp_size=1, pp_size=1, pp_max_micro_batch_size=None, pp_async_batch_depth=0, dp_size=1, load_balance_method='round_robin', attn_cp_size=1, moe_dp_size=1, dcp_comm_backend='ag_rs', dcp_replicate_q_proj=None, dwdp_size=1, enable_prefill_cp=False, cp_strategy=None, enable_dsa_cache_layer_split=False, enable_dsa_prefill_context_parallel=False, dsa_prefill_cp_mode='round-robin-split', enable_prefill_context_parallel=False, prefill_cp_mode='in-seq-split', enable_cp_decode_attn_tp=False, enable_dp_attention=False, enable_dp_attention_local_control_broadcast=False, enable_dp_lm_head=False, enable_attn_tp_input_scattered=False, disable_attn_tp_gather=False, enable_p2p_check=False, device='cuda', base_gpu_id=0, gpu_id_step=1, random_seed=434112381, watchdog_timeout=300, soft_watchdog_timeout=None, sleep_on_idle=False, use_ray=False, custom_sigquit_handler=None, numa_node=None, gc_threshold=None, host='0.0.0.0', port=30000, fastapi_root_path='', smg_grpc_mode=False, grpc_mode=False, grpc_port=None, sidecar=None, sidecar_args=None, skip_server_warmup=False, warmups=None, enable_http2=False, ssl_keyfile=None, ssl_certfile=None, ssl_ca_certs=None, ssl_keyfile_password=None, enable_ssl_refresh=False, api_key=None, admin_api_key=None, served_model_name='kimi-k3', weight_version='default', chat_template=None, hf_chat_template_name=None, completion_template=None, file_storage_path='sglang_storage', enable_cache_report=False, reasoning_parser=None, default_chat_template_kwargs=None, strip_thinking_cache=False, enable_strict_thinking=False, tool_call_parser=None, tool_server=None, sampling_defaults='model', asr_max_buffer_seconds=60, asr_max_concurrent_sessions=32, preferred_sampling_params=None, allow_auto_truncate=False, stream_interval=1, batch_notify_size=16, stream_response_default_include_usage=False, incremental_streaming_output=False, enable_streaming_session=False, enable_session_radix_cache=False, log_level='info', log_level_http=None, log_requests=False, log_requests_level=2, log_requests_format='text', log_requests_target=None, uvicorn_access_log_exclude_prefixes=[], crash_dump_folder=None, show_time_cost=False, enable_metrics=False, smg_http_sidecar_port=None, enable_mfu_metrics=False, enable_metrics_for_all_schedulers=False, load_snapshot_publish_interval=15, tokenizer_metrics_custom_labels_header='x-custom-labels', tokenizer_metrics_allowed_custom_labels=None, extra_metric_labels=None, bucket_time_to_first_token=None, bucket_inter_token_latency=None, bucket_e2e_request_latency=None, prompt_tokens_buckets=None, generation_tokens_buckets=None, gc_warning_threshold_secs=0.0, decode_log_interval=40, enable_request_time_stats_logging=False, kv_events_config=None, enable_forward_pass_metrics=False, forward_pass_metrics_worker_id='', forward_pass_metrics_ipc_name=None, enable_trace=False, trace_modules='request', otlp_traces_endpoint='localhost:4317', export_metrics_to_file=False, export_metrics_to_file_dir=None, stat_loggers=None, constrained_json_whitespace_pattern=None, constrained_json_disable_any_whitespace=False, attention_backend='triton', decode_attention_backend=None, prefill_attention_backend=None, sampling_backend='flashinfer', grammar_backend='xgrammar', radix_cache_backend=None, mm_attention_backend=None, fp8_gemm_runner_backend='auto', fp4_gemm_runner_backend='auto', bf16_gemm_backend='auto', dsa_prefill_backend=None, dsa_decode_backend=None, dsa_paged_mqa_logits_backend='auto', dsa_topk_backend='sgl-kernel', disable_flashinfer_autotune=False, flashinfer_autotune_skip_ops=None, mamba_backend='triton', cuda_graph_config=CudaGraphConfig(decode=PhaseConfig(backend='full', max_bs=16, bs=[1, 2, 4, 8, 12, 16], tc_compiler='eager', full_prefill_max_req=None), prefill=PhaseConfig(backend='disabled', max_bs=2048, bs=[4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], tc_compiler='eager', full_prefill_max_req=None)), cuda_graph_backend_decode=None, cuda_graph_backend_prefill=None, cuda_graph_max_bs_decode=16, cuda_graph_max_bs_prefill=None, cuda_graph_bs_decode=None, cuda_graph_bs_prefill=None, cuda_graph_tc_compiler=None, disable_prefill_cuda_graph=False, disable_decode_cuda_graph=False, disable_cuda_graph=False, disable_cuda_graph_padding=False, enable_profile_cuda_graph=False, enable_cudagraph_gc=False, debug_cuda_graph=False, enable_layerwise_nvtx_marker=False, enable_nccl_nvls=False, enable_symm_mem=False, triton_attention_reduce_in_fp32=False, triton_attention_num_kv_splits=8, triton_attention_split_tile_size=None, flashinfer_mla_disable_ragged=False, enable_fused_qk_norm_rope=False, enable_precise_embedding_interpolation=False, enable_fused_moe_sum_all_reduce=False, enable_deepseek_v4_fp4_indexer=False, disable_custom_all_reduce=False, enable_mscclpp=False, enable_torch_symm_mem=False, enable_scattered_sconv=False, pre_warm_nccl=False, enable_quant_communications=False, enable_flashinfer_allreduce_fusion=False, enforce_disable_flashinfer_allreduce_fusion=False, flashinfer_allreduce_fusion_backend=None, enable_aiter_allreduce_fusion=False, enable_torch_compile=False, enable_torch_compile_debug_mode=False, torch_compile_max_bs=32, torchao_config='', speculative_algorithm=None, speculative_draft_model_path=None, speculative_draft_model_revision=None, speculative_draft_load_format=None, speculative_num_steps=None, speculative_eagle_topk=None, speculative_num_draft_tokens=None, speculative_dflash_block_size=None, speculative_dspark_block_size=None, speculative_dspark_sps_table_path=None, speculative_dspark_confidence_sts_path=None, speculative_dspark_align_verify_tokens_to_graph_tier=False, speculative_accept_threshold_single=1.0, speculative_accept_threshold_acc=1.0, speculative_use_rejection_sampling=False, speculative_token_map=None, speculative_attention_mode='prefill', speculative_draft_attention_backend=None, speculative_draft_window_size=None, speculative_moe_runner_backend='marlin', speculative_moe_a2a_backend=None, speculative_draft_model_quantization=None, speculative_skip_dp_mlp_sync=False, enable_multi_layer_eagle=False, speculative_adaptive=False, speculative_adaptive_config=None, decoupled_spec_bind_endpoint=None, decoupled_spec_connect_endpoints=None, decoupled_spec_rank=None, decoupled_spec_role='null', spec_trace_dir=None, speculative_ngram_min_bfs_breadth=1, speculative_ngram_max_bfs_breadth=10, speculative_ngram_match_type='BFS', speculative_ngram_max_trie_depth=18, speculative_ngram_capacity=10000000, speculative_ngram_external_corpus_path=None, speculative_ngram_external_sam_budget=0, speculative_ngram_external_corpus_max_tokens=10000000, ep_size=32, moe_a2a_backend='none', moe_runner_backend='marlin', flashinfer_mxfp4_moe_precision='default', deepep_mode='auto', fuseep_mode=2, deepep_dispatcher_output_dtype='auto', ep_num_redundant_experts=0, ep_dispatch_algorithm=None, init_expert_location='trivial', enable_eplb=False, eplb_algorithm='auto', eplb_rebalance_num_iterations=1000, eplb_rebalance_layers_per_chunk=None, eplb_min_rebalancing_utilization_threshold=1.0, expert_distribution_recorder_mode=None, expert_distribution_recorder_buffer_size=1000, enable_expert_distribution_metrics=False, deepep_config=None, moe_dense_tp_size=None, elastic_ep_backend=None, enable_elastic_expert_backup=False, mooncake_ib_device=None, enable_waterfill=False, ep_join_mode=None, ep_join_rank_offset=0, elastic_ep_initial_size=None, max_ep_size=None, elastic_ep_scale_timeout=600, elastic_ep_rejoin=False, disable_flashinfer_cutlass_moe_fp4_allgather=False, disable_shared_experts_fusion=False, enforce_shared_experts_fusion=False, max_mamba_cache_size=None, mamba_ssm_dtype=None, mamba_max_states_per_path=-1, enable_mamba_cache_stochastic_rounding=False, mamba_cache_philox_rounds=0, mamba_full_memory_ratio=0.21, mamba_radix_cache_strategy='extra_buffer_lazy', uses_mamba_radix_cache=False, mamba_track_interval=256, enable_int8_mamba_checkpoint=False, int8_mamba_ckpt_size=None, linear_attn_backend='triton', linear_attn_decode_backend=None, linear_attn_prefill_backend=None, linear_attn_verify_backend=None, enable_linear_replayssm=False, linear_replayssm_cache_len=16, enable_linear_replayssm_spec=False, enable_hierarchical_cache=False, hicache_ratio=2.0, hicache_size=0, hicache_write_policy='write_through', hicache_io_backend='kernel', hicache_mem_layout='page_first', hicache_storage_backend=None, hicache_storage_prefetch_policy='timeout', hicache_storage_backend_extra_config=None, enable_hisparse=False, hisparse_config=None, enable_broadcast_mm_inputs_process=False, enable_prefix_mm_cache=False, mm_enable_dp_encoder=False, mm_process_config={}, mm_processor_worker_num=0, mm_io_worker_num=0, limit_mm_data_per_request=None, enable_mm_global_cache=False, disable_fast_image_processor=False, mm_feature_transport='cpu', keep_mm_feature_on_device=False, enable_lora=None, enable_lora_overlap_loading=None, max_lora_rank=None, lora_target_modules=None, lora_paths=None, max_loaded_loras=None, max_loras_per_batch=8, lora_eviction_policy='lru', lora_backend='csgmv', max_lora_chunk_size=16, experts_shared_outer_loras=None, lora_use_virtual_experts=False, lora_strict_loading=False, lora_drain_wait_threshold=0.0, enable_two_batch_overlap=False, enable_single_batch_overlap=False, tbo_token_distribution_threshold=0.48, cpu_offload_gb=0, offload_group_size=-1, offload_num_in_group=1, offload_prefetch_step=1, offload_mode='cpu', enable_lmcache=False, lmcache_config_file=None, enable_flexkv=False, flexkv_config_file=None, kt_weight_path=None, kt_method='AMXINT4', kt_cpuinfer=None, kt_threadpool_count=2, kt_num_gpu_experts=None, kt_max_deferred_experts_per_token=None, dllm_algorithm=None, dllm_algorithm_config=None, dllm_fdfo=True, disaggregation_mode='decode', disaggregation_transfer_backend='mooncake', disaggregation_bootstrap_port=28800, disaggregation_ib_device='mlx5_0,mlx5_1,mlx5_2,mlx5_3', disaggregation_decode_enable_radix_cache=False, disaggregation_decode_enable_offload_kvcache=False, num_reserved_decode_tokens=512, disaggregation_decode_extra_slots=0, disaggregation_decode_polling_interval=1, optimistic_prefill_attempts=0, encoder_only=False, language_only=False, encoder_transfer_backend='zmq_to_tokenizer', encoder_urls=[], encoder_bootstrap_port=8997, encoder_register_urls=[], enable_adaptive_dispatch_to_encoder=False, enable_pdmux=False, pdmux_config_path=None, sm_group_num=8, custom_weight_loader=[], weight_loader_disable_mmap=False, weight_loader_prefetch_checkpoints=False, weight_loader_prefetch_num_threads=4, weight_loader_drop_cache_after_load=False, remote_instance_weight_loader_seed_instance_ip=None, remote_instance_weight_loader_seed_instance_service_port=None, remote_instance_weight_loader_send_weights_group_ports=None, remote_instance_weight_loader_backend='nccl', remote_instance_weight_loader_start_seed_via_transfer_engine=False, engine_info_bootstrap_port=6789, modelexpress_config=None, download_dir=None, model_checksum=None, delete_ckpt_after_loading=False, decrypted_config_file=None, decrypted_draft_config_file=None, checkpoint_engine_wait_weights_before_ready=False, enable_prefill_delayer=False, prefill_delayer_max_delay_passes=30, prefill_delayer_token_usage_low_watermark=None, prefill_delayer_forward_passes_buckets=None, prefill_delayer_wait_seconds_buckets=None, prefill_delayer_queue_min_ratio=None, prefill_delayer_max_delay_ms=None, min_free_slots_delay=None, enable_deterministic_inference=False, rl_on_policy_target=None, kv_canary='none', kv_canary_real_data='none', kv_canary_sweep_interval=0, enable_dynamic_batch_tokenizer=False, dynamic_batch_tokenizer_batch_size=32, dynamic_batch_tokenizer_batch_timeout=0.002, enable_tokenizer_batch_encode=False, disable_tokenizer_batch_decode=False, debug_tensor_dump_output_folder=None, debug_tensor_dump_layers=None, debug_tensor_dump_input_file=None, enable_memory_saver=False, enable_weights_cpu_backup=False, enable_draft_weights_cpu_backup=False, enable_custom_logit_processor=False, enable_return_hidden_states=False, enable_return_routed_experts=False, enable_return_indexer_topk=False, disable_outlines_disk_cache=False, enable_mis=False, weight_cache_mode='off', weight_cache_socket=None, weight_cache_timeout=1800, forward_hooks=None, msprobe_dump_config=None)
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:22 TP15 EP15] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:22 TP15 EP15] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:23 TP9 EP9] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:23 TP9 EP9] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:23 TP12 EP12] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:23 TP12 EP12] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:23 TP13 EP13] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:23 TP13 EP13] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:23 TP14 EP14] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:23 TP14 EP14] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:23 TP8 EP8] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:23 TP8 EP8] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:23 TP10 EP10] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:23 TP10 EP10] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:23 TP11 EP11] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:23 TP11 EP11] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:24 TP15 EP15] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:24 TP9 EP9] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:24 TP13 EP13] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:24 TP12 EP12] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:24 TP14 EP14] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:25 TP10 EP10] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:25 TP8 EP8] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:25 TP11 EP11] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:26.087226   242 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:26.087333   242 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:26.087350   242 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.6 port: 12001
+I0827 04:36:26.087409   242 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.6:16298
+I0827 04:36:26.087569   242 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:26.095995   236 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:26.096091   236 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:26.096107   236 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.6 port: 12001
+I0827 04:36:26.096150   236 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.6:15958
+I0827 04:36:26.096269   236 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:26.096457   240 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:26.096560   240 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:26.096577   240 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.6 port: 12001
+I0827 04:36:26.096626   240 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.6:16447
+I0827 04:36:26.096773   240 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:26.096876   241 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:26.096975   241 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:26.096992   241 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.6 port: 12001
+I0827 04:36:26.097035   241 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.6:16035
+I0827 04:36:26.097170   241 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:26.103180   241 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:26.107820   240 topology.cpp:127] Device mlx5_0 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:26.109326   239 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:26.109416   239 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:26.109428   239 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.6 port: 12001
+I0827 04:36:26.109464   239 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.6:16843
+I0827 04:36:26.109562   239 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:26.114297   242 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:26.114653   239 topology.cpp:127] Device mlx5_0 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:26.115540   237 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:26.115630   237 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:26.115645   237 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.6 port: 12001
+I0827 04:36:26.115686   237 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.6:15544
+I0827 04:36:26.115813   237 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:26.118220   240 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:26.121481   242 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:26.122418   237 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:26.126497   236 topology.cpp:127] Device mlx5_0 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:26.126574   235 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:26.126684   235 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:26.126703   235 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.6 port: 12001
+I0827 04:36:26.126758   235 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.6:15006
+I0827 04:36:26.126933   235 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:26.127116   240 topology.cpp:127] Device mlx5_2 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:26.127162   238 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:26.127270   238 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:26.127285   238 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.6 port: 12001
+I0827 04:36:26.127342   238 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.6:16140
+I0827 04:36:26.127489   238 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:26.128777   237 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:26.129559   242 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:26.130432   241 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:26.133821   238 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:26.134943   235 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:26.135799   240 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:26.136080   237 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:26.138710   241 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:26.139173   242 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:26.139978   240 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:26.140033   240 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:26.140090   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.140504   235 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:26.142936   242 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:26.142961   242 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:26.143014   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.144655   237 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:26.146834   235 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:26.147805   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:26.147867   241 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:26.148357   239 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:26.148370   237 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:26.148394   237 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:26.148444   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.149503   240 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:06
+I0827 04:36:26.149536   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.150290   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:26.150381   236 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:26.151773   242 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:06
+I0827 04:36:26.151803   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.152055   241 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:26.152087   241 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:26.152148   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.153594   235 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:26.156044   235 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:26.156075   235 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:26.156124   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.157910   238 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:26.159448   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:26.159547   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:26.160912   242 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:06
+I0827 04:36:26.160940   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.161144   241 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:06
+I0827 04:36:26.161176   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.162882   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:26.164348   235 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:06
+I0827 04:36:26.164378   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.165159   238 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:26.168772   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:26.170400   241 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:06
+I0827 04:36:26.170428   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.172459   238 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:26.174376   239 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:26.175243   238 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:26.175272   238 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:26.175325   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.178375   236 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:26.178763   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:26.178785   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:26.180411   237 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:06
+I0827 04:36:26.180462   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.180604   240 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:06
+I0827 04:36:26.180672   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.181643   239 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:26.182818   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:26.184404   238 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:06
+I0827 04:36:26.184437   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.184495   239 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:26.184537   239 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:26.184595   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.185601   236 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:26.186753   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:26.188311   242 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:06
+I0827 04:36:26.188344   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.188390   236 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:26.188416   236 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:26.188470   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.190752   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:26.192051   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:26.192406   235 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:06
+I0827 04:36:26.192444   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.193856   240 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:06
+I0827 04:36:26.193897   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.194197   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:26.195964   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:26.195997   238 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:06
+I0827 04:36:26.196039   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.196040   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:26.197574   242 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:06
+I0827 04:36:26.197599   236 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:06
+I0827 04:36:26.197629   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.197691   242 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:26 TP15 EP15] Init torch distributed begin.
+I0827 04:36:26.198544   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:26.200238   241 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:06
+I0827 04:36:26.200268   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.204000   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:26.205737   238 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:06
+I0827 04:36:26.205766   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.210784   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:26.212373   237 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:06
+I0827 04:36:26.212417   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.218751   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:26.220386   235 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:06
+I0827 04:36:26.220422   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.222854   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:26.223088   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:26.223217   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:26.223345   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:26.224500   241 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:06
+I0827 04:36:26.224536   240 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:06
+I0827 04:36:26.224612   241 transfer_engine_impl.cpp:388] installTransport, type=rdma
+I0827 04:36:26.224682   240 transfer_engine_impl.cpp:388] installTransport, type=rdma
+I0827 04:36:26.224973   236 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:06
+I0827 04:36:26.225018   236 rdma_context.cpp:205] Using SIEVE endpoint store
+[2026-08-27 04:36:26 TP14 EP14] Init torch distributed begin.
+[2026-08-27 04:36:26 TP13 EP13] Init torch distributed begin.
+I0827 04:36:26.225459   239 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:06
+I0827 04:36:26.225523   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.230858   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:26.232468   238 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:06
+I0827 04:36:26.232614   238 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:26 TP11 EP11] Init torch distributed begin.
+I0827 04:36:26.233065   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:26.234685   236 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:06
+I0827 04:36:26.234727   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.238961   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:26.240710   237 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:06
+I0827 04:36:26.240758   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.242511   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:26.244089   236 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:06
+I0827 04:36:26.244236   236 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:26 TP9 EP9] Init torch distributed begin.
+I0827 04:36:26.247059   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:26.248764   235 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:06
+I0827 04:36:26.248920   235 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:26 TP8 EP8] Init torch distributed begin.
+I0827 04:36:26.250787   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:26.252344   239 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:06
+I0827 04:36:26.252393   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.266752   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:26.268360   237 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:06
+I0827 04:36:26.268492   237 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:26 TP10 EP10] Init torch distributed begin.
+I0827 04:36:26.278649   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:26.280236   239 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:06
+I0827 04:36:26.280287   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.306850   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:26.308480   239 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:06
+I0827 04:36:26.308605   239 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:26 TP12 EP12] Init torch distributed begin.
+[2026-08-27 04:36:48 TP14 EP14] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP15 EP15] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP8 EP8] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP11 EP11] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP9 EP9] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP13 EP13] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP10 EP10] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP12 EP12] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP15 EP15] Init torch distributed ends. elapsed=22.03 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP14 EP14] Init torch distributed ends. elapsed=22.01 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP13 EP13] Init torch distributed ends. elapsed=22.01 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP12 EP12] Init torch distributed ends. elapsed=21.92 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP11 EP11] Init torch distributed ends. elapsed=22.00 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP9 EP9] Init torch distributed ends. elapsed=21.99 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP10 EP10] Init torch distributed ends. elapsed=21.96 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP8 EP8] Init torch distributed ends. elapsed=21.98 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP14 EP14] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP15 EP15] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP9 EP9] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP13 EP13] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP8 EP8] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP10 EP10] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP12 EP12] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP14 EP14] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP15 EP15] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP9 EP9] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP10 EP10] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP13 EP13] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP8 EP8] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP12 EP12] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP11 EP11] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP11 EP11] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP13 EP13] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP14 EP14] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP15 EP15] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP12 EP12] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP10 EP10] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP8 EP8] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP9 EP9] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP13 EP13] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP12 EP12] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP14 EP14] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP15 EP15] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP10 EP10] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP8 EP8] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP9 EP9] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP11 EP11] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP11 EP11] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:50 TP8 EP8] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP9 EP9] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP11 EP11] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP12 EP12] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP13 EP13] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP14 EP14] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP10 EP10] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP15 EP15] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:38:57 TP14 EP14] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:57 TP14 EP14] Load weight end. elapsed=129.27 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP9 EP9] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP9 EP9] Load weight end. elapsed=129.52 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP10 EP10] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP10 EP10] Load weight end. elapsed=129.56 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP8 EP8] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP8 EP8] Load weight end. elapsed=129.57 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP13 EP13] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP13 EP13] Load weight end. elapsed=129.58 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP11 EP11] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP11 EP11] Load weight end. elapsed=129.64 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP15 EP15] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP15 EP15] Load weight end. elapsed=129.77 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP12 EP12] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP12 EP12] Load weight end. elapsed=129.90 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:39:01 TP14 EP14] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP14 EP14] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP11 EP11] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP11 EP11] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP8 EP8] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP8 EP8] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP13 EP13] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP13 EP13] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP9 EP9] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP9 EP9] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP15 EP15] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP15 EP15] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP12 EP12] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP12 EP12] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP10 EP10] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP10 EP10] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:02 TP14 EP14] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP15 EP15] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP12 EP12] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP13 EP13] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP10 EP10] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP9 EP9] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP11 EP11] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP8 EP8] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP15 EP15] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP12 EP12] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP10 EP10] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP8 EP8] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP11 EP11] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP13 EP13] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP9 EP9] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP14 EP14] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP9 EP9] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP12 EP12] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP9 EP9] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP10 EP10] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP12 EP12] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP13 EP13] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP15 EP15] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP10 EP10] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP11 EP11] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP13 EP13] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP14 EP14] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP15 EP15] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP11 EP11] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP8 EP8] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP14 EP14] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP8 EP8] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP14 EP14] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP14 EP14] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP9 EP9] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP9 EP9] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP12 EP12] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP12 EP12] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:39:02 TP10 EP10] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP10 EP10] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP8 EP8] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP8 EP8] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:39:02 TP11 EP11] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP11 EP11] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:39:02 TP13 EP13] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP13 EP13] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP15 EP15] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP15 EP15] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:42:56 TP8 EP8] Capture target decode CUDA graph end. elapsed=233.78 s, mem usage=0.60 GB, avail mem=8.52 GB.
+[2026-08-27 04:42:56 TP13 EP13] Capture target decode CUDA graph end. elapsed=233.78 s, mem usage=0.58 GB, avail mem=8.58 GB.
+[2026-08-27 04:42:56 TP8 EP8] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP13 EP13] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP14 EP14] Capture target decode CUDA graph end. elapsed=233.81 s, mem usage=0.56 GB, avail mem=8.59 GB.
+[2026-08-27 04:42:56 TP10 EP10] Capture target decode CUDA graph end. elapsed=233.79 s, mem usage=0.56 GB, avail mem=8.59 GB.
+[2026-08-27 04:42:56 TP14 EP14] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP12 EP12] Capture target decode CUDA graph end. elapsed=233.80 s, mem usage=0.60 GB, avail mem=8.52 GB.
+[2026-08-27 04:42:56 TP10 EP10] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP12 EP12] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP11 EP11] Capture target decode CUDA graph end. elapsed=233.80 s, mem usage=0.62 GB, avail mem=8.51 GB.
+[2026-08-27 04:42:56 TP11 EP11] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP9 EP9] Capture target decode CUDA graph end. elapsed=233.91 s, mem usage=0.58 GB, avail mem=8.58 GB.
+[2026-08-27 04:42:56 TP9 EP9] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP15 EP15] Capture target decode CUDA graph end. elapsed=233.94 s, mem usage=0.62 GB, avail mem=8.51 GB.
+[2026-08-27 04:42:56 TP15 EP15] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+W0827 04:42:57.337543  2432 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5972aec0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.337602  2433 memory_location.cpp:73] Failed to get NUMA node, addr: 0x594a1440, len: 23168: Operation not permitted [1]
+W0827 04:42:57.337656  2434 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59f51680, len: 23168: Operation not permitted [1]
+W0827 04:42:57.337674  2435 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59ebed40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.337788  2436 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5b797200, len: 185344: Operation not permitted [1]
+W0827 04:42:57.337911  2437 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5b7c4640, len: 185344: Operation not permitted [1]
+W0827 04:42:57.337891  2438 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59639800, len: 23168: Operation not permitted [1]
+W0827 04:42:57.337962  2439 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59435700, len: 23168: Operation not permitted [1]
+W0827 04:42:57.338008  2440 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59660c80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.338057  2441 memory_location.cpp:73] Failed to get NUMA node, addr: 0x594c9400, len: 46336: Operation not permitted [1]
+W0827 04:42:57.338071  2442 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59bb1f00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.338107  2443 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59f5fc80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.338161  2445 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5953b700, len: 23168: Operation not permitted [1]
+W0827 04:42:57.338165  2444 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59812080, len: 23168: Operation not permitted [1]
+W0827 04:42:57.338204  2446 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5baddac0, len: 185344: Operation not permitted [1]
+W0827 04:42:57.338325  2448 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59994940, len: 23168: Operation not permitted [1]
+W0827 04:42:57.338340  2447 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5bb0af00, len: 185344: Operation not permitted [1]
+W0827 04:42:57.338538  2449 memory_location.cpp:73] Failed to get NUMA node, addr: 0x45ba2780, len: 46336: Operation not permitted [1]
+W0827 04:42:57.338596  2450 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59ac5f00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.338649  2451 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59984000, len: 23168: Operation not permitted [1]
+W0827 04:42:57.381100  2729 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59997640, len: 23168: Operation not permitted [1]
+W0827 04:42:57.381177  2728 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59a855c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.381280  2730 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59953e40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.381347  2732 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5b87eac0, len: 185344: Operation not permitted [1]
+W0827 04:42:57.381466  2731 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59971000, len: 23168: Operation not permitted [1]
+W0827 04:42:57.381542  2733 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5b8abf00, len: 185344: Operation not permitted [1]
+W0827 04:42:57.381620  2734 memory_location.cpp:73] Failed to get NUMA node, addr: 0x599b8540, len: 23168: Operation not permitted [1]
+W0827 04:42:57.381628  2735 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5927e3c0, len: 46336: Operation not permitted [1]
+W0827 04:42:57.381863  2736 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59900b40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.381878  2737 memory_location.cpp:73] Failed to get NUMA node, addr: 0x598b4f80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.421048  2876 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58bb5680, len: 23168: Operation not permitted [1]
+W0827 04:42:57.421196  2878 memory_location.cpp:73] Failed to get NUMA node, addr: 0x592f7a00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.421193  2877 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59321cc0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.421262  2881 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ab93340, len: 185344: Operation not permitted [1]
+W0827 04:42:57.421321  2879 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5929c2c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.421382  2880 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ab65f00, len: 185344: Operation not permitted [1]
+W0827 04:42:57.421423  2882 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58aa4dc0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.421440  2883 memory_location.cpp:73] Failed to get NUMA node, addr: 0x585179c0, len: 46336: Operation not permitted [1]
+W0827 04:42:57.421434  2884 memory_location.cpp:73] Failed to get NUMA node, addr: 0x588b0240, len: 23168: Operation not permitted [1]
+W0827 04:42:57.421594  2885 memory_location.cpp:73] Failed to get NUMA node, addr: 0x588d0800, len: 23168: Operation not permitted [1]
+W0827 04:42:57.423333  2900 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58bc9040, len: 23168: Operation not permitted [1]
+W0827 04:42:57.423418  2903 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58a57e40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.423974  2913 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ae07a80, len: 185344: Operation not permitted [1]
+W0827 04:42:57.424072  2911 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5adda640, len: 185344: Operation not permitted [1]
+W0827 04:42:57.424093  2908 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58f8c900, len: 23168: Operation not permitted [1]
+W0827 04:42:57.424189  2924 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58d795c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.424269  2906 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58e657c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.424296  2915 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58c46ac0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.424311  2922 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58e39680, len: 23168: Operation not permitted [1]
+W0827 04:42:57.424809  2930 memory_location.cpp:73] Failed to get NUMA node, addr: 0x591a7f40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.424866  2932 memory_location.cpp:73] Failed to get NUMA node, addr: 0x593dee00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.425096  2934 memory_location.cpp:73] Failed to get NUMA node, addr: 0x596af5c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.425186  2935 memory_location.cpp:73] Failed to get NUMA node, addr: 0x595b2ac0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.425271  2938 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5b4f5f80, len: 185344: Operation not permitted [1]
+W0827 04:42:57.425292  2937 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5b4c8b40, len: 185344: Operation not permitted [1]
+W0827 04:42:57.425374  2939 memory_location.cpp:73] Failed to get NUMA node, addr: 0x594e07c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.425433  2942 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58ec6dc0, len: 46336: Operation not permitted [1]
+W0827 04:42:57.425568  2944 memory_location.cpp:73] Failed to get NUMA node, addr: 0x596a8180, len: 23168: Operation not permitted [1]
+W0827 04:42:57.425750  2919 memory_location.cpp:73] Failed to get NUMA node, addr: 0x45850900, len: 46336: Operation not permitted [1]
+W0827 04:42:57.425791  2946 memory_location.cpp:73] Failed to get NUMA node, addr: 0x592c6f00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.427150  2998 memory_location.cpp:73] Failed to get NUMA node, addr: 0x599c15c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.427410  2989 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59750440, len: 23168: Operation not permitted [1]
+W0827 04:42:57.427572  3010 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5bad0cc0, len: 185344: Operation not permitted [1]
+W0827 04:42:57.427641  3014 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a39cd80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.427647  3024 memory_location.cpp:73] Failed to get NUMA node, addr: 0x464ff140, len: 23168: Operation not permitted [1]
+W0827 04:42:57.427655  3002 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a377380, len: 23168: Operation not permitted [1]
+W0827 04:42:57.427663  2993 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59876080, len: 23168: Operation not permitted [1]
+W0827 04:42:57.427681  3007 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5baa3880, len: 185344: Operation not permitted [1]
+W0827 04:42:57.427703  3025 memory_location.cpp:73] Failed to get NUMA node, addr: 0x463cd480, len: 23168: Operation not permitted [1]
+W0827 04:42:57.427706  3018 memory_location.cpp:73] Failed to get NUMA node, addr: 0x464e6b00, len: 46336: Operation not permitted [1]
+W0827 04:42:57.586287  3470 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a4fe500, len: 23168: Operation not permitted [1]
+W0827 04:42:57.586316  3471 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a8129c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.586287  3468 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a530a00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.586329  3469 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a72a280, len: 23168: Operation not permitted [1]
+W0827 04:42:57.586364  3472 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5c756140, len: 185344: Operation not permitted [1]
+W0827 04:42:57.586422  3474 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a287d00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.586429  3473 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5c783580, len: 185344: Operation not permitted [1]
+W0827 04:42:57.586477  3476 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a731080, len: 23168: Operation not permitted [1]
+W0827 04:42:57.586475  3475 memory_location.cpp:73] Failed to get NUMA node, addr: 0x46a69900, len: 46336: Operation not permitted [1]
+W0827 04:42:57.586767  3477 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a2cd580, len: 23168: Operation not permitted [1]
+[2026-08-27 04:42:59] Dummy health check server started in background thread at 0.0.0.0:30000
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/ready_d_node2.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/ready_d_node2.log
new file mode 100644
index 0000000..97f23a1
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/ready_d_node2.log
@@ -0,0 +1,498 @@
+/usr/local/lib/python3.12/dist-packages/torchao/quantization/quant_api.py:1731: SyntaxWarning: invalid escape sequence '\.'
+  """Configuration class for applying different quantization configs to modules or parameters based on their fully qualified names (FQNs).
+/sgl-workspace/sglang/python/sglang/launch_server.py:56: UserWarning: 'python -m sglang.launch_server' is still supported, but 'sglang serve' is the recommended entrypoint.
+  Example: sglang serve --model-path <model> [options]
+  warnings.warn(
+[2026-08-27 04:35:59] KV cache is forced as chunk cache for decode server
+[2026-08-27 04:36:03] Breakable CUDA graph is incompatible with MLA attention; disabling prefill CUDA graph.
+[2026-08-27 04:36:03] Attention backend not specified. Use triton backend by default.
+[2026-08-27 04:36:06] server_args=ServerArgs(model_path='/data/hf_models/Kimi-K3', tokenizer_path='/data/hf_models/Kimi-K3', tokenizer_mode='auto', tokenizer_backend='huggingface', tokenizer_worker_num=1, detokenizer_worker_num=1, skip_tokenizer_init=False, load_format='auto', model_loader_extra_config='{}', trust_remote_code=True, context_length=None, is_embedding=False, enable_multimodal=None, revision=None, model_impl='auto', model_config_parser='auto', json_model_override_args='{}', dtype='auto', quantization=None, quantization_param_path=None, kv_cache_dtype='auto', enable_fp32_lm_head=False, modelopt_quant=None, modelopt_checkpoint_restore_path=None, modelopt_checkpoint_save_path=None, modelopt_export_path=None, quantize_and_serve=False, rl_quant_profile=None, enable_tf32_matmul=False, mem_fraction_static=0.88, max_running_requests=None, max_queued_requests=None, max_total_tokens=None, chunked_prefill_size=8192, enable_dynamic_chunking=False, max_prefill_tokens=16384, prefill_max_requests=None, schedule_policy='fcfs', enable_priority_scheduling=False, disable_priority_preemption=False, default_priority_value=None, abort_on_priority_when_disabled=False, schedule_low_priority_values_first=False, priority_scheduling_preemption_threshold=10, retraction_policy='length', schedule_conservativeness=1.0, page_size=64, swa_full_tokens_ratio=0.8, disable_hybrid_swa_memory=False, radix_eviction_policy='lru', prefill_only_disable_kv_cache=False, disable_radix_cache=True, enable_page_major_kv_layout=False, enable_unified_memory=False, disable_chunked_prefix_cache=False, disable_overlap_schedule=False, num_continuous_decode_steps=1, scheduler_recv_interval=1, enable_mixed_chunk=False, nccl_port=None, dist_timeout=3600, dist_init_addr='174.1.60.5:20000', nnodes=4, node_rank=2, tp_size=32, dcp_size=1, pp_size=1, pp_max_micro_batch_size=None, pp_async_batch_depth=0, dp_size=1, load_balance_method='round_robin', attn_cp_size=1, moe_dp_size=1, dcp_comm_backend='ag_rs', dcp_replicate_q_proj=None, dwdp_size=1, enable_prefill_cp=False, cp_strategy=None, enable_dsa_cache_layer_split=False, enable_dsa_prefill_context_parallel=False, dsa_prefill_cp_mode='round-robin-split', enable_prefill_context_parallel=False, prefill_cp_mode='in-seq-split', enable_cp_decode_attn_tp=False, enable_dp_attention=False, enable_dp_attention_local_control_broadcast=False, enable_dp_lm_head=False, enable_attn_tp_input_scattered=False, disable_attn_tp_gather=False, enable_p2p_check=False, device='cuda', base_gpu_id=0, gpu_id_step=1, random_seed=212084647, watchdog_timeout=300, soft_watchdog_timeout=None, sleep_on_idle=False, use_ray=False, custom_sigquit_handler=None, numa_node=None, gc_threshold=None, host='0.0.0.0', port=30000, fastapi_root_path='', smg_grpc_mode=False, grpc_mode=False, grpc_port=None, sidecar=None, sidecar_args=None, skip_server_warmup=False, warmups=None, enable_http2=False, ssl_keyfile=None, ssl_certfile=None, ssl_ca_certs=None, ssl_keyfile_password=None, enable_ssl_refresh=False, api_key=None, admin_api_key=None, served_model_name='kimi-k3', weight_version='default', chat_template=None, hf_chat_template_name=None, completion_template=None, file_storage_path='sglang_storage', enable_cache_report=False, reasoning_parser=None, default_chat_template_kwargs=None, strip_thinking_cache=False, enable_strict_thinking=False, tool_call_parser=None, tool_server=None, sampling_defaults='model', asr_max_buffer_seconds=60, asr_max_concurrent_sessions=32, preferred_sampling_params=None, allow_auto_truncate=False, stream_interval=1, batch_notify_size=16, stream_response_default_include_usage=False, incremental_streaming_output=False, enable_streaming_session=False, enable_session_radix_cache=False, log_level='info', log_level_http=None, log_requests=False, log_requests_level=2, log_requests_format='text', log_requests_target=None, uvicorn_access_log_exclude_prefixes=[], crash_dump_folder=None, show_time_cost=False, enable_metrics=False, smg_http_sidecar_port=None, enable_mfu_metrics=False, enable_metrics_for_all_schedulers=False, load_snapshot_publish_interval=15, tokenizer_metrics_custom_labels_header='x-custom-labels', tokenizer_metrics_allowed_custom_labels=None, extra_metric_labels=None, bucket_time_to_first_token=None, bucket_inter_token_latency=None, bucket_e2e_request_latency=None, prompt_tokens_buckets=None, generation_tokens_buckets=None, gc_warning_threshold_secs=0.0, decode_log_interval=40, enable_request_time_stats_logging=False, kv_events_config=None, enable_forward_pass_metrics=False, forward_pass_metrics_worker_id='', forward_pass_metrics_ipc_name=None, enable_trace=False, trace_modules='request', otlp_traces_endpoint='localhost:4317', export_metrics_to_file=False, export_metrics_to_file_dir=None, stat_loggers=None, constrained_json_whitespace_pattern=None, constrained_json_disable_any_whitespace=False, attention_backend='triton', decode_attention_backend=None, prefill_attention_backend=None, sampling_backend='flashinfer', grammar_backend='xgrammar', radix_cache_backend=None, mm_attention_backend=None, fp8_gemm_runner_backend='auto', fp4_gemm_runner_backend='auto', bf16_gemm_backend='auto', dsa_prefill_backend=None, dsa_decode_backend=None, dsa_paged_mqa_logits_backend='auto', dsa_topk_backend='sgl-kernel', disable_flashinfer_autotune=False, flashinfer_autotune_skip_ops=None, mamba_backend='triton', cuda_graph_config=CudaGraphConfig(decode=PhaseConfig(backend='full', max_bs=16, bs=[1, 2, 4, 8, 12, 16], tc_compiler='eager', full_prefill_max_req=None), prefill=PhaseConfig(backend='disabled', max_bs=2048, bs=[4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], tc_compiler='eager', full_prefill_max_req=None)), cuda_graph_backend_decode=None, cuda_graph_backend_prefill=None, cuda_graph_max_bs_decode=16, cuda_graph_max_bs_prefill=None, cuda_graph_bs_decode=None, cuda_graph_bs_prefill=None, cuda_graph_tc_compiler=None, disable_prefill_cuda_graph=False, disable_decode_cuda_graph=False, disable_cuda_graph=False, disable_cuda_graph_padding=False, enable_profile_cuda_graph=False, enable_cudagraph_gc=False, debug_cuda_graph=False, enable_layerwise_nvtx_marker=False, enable_nccl_nvls=False, enable_symm_mem=False, triton_attention_reduce_in_fp32=False, triton_attention_num_kv_splits=8, triton_attention_split_tile_size=None, flashinfer_mla_disable_ragged=False, enable_fused_qk_norm_rope=False, enable_precise_embedding_interpolation=False, enable_fused_moe_sum_all_reduce=False, enable_deepseek_v4_fp4_indexer=False, disable_custom_all_reduce=False, enable_mscclpp=False, enable_torch_symm_mem=False, enable_scattered_sconv=False, pre_warm_nccl=False, enable_quant_communications=False, enable_flashinfer_allreduce_fusion=False, enforce_disable_flashinfer_allreduce_fusion=False, flashinfer_allreduce_fusion_backend=None, enable_aiter_allreduce_fusion=False, enable_torch_compile=False, enable_torch_compile_debug_mode=False, torch_compile_max_bs=32, torchao_config='', speculative_algorithm=None, speculative_draft_model_path=None, speculative_draft_model_revision=None, speculative_draft_load_format=None, speculative_num_steps=None, speculative_eagle_topk=None, speculative_num_draft_tokens=None, speculative_dflash_block_size=None, speculative_dspark_block_size=None, speculative_dspark_sps_table_path=None, speculative_dspark_confidence_sts_path=None, speculative_dspark_align_verify_tokens_to_graph_tier=False, speculative_accept_threshold_single=1.0, speculative_accept_threshold_acc=1.0, speculative_use_rejection_sampling=False, speculative_token_map=None, speculative_attention_mode='prefill', speculative_draft_attention_backend=None, speculative_draft_window_size=None, speculative_moe_runner_backend='marlin', speculative_moe_a2a_backend=None, speculative_draft_model_quantization=None, speculative_skip_dp_mlp_sync=False, enable_multi_layer_eagle=False, speculative_adaptive=False, speculative_adaptive_config=None, decoupled_spec_bind_endpoint=None, decoupled_spec_connect_endpoints=None, decoupled_spec_rank=None, decoupled_spec_role='null', spec_trace_dir=None, speculative_ngram_min_bfs_breadth=1, speculative_ngram_max_bfs_breadth=10, speculative_ngram_match_type='BFS', speculative_ngram_max_trie_depth=18, speculative_ngram_capacity=10000000, speculative_ngram_external_corpus_path=None, speculative_ngram_external_sam_budget=0, speculative_ngram_external_corpus_max_tokens=10000000, ep_size=32, moe_a2a_backend='none', moe_runner_backend='marlin', flashinfer_mxfp4_moe_precision='default', deepep_mode='auto', fuseep_mode=2, deepep_dispatcher_output_dtype='auto', ep_num_redundant_experts=0, ep_dispatch_algorithm=None, init_expert_location='trivial', enable_eplb=False, eplb_algorithm='auto', eplb_rebalance_num_iterations=1000, eplb_rebalance_layers_per_chunk=None, eplb_min_rebalancing_utilization_threshold=1.0, expert_distribution_recorder_mode=None, expert_distribution_recorder_buffer_size=1000, enable_expert_distribution_metrics=False, deepep_config=None, moe_dense_tp_size=None, elastic_ep_backend=None, enable_elastic_expert_backup=False, mooncake_ib_device=None, enable_waterfill=False, ep_join_mode=None, ep_join_rank_offset=0, elastic_ep_initial_size=None, max_ep_size=None, elastic_ep_scale_timeout=600, elastic_ep_rejoin=False, disable_flashinfer_cutlass_moe_fp4_allgather=False, disable_shared_experts_fusion=False, enforce_shared_experts_fusion=False, max_mamba_cache_size=None, mamba_ssm_dtype=None, mamba_max_states_per_path=-1, enable_mamba_cache_stochastic_rounding=False, mamba_cache_philox_rounds=0, mamba_full_memory_ratio=0.21, mamba_radix_cache_strategy='extra_buffer_lazy', uses_mamba_radix_cache=False, mamba_track_interval=256, enable_int8_mamba_checkpoint=False, int8_mamba_ckpt_size=None, linear_attn_backend='triton', linear_attn_decode_backend=None, linear_attn_prefill_backend=None, linear_attn_verify_backend=None, enable_linear_replayssm=False, linear_replayssm_cache_len=16, enable_linear_replayssm_spec=False, enable_hierarchical_cache=False, hicache_ratio=2.0, hicache_size=0, hicache_write_policy='write_through', hicache_io_backend='kernel', hicache_mem_layout='page_first', hicache_storage_backend=None, hicache_storage_prefetch_policy='timeout', hicache_storage_backend_extra_config=None, enable_hisparse=False, hisparse_config=None, enable_broadcast_mm_inputs_process=False, enable_prefix_mm_cache=False, mm_enable_dp_encoder=False, mm_process_config={}, mm_processor_worker_num=0, mm_io_worker_num=0, limit_mm_data_per_request=None, enable_mm_global_cache=False, disable_fast_image_processor=False, mm_feature_transport='cpu', keep_mm_feature_on_device=False, enable_lora=None, enable_lora_overlap_loading=None, max_lora_rank=None, lora_target_modules=None, lora_paths=None, max_loaded_loras=None, max_loras_per_batch=8, lora_eviction_policy='lru', lora_backend='csgmv', max_lora_chunk_size=16, experts_shared_outer_loras=None, lora_use_virtual_experts=False, lora_strict_loading=False, lora_drain_wait_threshold=0.0, enable_two_batch_overlap=False, enable_single_batch_overlap=False, tbo_token_distribution_threshold=0.48, cpu_offload_gb=0, offload_group_size=-1, offload_num_in_group=1, offload_prefetch_step=1, offload_mode='cpu', enable_lmcache=False, lmcache_config_file=None, enable_flexkv=False, flexkv_config_file=None, kt_weight_path=None, kt_method='AMXINT4', kt_cpuinfer=None, kt_threadpool_count=2, kt_num_gpu_experts=None, kt_max_deferred_experts_per_token=None, dllm_algorithm=None, dllm_algorithm_config=None, dllm_fdfo=True, disaggregation_mode='decode', disaggregation_transfer_backend='mooncake', disaggregation_bootstrap_port=28800, disaggregation_ib_device='mlx5_0,mlx5_1,mlx5_2,mlx5_3', disaggregation_decode_enable_radix_cache=False, disaggregation_decode_enable_offload_kvcache=False, num_reserved_decode_tokens=512, disaggregation_decode_extra_slots=0, disaggregation_decode_polling_interval=1, optimistic_prefill_attempts=0, encoder_only=False, language_only=False, encoder_transfer_backend='zmq_to_tokenizer', encoder_urls=[], encoder_bootstrap_port=8997, encoder_register_urls=[], enable_adaptive_dispatch_to_encoder=False, enable_pdmux=False, pdmux_config_path=None, sm_group_num=8, custom_weight_loader=[], weight_loader_disable_mmap=False, weight_loader_prefetch_checkpoints=False, weight_loader_prefetch_num_threads=4, weight_loader_drop_cache_after_load=False, remote_instance_weight_loader_seed_instance_ip=None, remote_instance_weight_loader_seed_instance_service_port=None, remote_instance_weight_loader_send_weights_group_ports=None, remote_instance_weight_loader_backend='nccl', remote_instance_weight_loader_start_seed_via_transfer_engine=False, engine_info_bootstrap_port=6789, modelexpress_config=None, download_dir=None, model_checksum=None, delete_ckpt_after_loading=False, decrypted_config_file=None, decrypted_draft_config_file=None, checkpoint_engine_wait_weights_before_ready=False, enable_prefill_delayer=False, prefill_delayer_max_delay_passes=30, prefill_delayer_token_usage_low_watermark=None, prefill_delayer_forward_passes_buckets=None, prefill_delayer_wait_seconds_buckets=None, prefill_delayer_queue_min_ratio=None, prefill_delayer_max_delay_ms=None, min_free_slots_delay=None, enable_deterministic_inference=False, rl_on_policy_target=None, kv_canary='none', kv_canary_real_data='none', kv_canary_sweep_interval=0, enable_dynamic_batch_tokenizer=False, dynamic_batch_tokenizer_batch_size=32, dynamic_batch_tokenizer_batch_timeout=0.002, enable_tokenizer_batch_encode=False, disable_tokenizer_batch_decode=False, debug_tensor_dump_output_folder=None, debug_tensor_dump_layers=None, debug_tensor_dump_input_file=None, enable_memory_saver=False, enable_weights_cpu_backup=False, enable_draft_weights_cpu_backup=False, enable_custom_logit_processor=False, enable_return_hidden_states=False, enable_return_routed_experts=False, enable_return_indexer_topk=False, disable_outlines_disk_cache=False, enable_mis=False, weight_cache_mode='off', weight_cache_socket=None, weight_cache_timeout=1800, forward_hooks=None, msprobe_dump_config=None)
+[2026-08-27 04:36:06] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:06] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:06] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:06] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:06] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:06] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:06] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:06] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:24 TP18 EP18] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:24 TP18 EP18] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:24 TP21 EP21] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:24 TP21 EP21] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:24 TP20 EP20] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:24 TP20 EP20] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:25 TP19 EP19] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:25 TP19 EP19] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:25 TP22 EP22] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:25 TP22 EP22] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:25 TP23 EP23] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:25 TP23 EP23] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:26 TP17 EP17] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:26 TP17 EP17] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:26 TP16 EP16] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:26 TP16 EP16] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:27 TP19 EP19] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:27 TP21 EP21] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:27 TP18 EP18] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:27 TP20 EP20] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:27 TP23 EP23] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:27 TP22 EP22] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:27 TP17 EP17] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:27 TP16 EP16] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:29.187384   239 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:29.187532   239 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:29.187558   239 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.7 port: 12001
+I0827 04:36:29.187624   239 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.7:16248
+I0827 04:36:29.187814   239 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:29.191661   235 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:29.191802   235 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:29.191823   235 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.7 port: 12001
+I0827 04:36:29.191901   235 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.7:16408
+I0827 04:36:29.192099   235 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:29.195215   242 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:29.195319   242 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:29.195333   242 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.7 port: 12001
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:29.195333   238 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:29.195374   242 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.7:15148
+I0827 04:36:29.195490   238 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:29.195513   238 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.7 port: 12001
+I0827 04:36:29.195519   242 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:29.195581   238 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.7:16909
+I0827 04:36:29.195815   238 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:29.198207   236 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:29.198346   236 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:29.198370   236 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.7 port: 12001
+I0827 04:36:29.198427   236 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.7:16738
+I0827 04:36:29.198611   236 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:29.200070   242 topology.cpp:127] Device mlx5_0 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:29.201086   237 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:29.201221   237 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:29.201239   237 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.7 port: 12001
+I0827 04:36:29.201300   237 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.7:16522
+I0827 04:36:29.201495   237 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:29.201511   240 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:29.201632   240 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:29.201648   240 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.7 port: 12001
+I0827 04:36:29.201701   240 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.7:16431
+I0827 04:36:29.201871   240 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:29.204236   241 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:29.204355   242 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:29.204380   241 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:29.204393   241 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.7 port: 12001
+I0827 04:36:29.204454   241 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.7:16508
+I0827 04:36:29.204633   241 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:29.204896   238 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:29.207675   237 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:29.208006   240 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:29.208549   242 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:29.212560   242 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:29.214218   242 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:29.214243   242 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:29.214290   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.214713   240 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:29.214757   237 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:29.219102   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:29.219870   239 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:29.220057   235 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:29.220343   242 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:07
+I0827 04:36:29.220369   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.222033   240 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:29.222069   237 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:29.223913   236 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:29.228577   235 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:29.228634   239 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:29.230391   237 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:29.230446   240 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:29.232455   236 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:29.233546   237 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:29.233578   237 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:29.233631   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.234635   240 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:29.234665   240 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:29.234716   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.235774   238 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:29.235827   241 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:29.237140   235 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:29.241117   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:29.241130   236 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:29.242069   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:29.242725   237 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:07
+I0827 04:36:29.242758   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.243661   240 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:07
+I0827 04:36:29.243691   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.245620   235 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:29.247138   238 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:29.247233   241 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:29.248172   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:29.248514   235 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:29.248550   235 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:29.248612   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.248719   236 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:29.249847   242 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:07
+I0827 04:36:29.249886   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.251354   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:29.251672   236 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:29.251719   236 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:29.251781   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.252889   240 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:07
+I0827 04:36:29.252921   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.255584   239 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:29.256263   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:29.257805   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:29.257980   235 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:07
+I0827 04:36:29.258016   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.258512   238 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:29.258567   241 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:29.259447   242 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:07
+I0827 04:36:29.259477   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.259642   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:29.261375   236 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:07
+I0827 04:36:29.261410   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.261518   238 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:29.261557   238 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:29.261615   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.261615   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:29.262753   239 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:29.263202   240 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:07
+I0827 04:36:29.263229   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.265475   239 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:29.265504   239 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:29.265558   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.269449   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:29.271157   236 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:07
+I0827 04:36:29.271186   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.271761   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:29.272920   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:29.273422   237 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:07
+I0827 04:36:29.273456   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.274566   239 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:07
+I0827 04:36:29.274600   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.281176   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:29.282341   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:29.282708   237 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:07
+I0827 04:36:29.282739   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.284137   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:29.284178   239 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:07
+I0827 04:36:29.284219   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.285904   235 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:07
+I0827 04:36:29.285953   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.288185   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:29.289810   242 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:07
+I0827 04:36:29.289940   242 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:29 TP23 EP23] Init torch distributed begin.
+I0827 04:36:29.291602   241 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:29.291882   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:29.291936   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:29.293576   240 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:07
+I0827 04:36:29.293699   240 transfer_engine_impl.cpp:388] installTransport, type=rdma
+I0827 04:36:29.293921   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+[2026-08-27 04:36:29 TP21 EP21] Init torch distributed begin.
+I0827 04:36:29.294121   238 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:07
+I0827 04:36:29.294169   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.294241   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:29.295475   239 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:07
+I0827 04:36:29.295507   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.295553   241 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:29.295591   241 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:29.295648   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.296002   235 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:07
+I0827 04:36:29.296041   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.296232   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:29.298113   236 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:07
+I0827 04:36:29.298154   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.303915   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:29.305596   235 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:07
+I0827 04:36:29.305743   235 transfer_engine_impl.cpp:388] installTransport, type=rdma
+I0827 04:36:29.306217   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+[2026-08-27 04:36:29 TP16 EP16] Init torch distributed begin.
+I0827 04:36:29.307994   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:29.308105   236 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:07
+I0827 04:36:29.308277   236 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:29 TP17 EP17] Init torch distributed begin.
+I0827 04:36:29.310081   237 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:07
+I0827 04:36:29.310261   237 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:29 TP18 EP18] Init torch distributed begin.
+I0827 04:36:29.320421   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:29.321812   239 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:07
+I0827 04:36:29.321919   239 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:29 TP20 EP20] Init torch distributed begin.
+I0827 04:36:29.323829   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:29.323894   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:29.325821   238 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:07
+I0827 04:36:29.325820   241 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:07
+I0827 04:36:29.325871   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.325872   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.360142   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:29.360253   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:29.362092   241 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:07
+I0827 04:36:29.362147   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.362201   238 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:07
+I0827 04:36:29.362243   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.392091   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:29.392130   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:29.394111   238 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:07
+I0827 04:36:29.394127   241 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:07
+I0827 04:36:29.394166   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.394286   238 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:29 TP19 EP19] Init torch distributed begin.
+I0827 04:36:29.420166   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:29.421864   241 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:07
+I0827 04:36:29.421993   241 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:29 TP22 EP22] Init torch distributed begin.
+[2026-08-27 04:36:48 TP23 EP23] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP22 EP22] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP21 EP21] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP20 EP20] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP19 EP19] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP18 EP18] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP17 EP17] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP16 EP16] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP22 EP22] Init torch distributed ends. elapsed=18.81 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP23 EP23] Init torch distributed ends. elapsed=18.94 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP21 EP21] Init torch distributed ends. elapsed=18.93 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP20 EP20] Init torch distributed ends. elapsed=18.91 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP19 EP19] Init torch distributed ends. elapsed=18.83 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP18 EP18] Init torch distributed ends. elapsed=18.92 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP17 EP17] Init torch distributed ends. elapsed=18.92 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP16 EP16] Init torch distributed ends. elapsed=18.92 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP21 EP21] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP16 EP16] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP22 EP22] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP20 EP20] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP23 EP23] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP18 EP18] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP19 EP19] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP22 EP22] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP20 EP20] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP21 EP21] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP16 EP16] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP23 EP23] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP19 EP19] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP18 EP18] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP17 EP17] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP17 EP17] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP20 EP20] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP22 EP22] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP18 EP18] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP23 EP23] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP19 EP19] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP21 EP21] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP20 EP20] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP16 EP16] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP18 EP18] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP22 EP22] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP19 EP19] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP21 EP21] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP23 EP23] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP16 EP16] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP17 EP17] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP17 EP17] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:50 TP23 EP23] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP22 EP22] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP21 EP21] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP19 EP19] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP18 EP18] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP20 EP20] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP16 EP16] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP17 EP17] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:38:53 TP16 EP16] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:53 TP16 EP16] Load weight end. elapsed=124.66 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:53 TP17 EP17] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:53 TP17 EP17] Load weight end. elapsed=124.69 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:53 TP23 EP23] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:53 TP23 EP23] Load weight end. elapsed=124.71 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:53 TP21 EP21] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:53 TP21 EP21] Load weight end. elapsed=124.74 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:53 TP18 EP18] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:53 TP18 EP18] Load weight end. elapsed=124.82 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:53 TP20 EP20] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:53 TP20 EP20] Load weight end. elapsed=124.82 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:53 TP22 EP22] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:53 TP22 EP22] Load weight end. elapsed=124.92 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:53 TP19 EP19] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:53 TP19 EP19] Load weight end. elapsed=125.03 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:39:01 TP17 EP17] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP17 EP17] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP23 EP23] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP23 EP23] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP16 EP16] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP16 EP16] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP18 EP18] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP18 EP18] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP22 EP22] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP22 EP22] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP21 EP21] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP21 EP21] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP20 EP20] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP20 EP20] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP19 EP19] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP19 EP19] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:02 TP23 EP23] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP18 EP18] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP21 EP21] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP22 EP22] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP20 EP20] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP17 EP17] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP19 EP19] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP16 EP16] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP21 EP21] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP22 EP22] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP23 EP23] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP17 EP17] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP16 EP16] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP20 EP20] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP18 EP18] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP19 EP19] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP21 EP21] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP23 EP23] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP22 EP22] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP17 EP17] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP23 EP23] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP21 EP21] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP20 EP20] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP16 EP16] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP22 EP22] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP17 EP17] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP20 EP20] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP16 EP16] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP18 EP18] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP19 EP19] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP18 EP18] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP19 EP19] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP23 EP23] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP23 EP23] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:39:02 TP17 EP17] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP17 EP17] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP16 EP16] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP16 EP16] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:39:02 TP21 EP21] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP21 EP21] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP22 EP22] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP22 EP22] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP20 EP20] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP20 EP20] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:39:02 TP18 EP18] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP18 EP18] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP19 EP19] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP19 EP19] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:42:56 TP19 EP19] Capture target decode CUDA graph end. elapsed=233.80 s, mem usage=0.62 GB, avail mem=8.51 GB.
+[2026-08-27 04:42:56 TP19 EP19] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP16 EP16] Capture target decode CUDA graph end. elapsed=233.84 s, mem usage=0.60 GB, avail mem=8.52 GB.
+[2026-08-27 04:42:56 TP16 EP16] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP22 EP22] Capture target decode CUDA graph end. elapsed=233.84 s, mem usage=0.56 GB, avail mem=8.59 GB.
+[2026-08-27 04:42:56 TP22 EP22] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP17 EP17] Capture target decode CUDA graph end. elapsed=233.88 s, mem usage=0.58 GB, avail mem=8.58 GB.
+[2026-08-27 04:42:56 TP17 EP17] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP21 EP21] Capture target decode CUDA graph end. elapsed=233.90 s, mem usage=0.58 GB, avail mem=8.58 GB.
+[2026-08-27 04:42:56 TP21 EP21] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP20 EP20] Capture target decode CUDA graph end. elapsed=233.91 s, mem usage=0.60 GB, avail mem=8.52 GB.
+[2026-08-27 04:42:56 TP20 EP20] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP23 EP23] Capture target decode CUDA graph end. elapsed=234.01 s, mem usage=0.62 GB, avail mem=8.51 GB.
+[2026-08-27 04:42:56 TP23 EP23] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP18 EP18] Capture target decode CUDA graph end. elapsed=234.00 s, mem usage=0.56 GB, avail mem=8.59 GB.
+[2026-08-27 04:42:56 TP18 EP18] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+W0827 04:42:57.619849  2423 memory_location.cpp:73] Failed to get NUMA node, addr: 0x9680d380, len: 23168: Operation not permitted [1]
+W0827 04:42:57.619855  2422 memory_location.cpp:73] Failed to get NUMA node, addr: 0x9696f4c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.619967  2424 memory_location.cpp:73] Failed to get NUMA node, addr: 0x966bf4c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.620213  2425 memory_location.cpp:73] Failed to get NUMA node, addr: 0x82b481c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.620281  2426 memory_location.cpp:73] Failed to get NUMA node, addr: 0x988cf580, len: 185344: Operation not permitted [1]
+W0827 04:42:57.620285  2427 memory_location.cpp:73] Failed to get NUMA node, addr: 0x988fc9c0, len: 185344: Operation not permitted [1]
+W0827 04:42:57.620325  2428 memory_location.cpp:73] Failed to get NUMA node, addr: 0x96988ec0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.620384  2429 memory_location.cpp:73] Failed to get NUMA node, addr: 0x96949540, len: 46336: Operation not permitted [1]
+W0827 04:42:57.620546  2430 memory_location.cpp:73] Failed to get NUMA node, addr: 0x96790780, len: 23168: Operation not permitted [1]
+W0827 04:42:57.620606  2431 memory_location.cpp:73] Failed to get NUMA node, addr: 0x967ccac0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.651996  2570 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6c8ac200, len: 23168: Operation not permitted [1]
+W0827 04:42:57.652108  2571 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6c6f5f80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.652212  2572 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6c8ea380, len: 23168: Operation not permitted [1]
+W0827 04:42:57.652417  2573 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6d2e8ec0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.652568  2574 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6ea4eb80, len: 185344: Operation not permitted [1]
+W0827 04:42:57.652587  2576 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6cb20d80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.652642  2577 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6c987340, len: 46336: Operation not permitted [1]
+W0827 04:42:57.652667  2575 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6ea7bfc0, len: 185344: Operation not permitted [1]
+W0827 04:42:57.652815  2578 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6c6eb200, len: 23168: Operation not permitted [1]
+W0827 04:42:57.652889  2579 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6cb0a2c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.713306  2681 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8a7feb00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.713331  2680 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ad55240, len: 23168: Operation not permitted [1]
+W0827 04:42:57.713303  2679 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ad28280, len: 23168: Operation not permitted [1]
+W0827 04:42:57.713424  2682 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8abc1200, len: 23168: Operation not permitted [1]
+W0827 04:42:57.713523  2683 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ca64980, len: 185344: Operation not permitted [1]
+W0827 04:42:57.713629  2685 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ca91dc0, len: 185344: Operation not permitted [1]
+W0827 04:42:57.713903  2687 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ac54c00, len: 46336: Operation not permitted [1]
+W0827 04:42:57.714316  2688 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ab93e40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.714335  2689 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8abe8700, len: 23168: Operation not permitted [1]
+W0827 04:42:57.714352  2686 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8a7f7880, len: 23168: Operation not permitted [1]
+W0827 04:42:57.717235  2716 memory_location.cpp:73] Failed to get NUMA node, addr: 0x739fc3c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.717247  2718 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ff359c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.717484  2725 memory_location.cpp:73] Failed to get NUMA node, addr: 0x75e8fa00, len: 185344: Operation not permitted [1]
+W0827 04:42:57.717736  2729 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ff46700, len: 46336: Operation not permitted [1]
+W0827 04:42:57.718014  2728 memory_location.cpp:73] Failed to get NUMA node, addr: 0x73b6a280, len: 23168: Operation not permitted [1]
+W0827 04:42:57.718112  2730 memory_location.cpp:73] Failed to get NUMA node, addr: 0x73dda340, len: 23168: Operation not permitted [1]
+W0827 04:42:57.718163  2731 memory_location.cpp:73] Failed to get NUMA node, addr: 0x73dbf000, len: 23168: Operation not permitted [1]
+W0827 04:42:57.718196  2721 memory_location.cpp:73] Failed to get NUMA node, addr: 0x73a8f640, len: 23168: Operation not permitted [1]
+W0827 04:42:57.718261  2723 memory_location.cpp:73] Failed to get NUMA node, addr: 0x73c5ba40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.724411  2727 memory_location.cpp:73] Failed to get NUMA node, addr: 0x75ebce40, len: 185344: Operation not permitted [1]
+W0827 04:42:57.755818  2858 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5e996800, len: 23168: Operation not permitted [1]
+W0827 04:42:57.755931  2859 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ec52140, len: 23168: Operation not permitted [1]
+W0827 04:42:57.756026  2861 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5e767ec0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.756074  2862 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60bbb440, len: 185344: Operation not permitted [1]
+W0827 04:42:57.756379  2865 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4aecca80, len: 46336: Operation not permitted [1]
+W0827 04:42:57.756517  2863 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60be8880, len: 185344: Operation not permitted [1]
+W0827 04:42:57.756536  2864 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5eb0d880, len: 23168: Operation not permitted [1]
+W0827 04:42:57.756650  2866 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ea7fe80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.757373  2860 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5eb4d800, len: 23168: Operation not permitted [1]
+W0827 04:42:57.757380  2867 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ea97dc0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.818434  3070 memory_location.cpp:73] Failed to get NUMA node, addr: 0x965568c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.818459  3075 memory_location.cpp:73] Failed to get NUMA node, addr: 0x96700fc0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.818470  3072 memory_location.cpp:73] Failed to get NUMA node, addr: 0x96541bc0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.820174  3079 memory_location.cpp:73] Failed to get NUMA node, addr: 0x984fb440, len: 185344: Operation not permitted [1]
+W0827 04:42:57.822142  3084 memory_location.cpp:73] Failed to get NUMA node, addr: 0x962df200, len: 23168: Operation not permitted [1]
+W0827 04:42:57.822369  3086 memory_location.cpp:73] Failed to get NUMA node, addr: 0x831c0780, len: 46336: Operation not permitted [1]
+W0827 04:42:57.822384  3110 memory_location.cpp:73] Failed to get NUMA node, addr: 0x965cd3c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.822431  3082 memory_location.cpp:73] Failed to get NUMA node, addr: 0x98528880, len: 185344: Operation not permitted [1]
+W0827 04:42:57.822527  3127 memory_location.cpp:73] Failed to get NUMA node, addr: 0x9642bbc0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.829262  3077 memory_location.cpp:73] Failed to get NUMA node, addr: 0x9640b580, len: 23168: Operation not permitted [1]
+W0827 04:42:57.832078  3193 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ce567c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.832136  3194 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8cbe8d00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.832149  3195 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8cc94640, len: 23168: Operation not permitted [1]
+W0827 04:42:57.832294  3196 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8d79f140, len: 23168: Operation not permitted [1]
+W0827 04:42:57.832419  3199 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ce4bec0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.832562  3201 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8d7e4b40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.832683  3202 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8cbbd940, len: 23168: Operation not permitted [1]
+W0827 04:42:57.833209  3198 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ef88080, len: 185344: Operation not permitted [1]
+W0827 04:42:57.833267  3200 memory_location.cpp:73] Failed to get NUMA node, addr: 0x79c278c0, len: 46336: Operation not permitted [1]
+W0827 04:42:57.837312  3197 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ef5ac40, len: 185344: Operation not permitted [1]
+W0827 04:42:58.193480  3458 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59a23f40, len: 23168: Operation not permitted [1]
+W0827 04:42:58.193507  3459 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59aa70c0, len: 23168: Operation not permitted [1]
+W0827 04:42:58.193545  3460 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5999c100, len: 23168: Operation not permitted [1]
+W0827 04:42:58.193585  3461 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59bc3040, len: 23168: Operation not permitted [1]
+W0827 04:42:58.193631  3462 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5bd1e5c0, len: 185344: Operation not permitted [1]
+W0827 04:42:58.193667  3463 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5bd4ba00, len: 185344: Operation not permitted [1]
+W0827 04:42:58.193720  3464 memory_location.cpp:73] Failed to get NUMA node, addr: 0x599fa280, len: 23168: Operation not permitted [1]
+W0827 04:42:58.193746  3466 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59dd6100, len: 23168: Operation not permitted [1]
+W0827 04:42:58.193764  3467 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59cbd400, len: 23168: Operation not permitted [1]
+W0827 04:42:58.193776  3465 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59db0340, len: 46336: Operation not permitted [1]
+[2026-08-27 04:43:00] Dummy health check server started in background thread at 0.0.0.0:30000
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/ready_d_node3.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/ready_d_node3.log
new file mode 100644
index 0000000..9791774
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/ready_d_node3.log
@@ -0,0 +1,498 @@
+/usr/local/lib/python3.12/dist-packages/torchao/quantization/quant_api.py:1731: SyntaxWarning: invalid escape sequence '\.'
+  """Configuration class for applying different quantization configs to modules or parameters based on their fully qualified names (FQNs).
+/sgl-workspace/sglang/python/sglang/launch_server.py:56: UserWarning: 'python -m sglang.launch_server' is still supported, but 'sglang serve' is the recommended entrypoint.
+  Example: sglang serve --model-path <model> [options]
+  warnings.warn(
+[2026-08-27 04:36:00] KV cache is forced as chunk cache for decode server
+[2026-08-27 04:36:03] Breakable CUDA graph is incompatible with MLA attention; disabling prefill CUDA graph.
+[2026-08-27 04:36:03] Attention backend not specified. Use triton backend by default.
+[2026-08-27 04:36:05] server_args=ServerArgs(model_path='/data/hf_models/Kimi-K3', tokenizer_path='/data/hf_models/Kimi-K3', tokenizer_mode='auto', tokenizer_backend='huggingface', tokenizer_worker_num=1, detokenizer_worker_num=1, skip_tokenizer_init=False, load_format='auto', model_loader_extra_config='{}', trust_remote_code=True, context_length=None, is_embedding=False, enable_multimodal=None, revision=None, model_impl='auto', model_config_parser='auto', json_model_override_args='{}', dtype='auto', quantization=None, quantization_param_path=None, kv_cache_dtype='auto', enable_fp32_lm_head=False, modelopt_quant=None, modelopt_checkpoint_restore_path=None, modelopt_checkpoint_save_path=None, modelopt_export_path=None, quantize_and_serve=False, rl_quant_profile=None, enable_tf32_matmul=False, mem_fraction_static=0.88, max_running_requests=None, max_queued_requests=None, max_total_tokens=None, chunked_prefill_size=8192, enable_dynamic_chunking=False, max_prefill_tokens=16384, prefill_max_requests=None, schedule_policy='fcfs', enable_priority_scheduling=False, disable_priority_preemption=False, default_priority_value=None, abort_on_priority_when_disabled=False, schedule_low_priority_values_first=False, priority_scheduling_preemption_threshold=10, retraction_policy='length', schedule_conservativeness=1.0, page_size=64, swa_full_tokens_ratio=0.8, disable_hybrid_swa_memory=False, radix_eviction_policy='lru', prefill_only_disable_kv_cache=False, disable_radix_cache=True, enable_page_major_kv_layout=False, enable_unified_memory=False, disable_chunked_prefix_cache=False, disable_overlap_schedule=False, num_continuous_decode_steps=1, scheduler_recv_interval=1, enable_mixed_chunk=False, nccl_port=None, dist_timeout=3600, dist_init_addr='174.1.60.5:20000', nnodes=4, node_rank=3, tp_size=32, dcp_size=1, pp_size=1, pp_max_micro_batch_size=None, pp_async_batch_depth=0, dp_size=1, load_balance_method='round_robin', attn_cp_size=1, moe_dp_size=1, dcp_comm_backend='ag_rs', dcp_replicate_q_proj=None, dwdp_size=1, enable_prefill_cp=False, cp_strategy=None, enable_dsa_cache_layer_split=False, enable_dsa_prefill_context_parallel=False, dsa_prefill_cp_mode='round-robin-split', enable_prefill_context_parallel=False, prefill_cp_mode='in-seq-split', enable_cp_decode_attn_tp=False, enable_dp_attention=False, enable_dp_attention_local_control_broadcast=False, enable_dp_lm_head=False, enable_attn_tp_input_scattered=False, disable_attn_tp_gather=False, enable_p2p_check=False, device='cuda', base_gpu_id=0, gpu_id_step=1, random_seed=297832107, watchdog_timeout=300, soft_watchdog_timeout=None, sleep_on_idle=False, use_ray=False, custom_sigquit_handler=None, numa_node=None, gc_threshold=None, host='0.0.0.0', port=30000, fastapi_root_path='', smg_grpc_mode=False, grpc_mode=False, grpc_port=None, sidecar=None, sidecar_args=None, skip_server_warmup=False, warmups=None, enable_http2=False, ssl_keyfile=None, ssl_certfile=None, ssl_ca_certs=None, ssl_keyfile_password=None, enable_ssl_refresh=False, api_key=None, admin_api_key=None, served_model_name='kimi-k3', weight_version='default', chat_template=None, hf_chat_template_name=None, completion_template=None, file_storage_path='sglang_storage', enable_cache_report=False, reasoning_parser=None, default_chat_template_kwargs=None, strip_thinking_cache=False, enable_strict_thinking=False, tool_call_parser=None, tool_server=None, sampling_defaults='model', asr_max_buffer_seconds=60, asr_max_concurrent_sessions=32, preferred_sampling_params=None, allow_auto_truncate=False, stream_interval=1, batch_notify_size=16, stream_response_default_include_usage=False, incremental_streaming_output=False, enable_streaming_session=False, enable_session_radix_cache=False, log_level='info', log_level_http=None, log_requests=False, log_requests_level=2, log_requests_format='text', log_requests_target=None, uvicorn_access_log_exclude_prefixes=[], crash_dump_folder=None, show_time_cost=False, enable_metrics=False, smg_http_sidecar_port=None, enable_mfu_metrics=False, enable_metrics_for_all_schedulers=False, load_snapshot_publish_interval=15, tokenizer_metrics_custom_labels_header='x-custom-labels', tokenizer_metrics_allowed_custom_labels=None, extra_metric_labels=None, bucket_time_to_first_token=None, bucket_inter_token_latency=None, bucket_e2e_request_latency=None, prompt_tokens_buckets=None, generation_tokens_buckets=None, gc_warning_threshold_secs=0.0, decode_log_interval=40, enable_request_time_stats_logging=False, kv_events_config=None, enable_forward_pass_metrics=False, forward_pass_metrics_worker_id='', forward_pass_metrics_ipc_name=None, enable_trace=False, trace_modules='request', otlp_traces_endpoint='localhost:4317', export_metrics_to_file=False, export_metrics_to_file_dir=None, stat_loggers=None, constrained_json_whitespace_pattern=None, constrained_json_disable_any_whitespace=False, attention_backend='triton', decode_attention_backend=None, prefill_attention_backend=None, sampling_backend='flashinfer', grammar_backend='xgrammar', radix_cache_backend=None, mm_attention_backend=None, fp8_gemm_runner_backend='auto', fp4_gemm_runner_backend='auto', bf16_gemm_backend='auto', dsa_prefill_backend=None, dsa_decode_backend=None, dsa_paged_mqa_logits_backend='auto', dsa_topk_backend='sgl-kernel', disable_flashinfer_autotune=False, flashinfer_autotune_skip_ops=None, mamba_backend='triton', cuda_graph_config=CudaGraphConfig(decode=PhaseConfig(backend='full', max_bs=16, bs=[1, 2, 4, 8, 12, 16], tc_compiler='eager', full_prefill_max_req=None), prefill=PhaseConfig(backend='disabled', max_bs=2048, bs=[4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], tc_compiler='eager', full_prefill_max_req=None)), cuda_graph_backend_decode=None, cuda_graph_backend_prefill=None, cuda_graph_max_bs_decode=16, cuda_graph_max_bs_prefill=None, cuda_graph_bs_decode=None, cuda_graph_bs_prefill=None, cuda_graph_tc_compiler=None, disable_prefill_cuda_graph=False, disable_decode_cuda_graph=False, disable_cuda_graph=False, disable_cuda_graph_padding=False, enable_profile_cuda_graph=False, enable_cudagraph_gc=False, debug_cuda_graph=False, enable_layerwise_nvtx_marker=False, enable_nccl_nvls=False, enable_symm_mem=False, triton_attention_reduce_in_fp32=False, triton_attention_num_kv_splits=8, triton_attention_split_tile_size=None, flashinfer_mla_disable_ragged=False, enable_fused_qk_norm_rope=False, enable_precise_embedding_interpolation=False, enable_fused_moe_sum_all_reduce=False, enable_deepseek_v4_fp4_indexer=False, disable_custom_all_reduce=False, enable_mscclpp=False, enable_torch_symm_mem=False, enable_scattered_sconv=False, pre_warm_nccl=False, enable_quant_communications=False, enable_flashinfer_allreduce_fusion=False, enforce_disable_flashinfer_allreduce_fusion=False, flashinfer_allreduce_fusion_backend=None, enable_aiter_allreduce_fusion=False, enable_torch_compile=False, enable_torch_compile_debug_mode=False, torch_compile_max_bs=32, torchao_config='', speculative_algorithm=None, speculative_draft_model_path=None, speculative_draft_model_revision=None, speculative_draft_load_format=None, speculative_num_steps=None, speculative_eagle_topk=None, speculative_num_draft_tokens=None, speculative_dflash_block_size=None, speculative_dspark_block_size=None, speculative_dspark_sps_table_path=None, speculative_dspark_confidence_sts_path=None, speculative_dspark_align_verify_tokens_to_graph_tier=False, speculative_accept_threshold_single=1.0, speculative_accept_threshold_acc=1.0, speculative_use_rejection_sampling=False, speculative_token_map=None, speculative_attention_mode='prefill', speculative_draft_attention_backend=None, speculative_draft_window_size=None, speculative_moe_runner_backend='marlin', speculative_moe_a2a_backend=None, speculative_draft_model_quantization=None, speculative_skip_dp_mlp_sync=False, enable_multi_layer_eagle=False, speculative_adaptive=False, speculative_adaptive_config=None, decoupled_spec_bind_endpoint=None, decoupled_spec_connect_endpoints=None, decoupled_spec_rank=None, decoupled_spec_role='null', spec_trace_dir=None, speculative_ngram_min_bfs_breadth=1, speculative_ngram_max_bfs_breadth=10, speculative_ngram_match_type='BFS', speculative_ngram_max_trie_depth=18, speculative_ngram_capacity=10000000, speculative_ngram_external_corpus_path=None, speculative_ngram_external_sam_budget=0, speculative_ngram_external_corpus_max_tokens=10000000, ep_size=32, moe_a2a_backend='none', moe_runner_backend='marlin', flashinfer_mxfp4_moe_precision='default', deepep_mode='auto', fuseep_mode=2, deepep_dispatcher_output_dtype='auto', ep_num_redundant_experts=0, ep_dispatch_algorithm=None, init_expert_location='trivial', enable_eplb=False, eplb_algorithm='auto', eplb_rebalance_num_iterations=1000, eplb_rebalance_layers_per_chunk=None, eplb_min_rebalancing_utilization_threshold=1.0, expert_distribution_recorder_mode=None, expert_distribution_recorder_buffer_size=1000, enable_expert_distribution_metrics=False, deepep_config=None, moe_dense_tp_size=None, elastic_ep_backend=None, enable_elastic_expert_backup=False, mooncake_ib_device=None, enable_waterfill=False, ep_join_mode=None, ep_join_rank_offset=0, elastic_ep_initial_size=None, max_ep_size=None, elastic_ep_scale_timeout=600, elastic_ep_rejoin=False, disable_flashinfer_cutlass_moe_fp4_allgather=False, disable_shared_experts_fusion=False, enforce_shared_experts_fusion=False, max_mamba_cache_size=None, mamba_ssm_dtype=None, mamba_max_states_per_path=-1, enable_mamba_cache_stochastic_rounding=False, mamba_cache_philox_rounds=0, mamba_full_memory_ratio=0.21, mamba_radix_cache_strategy='extra_buffer_lazy', uses_mamba_radix_cache=False, mamba_track_interval=256, enable_int8_mamba_checkpoint=False, int8_mamba_ckpt_size=None, linear_attn_backend='triton', linear_attn_decode_backend=None, linear_attn_prefill_backend=None, linear_attn_verify_backend=None, enable_linear_replayssm=False, linear_replayssm_cache_len=16, enable_linear_replayssm_spec=False, enable_hierarchical_cache=False, hicache_ratio=2.0, hicache_size=0, hicache_write_policy='write_through', hicache_io_backend='kernel', hicache_mem_layout='page_first', hicache_storage_backend=None, hicache_storage_prefetch_policy='timeout', hicache_storage_backend_extra_config=None, enable_hisparse=False, hisparse_config=None, enable_broadcast_mm_inputs_process=False, enable_prefix_mm_cache=False, mm_enable_dp_encoder=False, mm_process_config={}, mm_processor_worker_num=0, mm_io_worker_num=0, limit_mm_data_per_request=None, enable_mm_global_cache=False, disable_fast_image_processor=False, mm_feature_transport='cpu', keep_mm_feature_on_device=False, enable_lora=None, enable_lora_overlap_loading=None, max_lora_rank=None, lora_target_modules=None, lora_paths=None, max_loaded_loras=None, max_loras_per_batch=8, lora_eviction_policy='lru', lora_backend='csgmv', max_lora_chunk_size=16, experts_shared_outer_loras=None, lora_use_virtual_experts=False, lora_strict_loading=False, lora_drain_wait_threshold=0.0, enable_two_batch_overlap=False, enable_single_batch_overlap=False, tbo_token_distribution_threshold=0.48, cpu_offload_gb=0, offload_group_size=-1, offload_num_in_group=1, offload_prefetch_step=1, offload_mode='cpu', enable_lmcache=False, lmcache_config_file=None, enable_flexkv=False, flexkv_config_file=None, kt_weight_path=None, kt_method='AMXINT4', kt_cpuinfer=None, kt_threadpool_count=2, kt_num_gpu_experts=None, kt_max_deferred_experts_per_token=None, dllm_algorithm=None, dllm_algorithm_config=None, dllm_fdfo=True, disaggregation_mode='decode', disaggregation_transfer_backend='mooncake', disaggregation_bootstrap_port=28800, disaggregation_ib_device='mlx5_0,mlx5_1,mlx5_2,mlx5_3', disaggregation_decode_enable_radix_cache=False, disaggregation_decode_enable_offload_kvcache=False, num_reserved_decode_tokens=512, disaggregation_decode_extra_slots=0, disaggregation_decode_polling_interval=1, optimistic_prefill_attempts=0, encoder_only=False, language_only=False, encoder_transfer_backend='zmq_to_tokenizer', encoder_urls=[], encoder_bootstrap_port=8997, encoder_register_urls=[], enable_adaptive_dispatch_to_encoder=False, enable_pdmux=False, pdmux_config_path=None, sm_group_num=8, custom_weight_loader=[], weight_loader_disable_mmap=False, weight_loader_prefetch_checkpoints=False, weight_loader_prefetch_num_threads=4, weight_loader_drop_cache_after_load=False, remote_instance_weight_loader_seed_instance_ip=None, remote_instance_weight_loader_seed_instance_service_port=None, remote_instance_weight_loader_send_weights_group_ports=None, remote_instance_weight_loader_backend='nccl', remote_instance_weight_loader_start_seed_via_transfer_engine=False, engine_info_bootstrap_port=6789, modelexpress_config=None, download_dir=None, model_checksum=None, delete_ckpt_after_loading=False, decrypted_config_file=None, decrypted_draft_config_file=None, checkpoint_engine_wait_weights_before_ready=False, enable_prefill_delayer=False, prefill_delayer_max_delay_passes=30, prefill_delayer_token_usage_low_watermark=None, prefill_delayer_forward_passes_buckets=None, prefill_delayer_wait_seconds_buckets=None, prefill_delayer_queue_min_ratio=None, prefill_delayer_max_delay_ms=None, min_free_slots_delay=None, enable_deterministic_inference=False, rl_on_policy_target=None, kv_canary='none', kv_canary_real_data='none', kv_canary_sweep_interval=0, enable_dynamic_batch_tokenizer=False, dynamic_batch_tokenizer_batch_size=32, dynamic_batch_tokenizer_batch_timeout=0.002, enable_tokenizer_batch_encode=False, disable_tokenizer_batch_decode=False, debug_tensor_dump_output_folder=None, debug_tensor_dump_layers=None, debug_tensor_dump_input_file=None, enable_memory_saver=False, enable_weights_cpu_backup=False, enable_draft_weights_cpu_backup=False, enable_custom_logit_processor=False, enable_return_hidden_states=False, enable_return_routed_experts=False, enable_return_indexer_topk=False, disable_outlines_disk_cache=False, enable_mis=False, weight_cache_mode='off', weight_cache_socket=None, weight_cache_timeout=1800, forward_hooks=None, msprobe_dump_config=None)
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:16 TP30 EP30] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:16 TP30 EP30] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:19 TP30 EP30] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:19.321749   241 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:19.321810   241 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:19.321820   241 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.8 port: 12001
+I0827 04:36:19.321856   241 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.8:16569
+I0827 04:36:19.321940   241 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:19.343153   241 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:19.347820   241 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:19.373265   241 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:19.380316   241 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:19.383023   241 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:19.383050   241 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:19.383101   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:19.390141   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:19.391583   241 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:08
+I0827 04:36:19.391615   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:19.417495   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:19.418977   241 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:08
+I0827 04:36:19.419003   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:19.445492   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:19.446861   241 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:08
+I0827 04:36:19.446884   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:19.473483   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:19.474817   241 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:08
+I0827 04:36:19.474922   241 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:19 TP30 EP30] Init torch distributed begin.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:20 TP27 EP27] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:20 TP27 EP27] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:21 TP25 EP25] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:21 TP25 EP25] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:21 TP26 EP26] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:21 TP26 EP26] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:21 TP28 EP28] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:21 TP28 EP28] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:21 TP24 EP24] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:21 TP24 EP24] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:21 TP29 EP29] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:21 TP29 EP29] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:21 TP31 EP31] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:21 TP31 EP31] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:22 TP27 EP27] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:22.734625   238 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:22.734687   238 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:22.734699   238 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.8 port: 12001
+I0827 04:36:22.734735   238 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.8:16566
+I0827 04:36:22.734848   238 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:22.760079   238 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:22.768187   238 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:22.777189   238 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:22.786154   238 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:22.789806   238 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:22.789830   238 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:22.789880   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:22.797986   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:22.799417   238 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:08
+I0827 04:36:22.799446   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:22.825555   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:22.826912   238 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:08
+I0827 04:36:22.826946   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:22.853536   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:22.854885   238 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:08
+I0827 04:36:22.854909   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:22.881553   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:22.882910   238 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:08
+I0827 04:36:22.883030   238 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:22 TP27 EP27] Init torch distributed begin.
+[2026-08-27 04:36:22 TP25 EP25] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:22 TP28 EP28] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:22 TP26 EP26] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:23 TP29 EP29] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:23 TP24 EP24] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:23.476856   236 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:23.476951   236 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:23.476967   236 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.8 port: 12001
+I0827 04:36:23.477013   236 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.8:15425
+I0827 04:36:23.477150   236 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:23.477173   239 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:23.477272   239 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:23.477288   239 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.8 port: 12001
+I0827 04:36:23.477344   239 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.8:15710
+I0827 04:36:23.477483   239 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:23.483197   236 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:23.483803   239 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:23.488940   236 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:23.489802   239 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:23.494866   236 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:23.495663   239 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:23.500973   236 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:23.501806   239 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:23.503927   236 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:23.503952   236 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:23.503998   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.504750   239 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:23.504770   239 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:23.504812   239 rdma_context.cpp:205] Using SIEVE endpoint store
+[2026-08-27 04:36:23 TP31 EP31] Multimodal data loading enabled with 16 worker threads (auto).
+I0827 04:36:23.510515   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:23.511322   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:23.511830   236 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:08
+I0827 04:36:23.511855   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.512616   239 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:08
+I0827 04:36:23.512640   239 rdma_context.cpp:205] Using SIEVE endpoint store
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:23.522899   237 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:23.523005   237 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:23.523020   237 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.8 port: 12001
+I0827 04:36:23.523065   237 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.8:16748
+I0827 04:36:23.523202   237 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:23.529469   237 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:23.535485   237 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:23.537460   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:23.537528   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:23.538903   236 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:08
+I0827 04:36:23.538913   239 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:08
+I0827 04:36:23.538928   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.538945   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.541347   237 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:23.546439   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:23.547672   237 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:23.547796   239 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:08
+I0827 04:36:23.547820   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.550041   237 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:23.550061   237 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:23.550102   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.555119   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:23.556490   239 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:08
+I0827 04:36:23.556598   239 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:23 TP28 EP28] Init torch distributed begin.
+I0827 04:36:23.557152   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:23.558562   237 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:08
+I0827 04:36:23.558589   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.565423   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:23.566766   236 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:08
+I0827 04:36:23.566788   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.589463   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:23.590811   237 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:08
+I0827 04:36:23.590835   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.593444   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:23.594800   236 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:08
+I0827 04:36:23.594898   236 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:23 TP25 EP25] Init torch distributed begin.
+I0827 04:36:23.598161   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:23.599563   237 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:08
+I0827 04:36:23.599589   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.625564   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:23.626922   237 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:08
+I0827 04:36:23.627038   237 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:23 TP26 EP26] Init torch distributed begin.
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:23.899124   240 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:23.899245   240 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:23.899261   240 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.8 port: 12001
+I0827 04:36:23.899313   240 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.8:15127
+I0827 04:36:23.899492   240 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:23.907753   235 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:23.907863   235 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:23.907878   235 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.8 port: 12001
+I0827 04:36:23.907927   235 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.8:15330
+I0827 04:36:23.908092   235 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:23.914011   235 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:23.919618   235 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:23.925108   235 topology.cpp:127] Device mlx5_2 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:23.925450   242 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:23.925558   242 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:23.925576   242 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.8 port: 12001
+I0827 04:36:23.925631   242 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.8:16235
+I0827 04:36:23.925693   240 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:23.925761   242 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:23.930419   242 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:23.933247   240 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:23.934639   242 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:23.939190   242 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:23.940605   240 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:23.943866   242 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:23.945531   242 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:23.945554   242 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:23.945595   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.947860   240 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:23.949263   235 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:23.951108   240 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:23.951148   240 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:23.951216   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.952417   235 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:23.952445   235 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:23.952502   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.958555   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:23.959836   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:23.960163   240 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:08
+I0827 04:36:23.960197   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.961452   235 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:08
+I0827 04:36:23.961483   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.969228   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:23.969322   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:23.970860   235 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:08
+I0827 04:36:23.970890   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.971050   242 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:08
+I0827 04:36:23.971084   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.979120   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:23.980706   242 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:08
+I0827 04:36:23.980736   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.989722   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:23.991313   240 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:08
+I0827 04:36:23.991346   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:24.001633   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:24.003211   235 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:08
+I0827 04:36:24.003237   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:24.005753   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:24.007426   242 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:08
+I0827 04:36:24.007457   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:24.021715   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:24.023293   240 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:08
+I0827 04:36:24.023322   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:24.034039   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:24.034071   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:24.035696   235 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:08
+I0827 04:36:24.035816   235 transfer_engine_impl.cpp:388] installTransport, type=rdma
+I0827 04:36:24.035794   242 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:08
+I0827 04:36:24.035987   242 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:24 TP24 EP24] Init torch distributed begin.
+[2026-08-27 04:36:24 TP31 EP31] Init torch distributed begin.
+I0827 04:36:24.049829   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:24.051388   240 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:08
+I0827 04:36:24.051532   240 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:24 TP29 EP29] Init torch distributed begin.
+[2026-08-27 04:36:48 TP31 EP31] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP30 EP30] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP28 EP28] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP29 EP29] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP27 EP27] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP26 EP26] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP25 EP25] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP24 EP24] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP31 EP31] Init torch distributed ends. elapsed=24.20 s, mem usage=0.40 GB
+[2026-08-27 04:36:48 TP30 EP30] Init torch distributed ends. elapsed=28.76 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP29 EP29] Init torch distributed ends. elapsed=24.18 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP28 EP28] Init torch distributed ends. elapsed=24.68 s, mem usage=0.40 GB
+[2026-08-27 04:36:48 TP27 EP27] Init torch distributed ends. elapsed=25.35 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP26 EP26] Init torch distributed ends. elapsed=24.61 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP25 EP25] Init torch distributed ends. elapsed=24.64 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP24 EP24] Init torch distributed ends. elapsed=24.20 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP28 EP28] Load weight begin. avail mem=82.19 GB
+[2026-08-27 04:36:48 TP27 EP27] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP30 EP30] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP24 EP24] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP29 EP29] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP26 EP26] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP25 EP25] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP28 EP28] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP30 EP30] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP27 EP27] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP29 EP29] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP24 EP24] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP26 EP26] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP25 EP25] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP31 EP31] Load weight begin. avail mem=82.19 GB
+[2026-08-27 04:36:48 TP31 EP31] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP27 EP27] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP28 EP28] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP24 EP24] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP27 EP27] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP30 EP30] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP29 EP29] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP26 EP26] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP25 EP25] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP28 EP28] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP24 EP24] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP29 EP29] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP31 EP31] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP30 EP30] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP26 EP26] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP25 EP25] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP31 EP31] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:50 TP29 EP29] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP28 EP28] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP30 EP30] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP26 EP26] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP25 EP25] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP31 EP31] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP27 EP27] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP24 EP24] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:38:54 TP26 EP26] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:54 TP27 EP27] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:54 TP24 EP24] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:54 TP26 EP26] Load weight end. elapsed=125.61 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:54 TP29 EP29] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:54 TP27 EP27] Load weight end. elapsed=125.61 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:54 TP24 EP24] Load weight end. elapsed=125.61 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:54 TP29 EP29] Load weight end. elapsed=125.61 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:54 TP25 EP25] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:54 TP25 EP25] Load weight end. elapsed=125.64 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:54 TP28 EP28] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:54 TP28 EP28] Load weight end. elapsed=125.69 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:54 TP30 EP30] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:54 TP30 EP30] Load weight end. elapsed=125.71 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:54 TP31 EP31] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:54 TP31 EP31] Load weight end. elapsed=125.77 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:39:01 TP26 EP26] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP26 EP26] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP30 EP30] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP30 EP30] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP28 EP28] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP28 EP28] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP27 EP27] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP27 EP27] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP25 EP25] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP25 EP25] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP31 EP31] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP31 EP31] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP24 EP24] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP24 EP24] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP29 EP29] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP29 EP29] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:02 TP29 EP29] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP28 EP28] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP26 EP26] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP27 EP27] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP30 EP30] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP25 EP25] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP31 EP31] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP24 EP24] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP29 EP29] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP30 EP30] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP28 EP28] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP26 EP26] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP27 EP27] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP25 EP25] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP24 EP24] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP31 EP31] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP29 EP29] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP29 EP29] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP28 EP28] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP28 EP28] Memory pool end. avail mem=9.18 GB
+[2026-08-27 04:39:02 TP27 EP27] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP24 EP24] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP26 EP26] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP27 EP27] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP31 EP31] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP24 EP24] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP26 EP26] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP31 EP31] Memory pool end. avail mem=9.18 GB
+[2026-08-27 04:39:02 TP30 EP30] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP25 EP25] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP30 EP30] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP25 EP25] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP29 EP29] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP29 EP29] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP24 EP24] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP24 EP24] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:39:02 TP28 EP28] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP28 EP28] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.17 GB
+[2026-08-27 04:39:02 TP25 EP25] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP31 EP31] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP25 EP25] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP31 EP31] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.17 GB
+[2026-08-27 04:39:02 TP26 EP26] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP26 EP26] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP27 EP27] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP27 EP27] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:39:02 TP30 EP30] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP30 EP30] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:42:56 TP30 EP30] Capture target decode CUDA graph end. elapsed=233.69 s, mem usage=0.56 GB, avail mem=8.59 GB.
+[2026-08-27 04:42:56 TP30 EP30] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP27 EP27] Capture target decode CUDA graph end. elapsed=233.73 s, mem usage=0.62 GB, avail mem=8.51 GB.
+[2026-08-27 04:42:56 TP27 EP27] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP31 EP31] Capture target decode CUDA graph end. elapsed=233.74 s, mem usage=0.62 GB, avail mem=8.55 GB.
+[2026-08-27 04:42:56 TP31 EP31] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP26 EP26] Capture target decode CUDA graph end. elapsed=233.76 s, mem usage=0.56 GB, avail mem=8.59 GB.
+[2026-08-27 04:42:56 TP26 EP26] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP29 EP29] Capture target decode CUDA graph end. elapsed=233.82 s, mem usage=0.58 GB, avail mem=8.58 GB.
+[2026-08-27 04:42:56 TP29 EP29] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP28 EP28] Capture target decode CUDA graph end. elapsed=233.79 s, mem usage=0.60 GB, avail mem=8.56 GB.
+[2026-08-27 04:42:56 TP28 EP28] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP24 EP24] Capture target decode CUDA graph end. elapsed=233.88 s, mem usage=0.60 GB, avail mem=8.52 GB.
+[2026-08-27 04:42:56 TP24 EP24] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP25 EP25] Capture target decode CUDA graph end. elapsed=233.89 s, mem usage=0.58 GB, avail mem=8.58 GB.
+[2026-08-27 04:42:56 TP25 EP25] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+W0827 04:42:57.132246  2431 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5913c7c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.132292  2432 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58be65c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.132460  2435 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5af7a600, len: 185344: Operation not permitted [1]
+W0827 04:42:57.132488  2433 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58ed5900, len: 23168: Operation not permitted [1]
+W0827 04:42:57.132501  2436 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5afa7a40, len: 185344: Operation not permitted [1]
+W0827 04:42:57.132512  2437 memory_location.cpp:73] Failed to get NUMA node, addr: 0x590583c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.132552  2434 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58c15080, len: 23168: Operation not permitted [1]
+W0827 04:42:57.132607  2439 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59046600, len: 23168: Operation not permitted [1]
+W0827 04:42:57.132640  2438 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5afd4e80, len: 46336: Operation not permitted [1]
+W0827 04:42:57.132716  2440 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58e5fc80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.276726  2580 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a339340, len: 23168: Operation not permitted [1]
+W0827 04:42:57.276731  2579 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59705880, len: 23168: Operation not permitted [1]
+W0827 04:42:57.276762  2581 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a32b080, len: 23168: Operation not permitted [1]
+W0827 04:42:57.276839  2582 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a331fc0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.276887  2583 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5bb81840, len: 185344: Operation not permitted [1]
+W0827 04:42:57.276939  2584 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5bbaec80, len: 185344: Operation not permitted [1]
+W0827 04:42:57.276983  2585 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59b7b6c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.277042  2587 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59953f00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.277055  2586 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59517380, len: 46336: Operation not permitted [1]
+W0827 04:42:57.277086  2588 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59b87640, len: 23168: Operation not permitted [1]
+W0827 04:42:57.448414  2727 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a4bd3c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.448436  2728 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ac9eb00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.448488  2729 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a2fe240, len: 23168: Operation not permitted [1]
+W0827 04:42:57.448798  2735 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a6135c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.448812  2734 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59ed25c0, len: 46336: Operation not permitted [1]
+W0827 04:42:57.448840  2733 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a3c3080, len: 23168: Operation not permitted [1]
+W0827 04:42:57.448849  2736 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a181e80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.451195  2730 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5acaf240, len: 23168: Operation not permitted [1]
+W0827 04:42:57.451217  2731 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5c61b840, len: 185344: Operation not permitted [1]
+W0827 04:42:57.455173  2732 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5c648c80, len: 185344: Operation not permitted [1]
+W0827 04:42:57.460536  2816 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a391f40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.460651  2817 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ae89700, len: 23168: Operation not permitted [1]
+W0827 04:42:57.460843  2819 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ae0fa00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.460901  2826 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a4e2c80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.460979  2830 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ae7a300, len: 23168: Operation not permitted [1]
+W0827 04:42:57.461052  2821 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5c6d6880, len: 185344: Operation not permitted [1]
+W0827 04:42:57.461104  2820 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a22ed40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.462260  2823 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5c703cc0, len: 185344: Operation not permitted [1]
+W0827 04:42:57.463069  2827 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4678c140, len: 46336: Operation not permitted [1]
+W0827 04:42:57.463085  2829 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a645e80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.466511  2955 memory_location.cpp:73] Failed to get NUMA node, addr: 0x597592c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.466544  2958 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59908b40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.466676  2960 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a2f9c80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.466849  2953 memory_location.cpp:73] Failed to get NUMA node, addr: 0x599f9a80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.467304  2964 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5b9e4d40, len: 185344: Operation not permitted [1]
+W0827 04:42:57.470024  2966 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5993fa80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.471112  2970 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59898900, len: 23168: Operation not permitted [1]
+W0827 04:42:57.471313  2962 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5b9b7900, len: 185344: Operation not permitted [1]
+W0827 04:42:57.475107  2968 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5991bac0, len: 46336: Operation not permitted [1]
+W0827 04:42:57.475121  2972 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59b5ea40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.493731  3089 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59564740, len: 23168: Operation not permitted [1]
+W0827 04:42:57.493901  3097 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5b355380, len: 185344: Operation not permitted [1]
+W0827 04:42:57.494069  3104 memory_location.cpp:73] Failed to get NUMA node, addr: 0x45dbac00, len: 46336: Operation not permitted [1]
+W0827 04:42:57.494072  3105 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5902f500, len: 23168: Operation not permitted [1]
+W0827 04:42:57.494150  3094 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59498600, len: 23168: Operation not permitted [1]
+W0827 04:42:57.494194  3099 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5b3827c0, len: 185344: Operation not permitted [1]
+W0827 04:42:57.499953  3107 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59486740, len: 23168: Operation not permitted [1]
+W0827 04:42:57.502707  3102 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59016c80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.504236  3091 memory_location.cpp:73] Failed to get NUMA node, addr: 0x594d2200, len: 23168: Operation not permitted [1]
+W0827 04:42:57.504359  3095 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59354e40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.514688  3200 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5885e940, len: 23168: Operation not permitted [1]
+W0827 04:42:57.514716  3202 memory_location.cpp:73] Failed to get NUMA node, addr: 0x587d55c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.514720  3201 memory_location.cpp:73] Failed to get NUMA node, addr: 0x586bd5c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.514801  3204 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5aa56c00, len: 185344: Operation not permitted [1]
+W0827 04:42:57.514810  3205 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5aa84040, len: 185344: Operation not permitted [1]
+W0827 04:42:57.514727  3203 memory_location.cpp:73] Failed to get NUMA node, addr: 0x44d09c40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.514963  3206 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58692200, len: 23168: Operation not permitted [1]
+W0827 04:42:57.515017  3209 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58865100, len: 23168: Operation not permitted [1]
+W0827 04:42:57.515022  3207 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58ae8bc0, len: 46336: Operation not permitted [1]
+W0827 04:42:57.515030  3208 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58683a40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.548924  3433 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5895be80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.548955  3434 memory_location.cpp:73] Failed to get NUMA node, addr: 0x593aa600, len: 23168: Operation not permitted [1]
+W0827 04:42:57.549010  3435 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59399e80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.549088  3436 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58a4fa40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.549110  3437 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ab36800, len: 185344: Operation not permitted [1]
+W0827 04:42:57.549167  3438 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ab63c40, len: 185344: Operation not permitted [1]
+W0827 04:42:57.551182  3439 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5936aac0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.551239  3441 memory_location.cpp:73] Failed to get NUMA node, addr: 0x455ad780, len: 23168: Operation not permitted [1]
+W0827 04:42:57.551265  3442 memory_location.cpp:73] Failed to get NUMA node, addr: 0x594169c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.551280  3440 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5853c980, len: 46336: Operation not permitted [1]
+[2026-08-27 04:42:59] Dummy health check server started in background thread at 0.0.0.0:30000
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/ready_mc_master.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/ready_mc_master.log
new file mode 100644
index 0000000..cf99d27
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/ready_mc_master.log
@@ -0,0 +1,94 @@
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+I0827 04:28:16.835302    99 master.cpp:1366] Master service started on port 50051, max_threads=16, enable_metric_reporting=1, metrics_port=9003, default_kv_lease_ttl=10000, default_kv_soft_pin_ttl=1800000, allow_evict_soft_pinned_objects=1, eviction_ratio=0.05, eviction_high_watermark_ratio=0.9, enable_ha=0, enable_offload=0, enable_kv_events=0, kv_events_bind_endpoint=, kv_events_backend_id=, offload_on_evict=0, offload_force_evict=0, offloading_queue_limit=50000, offload_cap_ratio=0.5, ha_backend_type=etcd, ha_backend_connstring=, etcd_endpoints=, client_ttl=10, rpc_thread_num=16, rpc_port=50051, rpc_address=0.0.0.0, rpc_interface=, rpc_conn_timeout_seconds=0, rpc_enable_tcp_no_delay=1, rpc protocol=tcp, cluster_id=mooncake_cluster, root_fs_dir=, global_file_segment_size=9223372036854775807, memory_allocator=offset, enable_http_metadata_server=0, http_metadata_server_port=8080, http_metadata_server_host=0.0.0.0, enable_metadata_cleanup_on_timeout=0, put_start_discard_timeout_sec=30, put_start_release_timeout_sec=600, max_total_finished_tasks=10000, max_total_pending_tasks=10000, max_total_processing_tasks=10000, pending_task_timeout_sec=300, processing_task_timeout_sec=300, enable_snapshot=0, enable_snapshot_restore=0, snapshot_interval_seconds=600, snapshot_backup_dir=, snapshot_object_store_type=, snapshot_catalog_store_type=, snapshot_retention_count=2, max_retry_attempts=10, enable_cxl=0, cxl_path=/dev/dax0.0, cxl_size=8589934592
+I0827 04:28:16.866606   108 master_service.cpp:1520] Task cleanup thread started
+I0827 04:28:16.867491    99 master_admin_service.cpp:311] Master admin server started on port 9003
+I0827 04:28:16.869802   111 master_admin_service.cpp:302] Master Admin Metrics: role=standby, state=starting, service_ready=false, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:28:26.870203   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:28:36.870568   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:28:46.870873   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:28:56.871232   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:29:06.871615   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:29:16.871969   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:29:26.872316   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:29:36.872687   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:29:46.873068   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:29:56.873467   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:30:06.873914   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:30:16.874214   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:30:26.874567   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:30:36.874953   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:30:46.875332   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:30:56.875741   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:31:06.876119   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:31:16.876497   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:31:26.876866   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:31:36.877233   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:31:46.877605   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:31:56.877948   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:32:06.878437   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:32:16.879210   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:32:26.879711   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:32:36.880080   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:32:46.880565   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:32:56.881004   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:33:06.881398   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:33:16.881786   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:33:26.882170   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:33:36.882550   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:33:46.882939   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:33:56.883315   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:34:06.883746   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:34:16.884127   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:34:26.884568   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:34:36.884953   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:34:46.885334   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:34:56.885706   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:35:06.886085   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:35:16.886464   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:35:26.886835   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:35:36.887218   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:35:46.887596   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:35:56.887964   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:36:06.888356   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:36:16.888742   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:36:26.889113   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:36:36.889495   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:36:46.889882   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:36:56.890261   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:37:06.890656   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:37:16.891040   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:37:26.891430   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:37:36.891836   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:37:46.892215   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:37:56.892597   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:38:06.892982   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:38:16.893360   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:38:26.893747   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:38:36.894131   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:38:46.894515   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:38:56.894920   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:39:06.895303   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:39:16.895701   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:39:26.896085   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:39:36.896471   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:39:46.896852   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:39:56.897240   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:40:06.897629   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:40:16.898005   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:40:26.898384   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:40:36.898761   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:40:46.899144   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:40:56.899530   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:41:06.899937   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:41:16.900316   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:41:26.900695   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:41:36.901089   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:41:46.901479   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:41:56.901856   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:42:06.902242   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:42:16.902666   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:42:26.903064   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:42:36.903537   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:42:46.903987   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:42:56.904397   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:43:06.904863   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/ready_p_node0.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/ready_p_node0.log
new file mode 100644
index 0000000..85cfdec
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/ready_p_node0.log
@@ -0,0 +1,941 @@
+/usr/local/lib/python3.12/dist-packages/torchao/quantization/quant_api.py:1731: SyntaxWarning: invalid escape sequence '\.'
+  """Configuration class for applying different quantization configs to modules or parameters based on their fully qualified names (FQNs).
+/sgl-workspace/sglang/python/sglang/launch_server.py:56: UserWarning: 'python -m sglang.launch_server' is still supported, but 'sglang serve' is the recommended entrypoint.
+  Example: sglang serve --model-path <model> [options]
+  warnings.warn(
+[2026-08-27 04:28:54] Breakable CUDA graph is incompatible with MLA attention; disabling prefill CUDA graph.
+[2026-08-27 04:28:55] Attention backend not specified. Use triton backend by default.
+[2026-08-27 04:28:55] Pipeline parallelism is incompatible with overlap schedule.
+[2026-08-27 04:28:58] server_args=ServerArgs(model_path='/data/hf_models/Kimi-K3', tokenizer_path='/data/hf_models/Kimi-K3', tokenizer_mode='auto', tokenizer_backend='huggingface', tokenizer_worker_num=1, detokenizer_worker_num=1, skip_tokenizer_init=False, load_format='auto', model_loader_extra_config='{}', trust_remote_code=True, context_length=None, is_embedding=False, enable_multimodal=None, revision=None, model_impl='auto', model_config_parser='auto', json_model_override_args='{}', dtype='auto', quantization=None, quantization_param_path=None, kv_cache_dtype='auto', enable_fp32_lm_head=False, modelopt_quant=None, modelopt_checkpoint_restore_path=None, modelopt_checkpoint_save_path=None, modelopt_export_path=None, quantize_and_serve=False, rl_quant_profile=None, enable_tf32_matmul=False, mem_fraction_static=0.88, max_running_requests=None, max_queued_requests=None, max_total_tokens=None, chunked_prefill_size=8192, enable_dynamic_chunking=False, max_prefill_tokens=16384, prefill_max_requests=None, schedule_policy='fcfs', enable_priority_scheduling=False, disable_priority_preemption=False, default_priority_value=None, abort_on_priority_when_disabled=False, schedule_low_priority_values_first=False, priority_scheduling_preemption_threshold=10, retraction_policy='length', schedule_conservativeness=1.0, page_size=64, swa_full_tokens_ratio=0.8, disable_hybrid_swa_memory=False, radix_eviction_policy='lru', prefill_only_disable_kv_cache=False, disable_radix_cache=True, enable_page_major_kv_layout=False, enable_unified_memory=False, disable_chunked_prefix_cache=False, disable_overlap_schedule=True, num_continuous_decode_steps=1, scheduler_recv_interval=1, enable_mixed_chunk=False, nccl_port=None, dist_timeout=3600, dist_init_addr='174.1.60.1:20000', nnodes=4, node_rank=0, tp_size=4, dcp_size=1, pp_size=8, pp_max_micro_batch_size=None, pp_async_batch_depth=0, dp_size=1, load_balance_method='follow_bootstrap_room', attn_cp_size=1, moe_dp_size=1, dcp_comm_backend='ag_rs', dcp_replicate_q_proj=None, dwdp_size=1, enable_prefill_cp=False, cp_strategy=None, enable_dsa_cache_layer_split=False, enable_dsa_prefill_context_parallel=False, dsa_prefill_cp_mode='round-robin-split', enable_prefill_context_parallel=False, prefill_cp_mode='in-seq-split', enable_cp_decode_attn_tp=False, enable_dp_attention=False, enable_dp_attention_local_control_broadcast=False, enable_dp_lm_head=False, enable_attn_tp_input_scattered=False, disable_attn_tp_gather=False, enable_p2p_check=False, device='cuda', base_gpu_id=0, gpu_id_step=1, random_seed=984607766, watchdog_timeout=300, soft_watchdog_timeout=None, sleep_on_idle=False, use_ray=False, custom_sigquit_handler=None, numa_node=None, gc_threshold=None, host='0.0.0.0', port=30000, fastapi_root_path='', smg_grpc_mode=False, grpc_mode=False, grpc_port=None, sidecar=None, sidecar_args=None, skip_server_warmup=False, warmups=None, enable_http2=False, ssl_keyfile=None, ssl_certfile=None, ssl_ca_certs=None, ssl_keyfile_password=None, enable_ssl_refresh=False, api_key=None, admin_api_key=None, served_model_name='kimi-k3', weight_version='default', chat_template=None, hf_chat_template_name=None, completion_template=None, file_storage_path='sglang_storage', enable_cache_report=False, reasoning_parser=None, default_chat_template_kwargs=None, strip_thinking_cache=False, enable_strict_thinking=False, tool_call_parser=None, tool_server=None, sampling_defaults='model', asr_max_buffer_seconds=60, asr_max_concurrent_sessions=32, preferred_sampling_params=None, allow_auto_truncate=False, stream_interval=1, batch_notify_size=16, stream_response_default_include_usage=False, incremental_streaming_output=False, enable_streaming_session=False, enable_session_radix_cache=False, log_level='info', log_level_http=None, log_requests=False, log_requests_level=2, log_requests_format='text', log_requests_target=None, uvicorn_access_log_exclude_prefixes=[], crash_dump_folder=None, show_time_cost=False, enable_metrics=False, smg_http_sidecar_port=None, enable_mfu_metrics=False, enable_metrics_for_all_schedulers=False, load_snapshot_publish_interval=15, tokenizer_metrics_custom_labels_header='x-custom-labels', tokenizer_metrics_allowed_custom_labels=None, extra_metric_labels=None, bucket_time_to_first_token=None, bucket_inter_token_latency=None, bucket_e2e_request_latency=None, prompt_tokens_buckets=None, generation_tokens_buckets=None, gc_warning_threshold_secs=0.0, decode_log_interval=40, enable_request_time_stats_logging=False, kv_events_config=None, enable_forward_pass_metrics=False, forward_pass_metrics_worker_id='', forward_pass_metrics_ipc_name=None, enable_trace=False, trace_modules='request', otlp_traces_endpoint='localhost:4317', export_metrics_to_file=False, export_metrics_to_file_dir=None, stat_loggers=None, constrained_json_whitespace_pattern=None, constrained_json_disable_any_whitespace=False, attention_backend='triton', decode_attention_backend=None, prefill_attention_backend=None, sampling_backend='flashinfer', grammar_backend='xgrammar', radix_cache_backend=None, mm_attention_backend=None, fp8_gemm_runner_backend='auto', fp4_gemm_runner_backend='auto', bf16_gemm_backend='auto', dsa_prefill_backend=None, dsa_decode_backend=None, dsa_paged_mqa_logits_backend='auto', dsa_topk_backend='sgl-kernel', disable_flashinfer_autotune=False, flashinfer_autotune_skip_ops=None, mamba_backend='triton', cuda_graph_config=CudaGraphConfig(decode=PhaseConfig(backend='disabled', max_bs=16, bs=[1, 2, 4, 8, 12, 16], tc_compiler='eager', full_prefill_max_req=None), prefill=PhaseConfig(backend='disabled', max_bs=2048, bs=[4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], tc_compiler='eager', full_prefill_max_req=None)), cuda_graph_backend_decode=None, cuda_graph_backend_prefill=None, cuda_graph_max_bs_decode=16, cuda_graph_max_bs_prefill=None, cuda_graph_bs_decode=None, cuda_graph_bs_prefill=None, cuda_graph_tc_compiler=None, disable_prefill_cuda_graph=False, disable_decode_cuda_graph=False, disable_cuda_graph=False, disable_cuda_graph_padding=False, enable_profile_cuda_graph=False, enable_cudagraph_gc=False, debug_cuda_graph=False, enable_layerwise_nvtx_marker=False, enable_nccl_nvls=False, enable_symm_mem=False, triton_attention_reduce_in_fp32=False, triton_attention_num_kv_splits=8, triton_attention_split_tile_size=None, flashinfer_mla_disable_ragged=False, enable_fused_qk_norm_rope=False, enable_precise_embedding_interpolation=False, enable_fused_moe_sum_all_reduce=False, enable_deepseek_v4_fp4_indexer=False, disable_custom_all_reduce=False, enable_mscclpp=False, enable_torch_symm_mem=False, enable_scattered_sconv=False, pre_warm_nccl=False, enable_quant_communications=False, enable_flashinfer_allreduce_fusion=False, enforce_disable_flashinfer_allreduce_fusion=False, flashinfer_allreduce_fusion_backend=None, enable_aiter_allreduce_fusion=False, enable_torch_compile=False, enable_torch_compile_debug_mode=False, torch_compile_max_bs=32, torchao_config='', speculative_algorithm=None, speculative_draft_model_path=None, speculative_draft_model_revision=None, speculative_draft_load_format=None, speculative_num_steps=None, speculative_eagle_topk=None, speculative_num_draft_tokens=None, speculative_dflash_block_size=None, speculative_dspark_block_size=None, speculative_dspark_sps_table_path=None, speculative_dspark_confidence_sts_path=None, speculative_dspark_align_verify_tokens_to_graph_tier=False, speculative_accept_threshold_single=1.0, speculative_accept_threshold_acc=1.0, speculative_use_rejection_sampling=False, speculative_token_map=None, speculative_attention_mode='prefill', speculative_draft_attention_backend=None, speculative_draft_window_size=None, speculative_moe_runner_backend='flashinfer_mxfp4', speculative_moe_a2a_backend=None, speculative_draft_model_quantization=None, speculative_skip_dp_mlp_sync=False, enable_multi_layer_eagle=False, speculative_adaptive=False, speculative_adaptive_config=None, decoupled_spec_bind_endpoint=None, decoupled_spec_connect_endpoints=None, decoupled_spec_rank=None, decoupled_spec_role='null', spec_trace_dir=None, speculative_ngram_min_bfs_breadth=1, speculative_ngram_max_bfs_breadth=10, speculative_ngram_match_type='BFS', speculative_ngram_max_trie_depth=18, speculative_ngram_capacity=10000000, speculative_ngram_external_corpus_path=None, speculative_ngram_external_sam_budget=0, speculative_ngram_external_corpus_max_tokens=10000000, ep_size=4, moe_a2a_backend='none', moe_runner_backend='flashinfer_mxfp4', flashinfer_mxfp4_moe_precision='default', deepep_mode='auto', fuseep_mode=2, deepep_dispatcher_output_dtype='auto', ep_num_redundant_experts=0, ep_dispatch_algorithm=None, init_expert_location='trivial', enable_eplb=False, eplb_algorithm='auto', eplb_rebalance_num_iterations=1000, eplb_rebalance_layers_per_chunk=None, eplb_min_rebalancing_utilization_threshold=1.0, expert_distribution_recorder_mode=None, expert_distribution_recorder_buffer_size=1000, enable_expert_distribution_metrics=False, deepep_config=None, moe_dense_tp_size=None, elastic_ep_backend=None, enable_elastic_expert_backup=False, mooncake_ib_device=None, enable_waterfill=False, ep_join_mode=None, ep_join_rank_offset=0, elastic_ep_initial_size=None, max_ep_size=None, elastic_ep_scale_timeout=600, elastic_ep_rejoin=False, disable_flashinfer_cutlass_moe_fp4_allgather=False, disable_shared_experts_fusion=False, enforce_shared_experts_fusion=False, max_mamba_cache_size=None, mamba_ssm_dtype=None, mamba_max_states_per_path=-1, enable_mamba_cache_stochastic_rounding=False, mamba_cache_philox_rounds=0, mamba_full_memory_ratio=0.36, mamba_radix_cache_strategy='extra_buffer_lazy', uses_mamba_radix_cache=False, mamba_track_interval=256, enable_int8_mamba_checkpoint=False, int8_mamba_ckpt_size=None, linear_attn_backend='triton', linear_attn_decode_backend=None, linear_attn_prefill_backend=None, linear_attn_verify_backend=None, enable_linear_replayssm=False, linear_replayssm_cache_len=16, enable_linear_replayssm_spec=False, enable_hierarchical_cache=False, hicache_ratio=2.0, hicache_size=0, hicache_write_policy='write_through', hicache_io_backend='kernel', hicache_mem_layout='page_first', hicache_storage_backend=None, hicache_storage_prefetch_policy='timeout', hicache_storage_backend_extra_config=None, enable_hisparse=False, hisparse_config=None, enable_broadcast_mm_inputs_process=False, enable_prefix_mm_cache=False, mm_enable_dp_encoder=False, mm_process_config={}, mm_processor_worker_num=0, mm_io_worker_num=0, limit_mm_data_per_request=None, enable_mm_global_cache=False, disable_fast_image_processor=False, mm_feature_transport='cpu', keep_mm_feature_on_device=False, enable_lora=None, enable_lora_overlap_loading=None, max_lora_rank=None, lora_target_modules=None, lora_paths=None, max_loaded_loras=None, max_loras_per_batch=8, lora_eviction_policy='lru', lora_backend='csgmv', max_lora_chunk_size=16, experts_shared_outer_loras=None, lora_use_virtual_experts=False, lora_strict_loading=False, lora_drain_wait_threshold=0.0, enable_two_batch_overlap=False, enable_single_batch_overlap=False, tbo_token_distribution_threshold=0.48, cpu_offload_gb=0, offload_group_size=-1, offload_num_in_group=1, offload_prefetch_step=1, offload_mode='cpu', enable_lmcache=False, lmcache_config_file=None, enable_flexkv=False, flexkv_config_file=None, kt_weight_path=None, kt_method='AMXINT4', kt_cpuinfer=None, kt_threadpool_count=2, kt_num_gpu_experts=None, kt_max_deferred_experts_per_token=None, dllm_algorithm=None, dllm_algorithm_config=None, dllm_fdfo=True, disaggregation_mode='prefill', disaggregation_transfer_backend='mooncake', disaggregation_bootstrap_port=28800, disaggregation_ib_device='mlx5_0,mlx5_1,mlx5_2,mlx5_3', disaggregation_decode_enable_radix_cache=False, disaggregation_decode_enable_offload_kvcache=False, num_reserved_decode_tokens=512, disaggregation_decode_extra_slots=None, disaggregation_decode_polling_interval=1, optimistic_prefill_attempts=0, encoder_only=False, language_only=False, encoder_transfer_backend='zmq_to_tokenizer', encoder_urls=[], encoder_bootstrap_port=8997, encoder_register_urls=[], enable_adaptive_dispatch_to_encoder=False, enable_pdmux=False, pdmux_config_path=None, sm_group_num=8, custom_weight_loader=[], weight_loader_disable_mmap=False, weight_loader_prefetch_checkpoints=False, weight_loader_prefetch_num_threads=4, weight_loader_drop_cache_after_load=False, remote_instance_weight_loader_seed_instance_ip=None, remote_instance_weight_loader_seed_instance_service_port=None, remote_instance_weight_loader_send_weights_group_ports=None, remote_instance_weight_loader_backend='nccl', remote_instance_weight_loader_start_seed_via_transfer_engine=False, engine_info_bootstrap_port=6789, modelexpress_config=None, download_dir=None, model_checksum=None, delete_ckpt_after_loading=False, decrypted_config_file=None, decrypted_draft_config_file=None, checkpoint_engine_wait_weights_before_ready=False, enable_prefill_delayer=False, prefill_delayer_max_delay_passes=30, prefill_delayer_token_usage_low_watermark=None, prefill_delayer_forward_passes_buckets=None, prefill_delayer_wait_seconds_buckets=None, prefill_delayer_queue_min_ratio=None, prefill_delayer_max_delay_ms=None, min_free_slots_delay=None, enable_deterministic_inference=False, rl_on_policy_target=None, kv_canary='none', kv_canary_real_data='none', kv_canary_sweep_interval=0, enable_dynamic_batch_tokenizer=False, dynamic_batch_tokenizer_batch_size=32, dynamic_batch_tokenizer_batch_timeout=0.002, enable_tokenizer_batch_encode=False, disable_tokenizer_batch_decode=False, debug_tensor_dump_output_folder=None, debug_tensor_dump_layers=None, debug_tensor_dump_input_file=None, enable_memory_saver=False, enable_weights_cpu_backup=False, enable_draft_weights_cpu_backup=False, enable_custom_logit_processor=False, enable_return_hidden_states=False, enable_return_routed_experts=False, enable_return_indexer_topk=False, disable_outlines_disk_cache=False, enable_mis=False, weight_cache_mode='off', weight_cache_socket=None, weight_cache_timeout=1800, forward_hooks=None, msprobe_dump_config=None)
+[2026-08-27 04:28:58] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:58] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:58] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:58] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:58] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:58] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:58] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:58] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:01] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:01] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:01] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:03] Multimodal processor concurrency enabled with 2 isolated worker threads (auto).
+[2026-08-27 04:29:03] CommonKVBootstrapServer started successfully on 0.0.0.0:28800
+[2026-08-27 04:29:03] No HuggingFace chat template found
+[2026-08-27 04:29:03] No chat template found, defaulting to 'string' content format
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:15 PP0 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:15 PP0 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:16] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:16] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:16 PP1 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:16 PP1 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+/sgl-workspace/sglang/python/sglang/srt/utils/hf_transformers/tokenizer.py:423: UserWarning: Using a slow tokenizer. This might cause a significant slowdown. Consider using a fast tokenizer instead.
+  warnings.warn(
+[2026-08-27 04:29:16] Applying special tokens cache patch for Kimi tokenizer: <class 'transformers_modules.Kimi_hyphen_K3.caf40b2773f9029e.tokenization_kimi.TikTokenTokenizer'>
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:16 PP1 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:16 PP1 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:16 PP0 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:16 PP0 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:16 PP1 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:16 PP1 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:17 PP0 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:17 PP0 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:17 PP1 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:17 PP1 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:17 PP0 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:17 PP0 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:17 PP0 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:17.587481   237 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:17.587555   237 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:17.587566   237 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.1 port: 12001
+I0827 04:29:17.587602   237 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.1:16830
+I0827 04:29:17.587723   237 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:17.613691   237 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:17.620473   237 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:17.627439   237 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:17.653980   237 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:17.656663   237 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:17.656692   237 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:17.656746   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:17.663833   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:17.665359   237 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:01
+I0827 04:29:17.665387   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:17.690173   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:17.691654   237 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:01
+I0827 04:29:17.691689   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:17.718215   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:17.719650   237 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:01
+I0827 04:29:17.719686   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:17.746297   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:17.747779   237 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:01
+I0827 04:29:17.747891   237 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:17 PP0 TP1 EP1] Init torch distributed begin.
+[2026-08-27 04:29:17 PP1 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:18.219898   241 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:18.219964   241 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:18.219973   241 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.1 port: 12001
+I0827 04:29:18.220007   241 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.1:15425
+I0827 04:29:18.220109   241 transfer_engine_impl.cpp:258] Auto-discovering topology...
+[2026-08-27 04:29:18 PP1 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+I0827 04:29:18.244532   241 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:18.252967   241 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:18.261787   241 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:18.289944   241 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:18.293610   241 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:18.293633   241 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:18.293681   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.301677   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:18.303126   241 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:01
+I0827 04:29:18.303155   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.330181   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:18.331590   241 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:01
+I0827 04:29:18.331614   241 rdma_context.cpp:205] Using SIEVE endpoint store
+[2026-08-27 04:29:18 PP0 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+I0827 04:29:18.358213   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:18.359581   241 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:01
+I0827 04:29:18.359604   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.386204   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:18.387562   241 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:01
+I0827 04:29:18.387681   241 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:18 PP1 TP1 EP1] Init torch distributed begin.
+[2026-08-27 04:29:18 PP1 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:18.588883   242 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:18.588970   242 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:18.588984   242 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.1 port: 12001
+I0827 04:29:18.589022   242 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.1:15650
+I0827 04:29:18.589143   242 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:18.594419   242 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:18.599439   242 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:18.604553   242 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:18.609825   242 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:18.612036   242 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:18.612056   242 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:18.612097   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.618021   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:18.619308   242 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:01
+I0827 04:29:18.619333   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.646080   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:18.647465   242 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:01
+I0827 04:29:18.647495   242 rdma_context.cpp:205] Using SIEVE endpoint store
+[2026-08-27 04:29:18 PP0 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+I0827 04:29:18.674199   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:18.675598   242 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:01
+I0827 04:29:18.675621   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.682866   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:18.684276   242 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:01
+I0827 04:29:18.684391   242 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:18 PP1 TP2 EP2] Init torch distributed begin.
+[2026-08-27 04:29:18 PP1 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:18.695660   238 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:18.695755   238 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:18.695767   238 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.1 port: 12001
+I0827 04:29:18.695806   238 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.1:15123
+I0827 04:29:18.695909   238 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:18.701093   238 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:18.706161   238 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:18.711323   238 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:18.737910   238 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:18.740556   238 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:18.740602   238 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:18.740651   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.747717   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:18.749193   238 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:01
+I0827 04:29:18.749220   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.774145   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:18.775534   238 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:01
+I0827 04:29:18.775558   238 rdma_context.cpp:205] Using SIEVE endpoint store
+[2026-08-27 04:29:18 PP0 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+I0827 04:29:18.802196   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:18.803623   238 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:01
+I0827 04:29:18.803649   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.810976   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:18.812341   238 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:01
+I0827 04:29:18.812443   238 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:18 PP0 TP2 EP2] Init torch distributed begin.
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:19.084213   243 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:19.084332   243 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:19.084348   243 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.1 port: 12001
+I0827 04:29:19.084400   243 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.1:15411
+I0827 04:29:19.084602   243 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:19.114387   243 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:19.121773   243 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:19.146405   243 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:19.153759   243 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:19.156515   243 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:19.156548   243 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:19.156603   243 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.163982   243 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:19.165694   243 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:01
+I0827 04:29:19.165726   243 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.173307   243 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:19.174904   243 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:01
+I0827 04:29:19.174932   243 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.206454   243 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:19.208142   243 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:01
+I0827 04:29:19.208168   243 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.215909   243 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:19.217583   243 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:01
+I0827 04:29:19.217706   243 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:19 PP1 TP3 EP3] Init torch distributed begin.
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:19.239588   236 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:19.239699   236 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:19.239715   236 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.1 port: 12001
+I0827 04:29:19.239763   236 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.1:15605
+I0827 04:29:19.239933   236 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:19.240546   240 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:19.240644   240 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:19.240658   240 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.1 port: 12001
+I0827 04:29:19.240702   240 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.1:15487
+I0827 04:29:19.240862   240 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:19.242301   239 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:19.242408   239 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:19.242422   239 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.1 port: 12001
+I0827 04:29:19.242470   239 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.1:15552
+I0827 04:29:19.242636   239 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:19.246201   240 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:19.248451   239 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:19.251204   240 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:19.254226   239 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:19.256254   240 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:19.259852   239 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:19.261372   240 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:19.263411   240 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:19.263432   240 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:19.263473   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.265838   239 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:19.268059   239 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:19.268082   239 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:19.268126   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.269106   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:19.270298   236 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:19.270401   240 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:01
+I0827 04:29:19.270423   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.274480   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:19.275991   239 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:01
+I0827 04:29:19.276019   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.276877   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:19.278193   240 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:01
+I0827 04:29:19.278220   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.297820   236 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:19.302485   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:19.302623   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:19.303958   239 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:01
+I0827 04:29:19.303983   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.304121   240 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:01
+I0827 04:29:19.304147   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.311621   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:19.313341   240 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:01
+I0827 04:29:19.313478   240 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:19 PP1 TP0 EP0] Init torch distributed begin.
+I0827 04:29:19.322126   236 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:19.330281   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:19.331759   239 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:01
+I0827 04:29:19.331784   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.339274   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:19.340766   239 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:01
+I0827 04:29:19.340884   239 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:19 PP0 TP3 EP3] Init torch distributed begin.
+I0827 04:29:19.345958   236 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:19.348632   236 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:19.348659   236 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:19.348712   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.355962   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:19.417273   236 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:01
+I0827 04:29:19.417309   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.424894   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:19.426437   236 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:01
+I0827 04:29:19.426491   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.434231   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:19.435868   236 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:01
+I0827 04:29:19.435896   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.443578   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:19.445148   236 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:01
+I0827 04:29:19.445264   236 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:19 PP0 TP0 EP0] Init torch distributed begin.
+[2026-08-27 04:29:26 PP0 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:26 PP1 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:27 PP1 TP2 EP2] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP1 TP0 EP0] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP1 TP3 EP3] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP1 TP1 EP1] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP0 TP3 EP3] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP0 TP0 EP0] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP0 TP1 EP1] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP0 TP2 EP2] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP0 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:27 PP0 TP3 EP3] sglang is using nccl==2.28.9
+[2026-08-27 04:29:27 PP0 TP1 EP1] sglang is using nccl==2.28.9
+[2026-08-27 04:29:27 PP0 TP2 EP2] sglang is using nccl==2.28.9
+[2026-08-27 04:29:27 PP0 TP0 EP0] Init torch distributed ends. elapsed=8.22 s, mem usage=0.46 GB
+[2026-08-27 04:29:27 PP1 TP3 EP3] Init torch distributed ends. elapsed=8.45 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP1 TP1 EP1] Init torch distributed ends. elapsed=9.28 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP1 TP2 EP2] Init torch distributed ends. elapsed=8.98 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP1 TP0 EP0] Init torch distributed ends. elapsed=8.35 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP0 TP2 EP2] Init torch distributed ends. elapsed=8.85 s, mem usage=0.46 GB
+[2026-08-27 04:29:27 PP0 TP3 EP3] Init torch distributed ends. elapsed=8.32 s, mem usage=0.46 GB
+[2026-08-27 04:29:27 PP0 TP1 EP1] Init torch distributed ends. elapsed=9.92 s, mem usage=0.46 GB
+[2026-08-27 04:29:27 PP1 TP0 EP0] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP1 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP0 TP0 EP0] Load weight begin. avail mem=82.12 GB
+[2026-08-27 04:29:27 PP1 TP3 EP3] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP1 TP2 EP2] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP0 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP1 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP1 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP1 TP1 EP1] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP0 TP2 EP2] Load weight begin. avail mem=82.12 GB
+[2026-08-27 04:29:27 PP1 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP0 TP3 EP3] Load weight begin. avail mem=82.12 GB
+[2026-08-27 04:29:27 PP0 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP0 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP1 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:27 PP1 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:27 PP0 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:27 PP1 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:27 PP1 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:27 PP1 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:27 PP0 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:27 PP1 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:27 PP1 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:27 PP1 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:27 PP0 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:27 PP0 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:27 PP0 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:27 PP0 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:27 PP0 TP1 EP1] Load weight begin. avail mem=82.12 GB
+[2026-08-27 04:29:27 PP0 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP0 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP0 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+
Multi-thread loading shards:   0% Completed | 0/96 [00:00<?, ?it/s]
Multi-thread loading shards:   1% Completed | 1/96 [00:00<01:01,  1.55it/s]
Multi-thread loading shards:   2% Completed | 2/96 [00:09<08:49,  5.63s/it]
Multi-thread loading shards:   3% Completed | 3/96 [00:12<06:35,  4.25s/it]
Multi-thread loading shards:   4% Completed | 4/96 [00:14<05:26,  3.55s/it]
Multi-thread loading shards:   5% Completed | 5/96 [00:17<04:43,  3.11s/it]
Multi-thread loading shards:   6% Completed | 6/96 [00:19<04:17,  2.86s/it]
Multi-thread loading shards:   7% Completed | 7/96 [00:21<04:00,  2.71s/it]
Multi-thread loading shards:   8% Completed | 8/96 [00:24<03:49,  2.61s/it]
Multi-thread loading shards:   9% Completed | 9/96 [00:26<03:44,  2.58s/it]
Multi-thread loading shards:  10% Completed | 10/96 [00:29<03:40,  2.56s/it]
Multi-thread loading shards:  11% Completed | 11/96 [00:31<03:36,  2.54s/it]
Multi-thread loading shards:  12% Completed | 12/96 [00:32<02:43,  1.95s/it]
Multi-thread loading shards:  14% Completed | 13/96 [00:32<02:05,  1.52s/it]
Multi-thread loading shards:  15% Completed | 14/96 [00:33<01:33,  1.14s/it]
Multi-thread loading shards:  16% Completed | 15/96 [00:33<01:21,  1.01s/it]
Multi-thread loading shards:  17% Completed | 16/96 [00:34<01:19,  1.00it/s]
Multi-thread loading shards:  18% Completed | 17/96 [00:35<01:02,  1.26it/s]
Multi-thread loading shards:  19% Completed | 18/96 [00:35<00:51,  1.53it/s]
Multi-thread loading shards:  20% Completed | 19/96 [00:35<00:42,  1.80it/s]
Multi-thread loading shards:  21% Completed | 20/96 [00:36<00:37,  2.01it/s]
Multi-thread loading shards:  22% Completed | 21/96 [00:36<00:33,  2.23it/s]
Multi-thread loading shards:  23% Completed | 22/96 [00:36<00:30,  2.43it/s]
Multi-thread loading shards:  24% Completed | 23/96 [00:37<00:27,  2.68it/s]
Multi-thread loading shards:  25% Completed | 24/96 [00:37<00:25,  2.80it/s]
Multi-thread loading shards:  26% Completed | 25/96 [00:37<00:24,  2.89it/s]
Multi-thread loading shards:  27% Completed | 26/96 [00:38<00:23,  2.99it/s]
Multi-thread loading shards:  28% Completed | 27/96 [00:38<00:22,  3.01it/s]
Multi-thread loading shards:  29% Completed | 28/96 [00:38<00:22,  2.99it/s]
Multi-thread loading shards:  30% Completed | 29/96 [00:39<00:22,  2.98it/s]
Multi-thread loading shards:  31% Completed | 30/96 [00:39<00:22,  2.98it/s]
Multi-thread loading shards:  32% Completed | 31/96 [00:39<00:21,  3.01it/s]
Multi-thread loading shards:  33% Completed | 32/96 [00:40<00:21,  3.00it/s]
Multi-thread loading shards:  34% Completed | 33/96 [00:41<00:35,  1.79it/s]
Multi-thread loading shards:  35% Completed | 34/96 [00:41<00:30,  2.02it/s]
Multi-thread loading shards:  36% Completed | 35/96 [00:41<00:27,  2.23it/s]
Multi-thread loading shards:  38% Completed | 36/96 [00:42<00:24,  2.41it/s]
Multi-thread loading shards:  39% Completed | 37/96 [00:42<00:22,  2.57it/s]
Multi-thread loading shards:  40% Completed | 38/96 [00:42<00:21,  2.64it/s]
Multi-thread loading shards:  41% Completed | 39/96 [00:43<00:20,  2.73it/s]
Multi-thread loading shards:  42% Completed | 40/96 [00:43<00:20,  2.80it/s]
Multi-thread loading shards:  43% Completed | 41/96 [00:43<00:19,  2.79it/s]
Multi-thread loading shards:  44% Completed | 42/96 [00:44<00:19,  2.80it/s]
Multi-thread loading shards:  45% Completed | 43/96 [00:44<00:22,  2.39it/s]
Multi-thread loading shards:  46% Completed | 44/96 [00:45<00:20,  2.51it/s]
Multi-thread loading shards:  47% Completed | 45/96 [00:45<00:19,  2.64it/s]
Multi-thread loading shards:  48% Completed | 46/96 [00:45<00:18,  2.67it/s]
Multi-thread loading shards:  49% Completed | 47/96 [00:46<00:18,  2.61it/s]
Multi-thread loading shards:  50% Completed | 48/96 [00:46<00:17,  2.72it/s]
Multi-thread loading shards:  51% Completed | 49/96 [00:47<00:16,  2.84it/s]
Multi-thread loading shards:  52% Completed | 50/96 [00:47<00:16,  2.81it/s]
Multi-thread loading shards:  53% Completed | 51/96 [00:47<00:15,  2.93it/s]
Multi-thread loading shards:  54% Completed | 52/96 [00:48<00:21,  2.05it/s]
Multi-thread loading shards:  55% Completed | 53/96 [00:49<00:33,  1.30it/s]
Multi-thread loading shards:  56% Completed | 54/96 [00:50<00:27,  1.55it/s]
Multi-thread loading shards:  57% Completed | 55/96 [00:50<00:22,  1.82it/s]
Multi-thread loading shards:  58% Completed | 56/96 [00:50<00:19,  2.06it/s]
Multi-thread loading shards:  59% Completed | 57/96 [00:51<00:17,  2.25it/s]
Multi-thread loading shards:  60% Completed | 58/96 [00:51<00:15,  2.40it/s]
Multi-thread loading shards:  61% Completed | 59/96 [00:51<00:14,  2.56it/s]
Multi-thread loading shards:  62% Completed | 60/96 [00:52<00:13,  2.67it/s]
Multi-thread loading shards:  64% Completed | 61/96 [00:52<00:12,  2.77it/s]
Multi-thread loading shards:  65% Completed | 62/96 [00:52<00:11,  2.84it/s]
Multi-thread loading shards:  66% Completed | 63/96 [00:53<00:11,  2.89it/s]
Multi-thread loading shards:  67% Completed | 64/96 [00:53<00:11,  2.87it/s]
Multi-thread loading shards:  68% Completed | 65/96 [00:54<00:10,  2.89it/s]
Multi-thread loading shards:  69% Completed | 66/96 [00:54<00:10,  2.91it/s]
Multi-thread loading shards:  70% Completed | 67/96 [00:54<00:10,  2.90it/s]
Multi-thread loading shards:  71% Completed | 68/96 [00:55<00:09,  2.90it/s]
Multi-thread loading shards:  72% Completed | 69/96 [00:55<00:09,  2.89it/s]
Multi-thread loading shards:  73% Completed | 70/96 [00:56<00:14,  1.85it/s]
Multi-thread loading shards:  74% Completed | 71/96 [00:56<00:11,  2.10it/s]
Multi-thread loading shards:  75% Completed | 72/96 [00:57<00:10,  2.23it/s]
Multi-thread loading shards:  76% Completed | 73/96 [00:57<00:09,  2.39it/s]
Multi-thread loading shards:  77% Completed | 74/96 [00:57<00:08,  2.48it/s]
Multi-thread loading shards:  78% Completed | 75/96 [00:58<00:09,  2.15it/s]
Multi-thread loading shards:  79% Completed | 76/96 [00:59<00:10,  1.89it/s]
Multi-thread loading shards:  80% Completed | 77/96 [00:59<00:09,  2.05it/s]
Multi-thread loading shards:  81% Completed | 78/96 [00:59<00:08,  2.24it/s]
Multi-thread loading shards:  82% Completed | 79/96 [01:00<00:07,  2.39it/s]
Multi-thread loading shards:  83% Completed | 80/96 [01:00<00:06,  2.51it/s]
Multi-thread loading shards:  84% Completed | 81/96 [01:00<00:05,  2.65it/s]
Multi-thread loading shards:  85% Completed | 82/96 [01:01<00:04,  2.84it/s]
Multi-thread loading shards:  86% Completed | 83/96 [01:02<00:06,  1.99it/s]
Multi-thread loading shards:  88% Completed | 84/96 [01:02<00:06,  1.99it/s]
Multi-thread loading shards:  90% Completed | 86/96 [01:02<00:03,  3.27it/s]
Multi-thread loading shards:  92% Completed | 88/96 [01:02<00:01,  4.72it/s]
Multi-thread loading shards:  94% Completed | 90/96 [01:02<00:00,  6.49it/s]
Multi-thread loading shards:  96% Completed | 92/96 [01:03<00:00,  8.37it/s]
Multi-thread loading shards:  98% Completed | 94/96 [01:03<00:00,  8.91it/s]
Multi-thread loading shards: 100% Completed | 96/96 [01:03<00:00,  9.35it/s]
Multi-thread loading shards: 100% Completed | 96/96 [01:03<00:00,  1.51it/s]
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:33 PP0 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:33 PP0 TP2 EP2] Load weight end. elapsed=65.63 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=39.03 GB, mem usage=43.09 GB.
+[2026-08-27 04:30:33 PP0 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:33 PP0 TP0 EP0] Load weight end. elapsed=65.81 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=39.03 GB, mem usage=43.09 GB.
+[2026-08-27 04:30:35 PP0 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:35 PP0 TP3 EP3] Load weight end. elapsed=67.37 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=39.03 GB, mem usage=43.09 GB.
+[2026-08-27 04:30:35 PP0 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:35 PP0 TP1 EP1] Load weight end. elapsed=67.46 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=39.03 GB, mem usage=43.09 GB.
+[2026-08-27 04:30:35 PP0 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:35 PP0 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:35 PP0 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:35 PP0 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:35 PP0 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:35 PP0 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:35 PP0 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:35 PP0 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:36 PP1 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:36 PP1 TP3 EP3] Load weight end. elapsed=68.32 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=36.12 GB, mem usage=45.92 GB.
+[2026-08-27 04:30:36 PP1 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP1 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP1 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP1 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP1 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP1 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP1 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP1 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:37 PP1 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:37 PP1 TP1 EP1] Load weight end. elapsed=69.55 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=36.12 GB, mem usage=45.92 GB.
+[2026-08-27 04:30:37 PP1 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:37 PP1 TP0 EP0] Load weight end. elapsed=69.66 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=36.12 GB, mem usage=45.92 GB.
+[2026-08-27 04:30:39 PP1 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:39 PP1 TP2 EP2] Load weight end. elapsed=71.33 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=36.12 GB, mem usage=45.92 GB.
+[2026-08-27 04:30:39 PP1 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:39 PP1 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:39 PP1 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:39 PP1 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:39 PP1 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:39 PP1 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:39 PP1 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:39 PP1 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:53 PP0 TP0 EP0] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:53 PP1 TP3 EP3] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:53 PP1 TP2 EP2] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:53 PP1 TP1 EP1] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:53 PP1 TP0 EP0] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:53 PP0 TP2 EP2] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:53 PP0 TP3 EP3] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:53 PP0 TP1 EP1] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:53 PP0 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:53 PP1 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:53 PP1 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:53 PP1 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:53 PP0 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:53 PP1 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:53 PP0 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:53 PP0 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:53 PP0 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 8.16 GB
+[2026-08-27 04:30:53 PP0 TP0 EP0] Memory pool end. avail mem=30.14 GB
+[2026-08-27 04:30:53 PP0 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 8.16 GB
+[2026-08-27 04:30:53 PP0 TP2 EP2] Memory pool end. avail mem=30.14 GB
+[2026-08-27 04:30:53 PP1 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:53 PP1 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:53 PP1 TP0 EP0] Memory pool end. avail mem=23.24 GB
+[2026-08-27 04:30:53 PP1 TP2 EP2] Memory pool end. avail mem=23.24 GB
+[2026-08-27 04:30:53 PP0 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 8.16 GB
+[2026-08-27 04:30:53 PP1 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:53 PP0 TP3 EP3] Memory pool end. avail mem=30.14 GB
+[2026-08-27 04:30:53 PP1 TP1 EP1] Memory pool end. avail mem=23.24 GB
+[2026-08-27 04:30:53 PP1 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:53 PP0 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 8.16 GB
+[2026-08-27 04:30:53 PP1 TP3 EP3] Memory pool end. avail mem=23.24 GB
+[2026-08-27 04:30:53 PP0 TP1 EP1] Memory pool end. avail mem=30.14 GB
+[2026-08-27 04:30:54 PP0 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:30:54 PP0 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:30:54 PP0 TP0 EP0] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp0_pp0_dp0.json
+[2026-08-27 04:30:54 PP1 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:30:54 PP1 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:30:54 PP0 TP2 EP2] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp2_pp0_dp0.json
+[2026-08-27 04:30:54 PP1 TP2 EP2] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp2_pp1_dp0.json
+[2026-08-27 04:30:54 PP1 TP0 EP0] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp0_pp1_dp0.json
+[2026-08-27 04:30:54 PP1 TP1 EP1] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp1_pp1_dp0.json
+[2026-08-27 04:30:54 PP1 TP3 EP3] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp3_pp1_dp0.json
+[2026-08-27 04:30:54 PP0 TP1 EP1] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp1_pp0_dp0.json
+[2026-08-27 04:30:54 PP0 TP3 EP3] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp3_pp0_dp0.json
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.46profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.32profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.51profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  6.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.09profile/s]
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.07profile/s]
+[TensorRT-LLM][INFO] Set logger level to INFO
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.48profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.11profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.29profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.37profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.16profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.82profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.18profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.58profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:01<00:02,  1.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.66profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.03profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.49profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:01<00:01,  2.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.25profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:01,  2.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:02<00:00,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:02<00:00,  3.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.20profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.01profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  9.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  9.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  9.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 12.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 12.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 11.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 11.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.78profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.16profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  3.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.75profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.50profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.92profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.68profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  3.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.97profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.87profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 10.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.51profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.74profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 11.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.30profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.70profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.72profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.63profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.53profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.29profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.58profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  8.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 10.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.71profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.62profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.43profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.18profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.29profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.37profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.53profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.37profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.35profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.00profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.71profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.60profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.78profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.79profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  8.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.86profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  8.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.16profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.78profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  3.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.15profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.94profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.93profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.25profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.23profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.97profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.79profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.96profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.64profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.12profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.75profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.12profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.25profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.91profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.05profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.66profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.90profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  9.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.73profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.55profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.49profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.13profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.74profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.17profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  3.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.72profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.57profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.59profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.64profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.83profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 13.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.49profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 12.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  9.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  7.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.32profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.93profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.29profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.72profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.75profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.46profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.51profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  3.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.90profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.16profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.72profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.63profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.83profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.70profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.59profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.55profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  6.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.93profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.33profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.75profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.30profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  6.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  7.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.79profile/s]
+[2026-08-27 04:31:21 PP0 TP1 EP1] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.53profile/s][2026-08-27 04:31:21 PP0 TP0 EP0] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.97profile/s]
+[2026-08-27 04:31:21 PP0 TP3 EP3] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.43profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.06profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s][2026-08-27 04:31:21 PP0 TP2 EP2] FlashInfer autotune completed.
+[2026-08-27 04:31:21 PP0 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:21 PP0 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:21 PP0 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:21 PP0 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:21 PP0 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:21 PP0 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:21 PP0 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:21 PP0 TP0 EP0] max_total_num_tokens=3801408, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=54, context_len=1048576, available_gpu_mem=27.35 GB
+[2026-08-27 04:31:21 PP0 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.86profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.58profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:01<00:07,  1.18s/profile]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  2.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.27profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.55profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  3.96profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.86profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  6.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  7.01profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.23profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.05profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  6.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  7.06profile/s]
+[2026-08-27 04:31:26 PP1 TP1 EP1] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.40profile/s]
+[2026-08-27 04:31:26 PP1 TP2 EP2] FlashInfer autotune completed.
+[2026-08-27 04:31:26 PP1 TP0 EP0] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.81profile/s]
+[2026-08-27 04:31:26 PP1 TP3 EP3] FlashInfer autotune completed.
+[2026-08-27 04:31:26 PP1 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:26 PP1 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:26 PP1 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:26 PP1 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:26 PP1 TP0 EP0] max_total_num_tokens=3801408, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=54, context_len=1048576, available_gpu_mem=20.45 GB
+[2026-08-27 04:31:26 PP1 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:26 PP1 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:26 PP1 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:26 PP1 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+W0827 04:31:30.805795  2607 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fe48940, len: 6912: Operation not permitted [1]
+W0827 04:31:30.805797  2608 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4f684400, len: 6912: Operation not permitted [1]
+W0827 04:31:30.805945  2609 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4f962700, len: 6912: Operation not permitted [1]
+W0827 04:31:30.805979  2610 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4f9e8a00, len: 6912: Operation not permitted [1]
+W0827 04:31:30.806130  2612 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fe5f640, len: 55296: Operation not permitted [1]
+W0827 04:31:30.806133  2611 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4f83f000, len: 55296: Operation not permitted [1]
+W0827 04:31:30.806264  2613 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fe328c0, len: 6912: Operation not permitted [1]
+W0827 04:31:30.806334  2614 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fa75cc0, len: 13824: Operation not permitted [1]
+W0827 04:31:30.806449  2615 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fa87f80, len: 6912: Operation not permitted [1]
+W0827 04:31:30.806493  2616 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fa29a80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.203712  2635 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50c65e80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.203859  2636 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50c6bb40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.203975  2637 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50d1ffc0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.204025  2638 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50a0c300, len: 6912: Operation not permitted [1]
+W0827 04:31:31.204075  2640 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50de1940, len: 55296: Operation not permitted [1]
+W0827 04:31:31.204082  2639 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50dc93c0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.204438  2641 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50c87940, len: 6912: Operation not permitted [1]
+W0827 04:31:31.204561  2642 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50cffb00, len: 13824: Operation not permitted [1]
+W0827 04:31:31.204641  2644 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50dc7380, len: 6912: Operation not permitted [1]
+W0827 04:31:31.204722  2643 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50c243c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.323861  2661 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50b13c40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.323863  2662 memory_location.cpp:73] Failed to get NUMA node, addr: 0x511191c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.323912  2663 memory_location.cpp:73] Failed to get NUMA node, addr: 0x511b37c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.324069  2664 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50b2da80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.324110  2666 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5115c980, len: 55296: Operation not permitted [1]
+W0827 04:31:31.324155  2665 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5116ac40, len: 55296: Operation not permitted [1]
+W0827 04:31:31.324479  2667 memory_location.cpp:73] Failed to get NUMA node, addr: 0x511a7e80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.324587  2668 memory_location.cpp:73] Failed to get NUMA node, addr: 0x511b8cc0, len: 13824: Operation not permitted [1]
+W0827 04:31:31.324621  2669 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50fc6800, len: 6912: Operation not permitted [1]
+W0827 04:31:31.324682  2670 memory_location.cpp:73] Failed to get NUMA node, addr: 0x511bc780, len: 6912: Operation not permitted [1]
+W0827 04:31:31.543082  2689 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4902e440, len: 6912: Operation not permitted [1]
+W0827 04:31:31.543203  2690 memory_location.cpp:73] Failed to get NUMA node, addr: 0x49484200, len: 6912: Operation not permitted [1]
+W0827 04:31:31.543314  2691 memory_location.cpp:73] Failed to get NUMA node, addr: 0x49119600, len: 6912: Operation not permitted [1]
+W0827 04:31:31.543375  2692 memory_location.cpp:73] Failed to get NUMA node, addr: 0x493dc780, len: 6912: Operation not permitted [1]
+W0827 04:31:31.543467  2693 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51ac5fc0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.543601  2694 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51a9e780, len: 55296: Operation not permitted [1]
+W0827 04:31:31.543751  2695 memory_location.cpp:73] Failed to get NUMA node, addr: 0x49149a40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.543912  2696 memory_location.cpp:73] Failed to get NUMA node, addr: 0x518c0840, len: 13824: Operation not permitted [1]
+W0827 04:31:31.543947  2697 memory_location.cpp:73] Failed to get NUMA node, addr: 0x49303540, len: 6912: Operation not permitted [1]
+W0827 04:31:31.543990  2698 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51951a80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681525  2717 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50376e40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681525  2718 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5058ee40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681525  2716 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5008b980, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681548  2715 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fe0e840, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681727  2719 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5040eb80, len: 55296: Operation not permitted [1]
+W0827 04:31:31.681779  2724 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50209ac0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681811  2720 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5041c3c0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.681828  2722 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5048cc40, len: 13824: Operation not permitted [1]
+W0827 04:31:31.681844  2723 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fde7a00, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681851  2721 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50386580, len: 6912: Operation not permitted [1]
+W0827 04:31:31.730942  2745 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fce4240, len: 6912: Operation not permitted [1]
+W0827 04:31:31.730918  2744 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fed7d40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.730950  2743 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fdcaf80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.731007  2748 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5074c400, len: 55296: Operation not permitted [1]
+W0827 04:31:31.731073  2747 memory_location.cpp:73] Failed to get NUMA node, addr: 0x507bd9c0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.731086  2746 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4f931480, len: 6912: Operation not permitted [1]
+W0827 04:31:31.731294  2749 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd958c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.731361  2751 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd31d80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.731361  2750 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4ffabe00, len: 13824: Operation not permitted [1]
+W0827 04:31:31.731566  2752 memory_location.cpp:73] Failed to get NUMA node, addr: 0x508527c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.808167  2778 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd3ce40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.808190  2777 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd20740, len: 6912: Operation not permitted [1]
+W0827 04:31:31.808192  2776 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4ff86c80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.808637  2782 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fe88780, len: 6912: Operation not permitted [1]
+W0827 04:31:31.808724  2780 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fe48c80, len: 55296: Operation not permitted [1]
+W0827 04:31:31.808760  2781 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fe564c0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.809056  2785 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4ffa7e40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.809049  2779 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4feaa740, len: 6912: Operation not permitted [1]
+W0827 04:31:31.809075  2784 memory_location.cpp:73] Failed to get NUMA node, addr: 0x502fabc0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.809239  2783 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4ffa1f80, len: 13824: Operation not permitted [1]
+W0827 04:31:31.851475  2807 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4ff6fcc0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.851596  2808 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fc0a280, len: 6912: Operation not permitted [1]
+W0827 04:31:31.851615  2809 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd7d700, len: 6912: Operation not permitted [1]
+W0827 04:31:31.851732  2810 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd761c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.851951  2811 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fde6cc0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.851970  2812 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4feeb000, len: 55296: Operation not permitted [1]
+W0827 04:31:31.851962  2813 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fbe2680, len: 6912: Operation not permitted [1]
+W0827 04:31:31.851987  2816 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fbf07c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.852048  2815 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4ffa9940, len: 6912: Operation not permitted [1]
+W0827 04:31:31.852046  2814 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd235c0, len: 13824: Operation not permitted [1]
+[2026-08-27 04:31:32] INFO:     Started server process [1]
+[2026-08-27 04:31:32] INFO:     Waiting for application startup.
+[2026-08-27 04:31:32] Calling super().encode with {'add_special_tokens': False}
+[2026-08-27 04:31:32] Calling super().encode with {'add_special_tokens': False}
+[2026-08-27 04:31:32] INFO:     Application startup complete.
+[2026-08-27 04:31:32] INFO:     Uvicorn running on http://0.0.0.0:30000 (Press CTRL+C to quit)
+[2026-08-27 04:31:33] INFO:     127.0.0.1:60348 - "GET /model_info HTTP/1.1" 200 OK
+[2026-08-27 04:31:33] Start of pd disaggregation warmup ...
+[2026-08-27 04:31:37] INFO:     174.1.60.1:33088 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:38] INFO:     174.1.60.1:33092 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:39] INFO:     174.1.60.1:33106 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:40] INFO:     174.1.60.1:33110 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:41] INFO:     174.1.60.1:42028 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:42] INFO:     174.1.60.1:42030 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:43] INFO:     174.1.60.1:42046 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:44] INFO:     174.1.60.1:42050 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:45] INFO:     174.1.60.1:42060 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:46] INFO:     174.1.60.1:42072 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:47] INFO:     174.1.60.1:42086 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:48] INFO:     174.1.60.1:42092 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:49] INFO:     174.1.60.1:42098 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:50] INFO:     174.1.60.1:42100 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:51] INFO:     174.1.60.1:46852 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:52] INFO:     174.1.60.1:46856 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:53] INFO:     174.1.60.1:46872 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:54] INFO:     174.1.60.1:46876 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:55] INFO:     174.1.60.1:46886 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:56] INFO:     174.1.60.1:46890 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:57] INFO:     174.1.60.1:46898 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:58] INFO:     174.1.60.1:46912 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:59] INFO:     174.1.60.1:46928 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:00] INFO:     174.1.60.1:46936 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:01] INFO:     174.1.60.1:57672 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:02] INFO:     174.1.60.1:57676 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:03] INFO:     174.1.60.1:57686 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:04] INFO:     174.1.60.1:57688 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:05] INFO:     174.1.60.1:57696 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:06] INFO:     174.1.60.1:57708 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:12] INFO:     174.1.60.1:39714 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:13] INFO:     174.1.60.1:39728 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:14] INFO:     174.1.60.1:39736 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:15] INFO:     174.1.60.1:39746 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:16] INFO:     174.1.60.1:39760 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:17] INFO:     174.1.60.1:39772 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:18] INFO:     174.1.60.1:39782 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:19] INFO:     174.1.60.1:39786 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:20] INFO:     174.1.60.1:39800 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:21] INFO:     174.1.60.1:41036 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:22] INFO:     174.1.60.1:41052 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:23] INFO:     174.1.60.1:41058 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:24] INFO:     174.1.60.1:41068 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:25] INFO:     174.1.60.1:41074 - "GET /health HTTP/1.1" 503 Service Unavailable
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank1]:[W827 04:32:26.019157362 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank0]:[W827 04:32:26.019963648 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank3]:[W827 04:32:26.020028119 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank2]:[W827 04:32:26.020057305 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank6]:[W827 04:32:26.045976535 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank7]:[W827 04:32:26.046176399 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank5]:[W827 04:32:26.048033943 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank4]:[W827 04:32:26.048331224 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[2026-08-27 04:32:26] INFO:     174.1.60.1:41078 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:27] INFO:     174.1.60.1:41090 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:28] INFO:     174.1.60.1:41102 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:29] INFO:     174.1.60.1:41104 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:30] INFO:     174.1.60.1:41120 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:31] INFO:     174.1.60.1:57002 - "GET /health HTTP/1.1" 503 Service Unavailable
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank6]:[W827 04:32:31.709756552 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank7]:[W827 04:32:31.709839299 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank5]:[W827 04:32:31.709991082 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank4]:[W827 04:32:31.710123265 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[2026-08-27 04:32:32] INFO:     174.1.60.1:57016 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:33] INFO:     174.1.60.1:57030 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:34] INFO:     174.1.60.1:57046 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:35] INFO:     174.1.60.1:57056 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:36] INFO:     174.1.60.1:57066 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:37] INFO:     174.1.60.1:57070 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:38] INFO:     174.1.60.1:57072 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:39] INFO:     174.1.60.1:57082 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:40] INFO:     174.1.60.1:57098 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:41] INFO:     174.1.60.1:56606 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:50] INFO:     174.1.60.1:56620 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:51] INFO:     174.1.60.1:45658 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:52] INFO:     174.1.60.1:45672 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:53] INFO:     174.1.60.1:45674 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:54] INFO:     174.1.60.1:45690 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:55] INFO:     174.1.60.1:45704 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:56] INFO:     174.1.60.1:45708 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:57] INFO:     174.1.60.1:45714 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:58] INFO:     174.1.60.1:45722 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:00] INFO:     174.1.60.1:45736 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:01] INFO:     174.1.60.1:45746 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:02] INFO:     174.1.60.1:33310 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:03] INFO:     174.1.60.1:33320 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:04] INFO:     174.1.60.1:33324 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:05] INFO:     174.1.60.1:33328 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:06] INFO:     174.1.60.1:33336 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:07] INFO:     174.1.60.1:33350 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:08] INFO:     174.1.60.1:33354 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:09] INFO:     174.1.60.1:33356 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:10] INFO:     174.1.60.1:33368 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:11] INFO:     174.1.60.1:39318 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:12] INFO:     174.1.60.1:39328 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:13] INFO:     174.1.60.1:39336 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:14] INFO:     174.1.60.1:39344 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:15] INFO:     174.1.60.1:39346 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:16] INFO:     174.1.60.1:39352 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:17] INFO:     174.1.60.1:39364 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:18] INFO:     174.1.60.1:39378 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:19] INFO:     174.1.60.1:39390 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:20] INFO:     174.1.60.1:39402 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:25] INFO:     174.1.60.1:37792 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:26] INFO:     174.1.60.1:37804 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:27] INFO:     174.1.60.1:37812 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:28] INFO:     174.1.60.1:37826 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:29] INFO:     174.1.60.1:37830 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:30] INFO:     174.1.60.1:37844 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:31] INFO:     174.1.60.1:44208 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:32] INFO:     174.1.60.1:44220 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:33] INFO:     174.1.60.1:44226 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:34] INFO:     174.1.60.1:44228 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:35] INFO:     174.1.60.1:44238 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:37] INFO:     174.1.60.1:44248 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:38] INFO:     174.1.60.1:44258 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:39] INFO:     174.1.60.1:44272 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:40] INFO:     174.1.60.1:44276 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:41] INFO:     174.1.60.1:44194 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:42] INFO:     174.1.60.1:44198 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:43] INFO:     174.1.60.1:44212 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:44] INFO:     174.1.60.1:44216 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:45] INFO:     174.1.60.1:44222 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:46] INFO:     174.1.60.1:44236 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:47] INFO:     174.1.60.1:44246 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:48] INFO:     174.1.60.1:44252 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:49] INFO:     174.1.60.1:44264 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:50] INFO:     174.1.60.1:44278 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:51] INFO:     174.1.60.1:35728 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:52] INFO:     174.1.60.1:35736 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:53] INFO:     174.1.60.1:35752 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:54] INFO:     174.1.60.1:35754 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:55] INFO:     174.1.60.1:35758 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:04] INFO:     174.1.60.1:44514 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:05] INFO:     174.1.60.1:44518 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:06] INFO:     174.1.60.1:44520 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:07] INFO:     174.1.60.1:44526 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:08] INFO:     174.1.60.1:44530 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:09] INFO:     174.1.60.1:44546 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:10] INFO:     174.1.60.1:44550 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:11] INFO:     174.1.60.1:45098 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:12] INFO:     127.0.0.1:59722 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:12] INFO:     174.1.60.1:45106 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:13] INFO:     174.1.60.1:45122 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:14] INFO:     174.1.60.1:45124 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:15] INFO:     174.1.60.1:45136 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:16] INFO:     174.1.60.1:45146 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:17] INFO:     174.1.60.1:45152 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:19] INFO:     174.1.60.1:45158 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:20] INFO:     174.1.60.1:45170 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:21] INFO:     174.1.60.1:45182 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:22] INFO:     174.1.60.1:39946 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:23] INFO:     174.1.60.1:39958 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:24] INFO:     174.1.60.1:39962 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:25] INFO:     174.1.60.1:39978 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:26] INFO:     174.1.60.1:39990 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:27] INFO:     174.1.60.1:39994 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:28] INFO:     174.1.60.1:39996 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:29] INFO:     174.1.60.1:40002 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:30] INFO:     174.1.60.1:40012 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:31] INFO:     174.1.60.1:46978 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:32] INFO:     174.1.60.1:46988 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:33] INFO:     174.1.60.1:46996 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:34] INFO:     174.1.60.1:47004 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:39] INFO:     174.1.60.1:47020 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:40] INFO:     174.1.60.1:47022 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:41] INFO:     174.1.60.1:59158 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:42] INFO:     174.1.60.1:59170 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:43] INFO:     174.1.60.1:59172 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:44] INFO:     174.1.60.1:59180 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:45] INFO:     174.1.60.1:59184 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:46] INFO:     174.1.60.1:59186 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:47] INFO:     174.1.60.1:59200 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:48] INFO:     174.1.60.1:59202 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:49] INFO:     174.1.60.1:59214 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:50] INFO:     174.1.60.1:59220 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:52] INFO:     174.1.60.1:40566 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:53] INFO:     174.1.60.1:40576 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:54] INFO:     174.1.60.1:40588 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:55] INFO:     174.1.60.1:40596 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:56] INFO:     174.1.60.1:40598 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:57] INFO:     174.1.60.1:40604 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:58] INFO:     174.1.60.1:40608 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:59] INFO:     174.1.60.1:40610 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:00] INFO:     174.1.60.1:40622 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:01] INFO:     174.1.60.1:41102 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:02] INFO:     174.1.60.1:41108 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:03] INFO:     174.1.60.1:41122 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:04] INFO:     174.1.60.1:41124 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:05] INFO:     174.1.60.1:41134 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:06] INFO:     174.1.60.1:41140 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:07] INFO:     174.1.60.1:41142 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:08] INFO:     174.1.60.1:41158 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:09] INFO:     174.1.60.1:41160 - "GET /health HTTP/1.1" 503 Service Unavailable
+[rank3]:[W827 04:35:10.359210278 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank2]:[W827 04:35:10.359272236 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank0]:[W827 04:35:10.360047094 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank1]:[W827 04:35:10.370610409 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[2026-08-27 04:35:10 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.28
+[2026-08-27 04:35:10 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.29
+[2026-08-27 04:35:10] INFO:     127.0.0.1:60364 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 04:35:10] Disaggregation warmup requests completed for all 1 DP ranks
+[2026-08-27 04:35:10] End of disaggregation warmup
+[2026-08-27 04:35:10] The server is fired up and ready to roll!
+[2026-08-27 04:35:29 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.39
+[2026-08-27 04:35:29 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.39
+[2026-08-27 04:35:31 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25.83
+[2026-08-27 04:35:31 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25.84
+[2026-08-27 04:35:32] INFO:     174.1.60.1:49902 - "GET /health HTTP/1.1" 200 OK
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/ready_p_node1.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/ready_p_node1.log
new file mode 100644
index 0000000..88def47
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/ready_p_node1.log
@@ -0,0 +1,745 @@
+/usr/local/lib/python3.12/dist-packages/torchao/quantization/quant_api.py:1731: SyntaxWarning: invalid escape sequence '\.'
+  """Configuration class for applying different quantization configs to modules or parameters based on their fully qualified names (FQNs).
+/sgl-workspace/sglang/python/sglang/launch_server.py:56: UserWarning: 'python -m sglang.launch_server' is still supported, but 'sglang serve' is the recommended entrypoint.
+  Example: sglang serve --model-path <model> [options]
+  warnings.warn(
+[2026-08-27 04:28:43] Breakable CUDA graph is incompatible with MLA attention; disabling prefill CUDA graph.
+[2026-08-27 04:28:43] Attention backend not specified. Use triton backend by default.
+[2026-08-27 04:28:43] Pipeline parallelism is incompatible with overlap schedule.
+[2026-08-27 04:28:46] server_args=ServerArgs(model_path='/data/hf_models/Kimi-K3', tokenizer_path='/data/hf_models/Kimi-K3', tokenizer_mode='auto', tokenizer_backend='huggingface', tokenizer_worker_num=1, detokenizer_worker_num=1, skip_tokenizer_init=False, load_format='auto', model_loader_extra_config='{}', trust_remote_code=True, context_length=None, is_embedding=False, enable_multimodal=None, revision=None, model_impl='auto', model_config_parser='auto', json_model_override_args='{}', dtype='auto', quantization=None, quantization_param_path=None, kv_cache_dtype='auto', enable_fp32_lm_head=False, modelopt_quant=None, modelopt_checkpoint_restore_path=None, modelopt_checkpoint_save_path=None, modelopt_export_path=None, quantize_and_serve=False, rl_quant_profile=None, enable_tf32_matmul=False, mem_fraction_static=0.88, max_running_requests=None, max_queued_requests=None, max_total_tokens=None, chunked_prefill_size=8192, enable_dynamic_chunking=False, max_prefill_tokens=16384, prefill_max_requests=None, schedule_policy='fcfs', enable_priority_scheduling=False, disable_priority_preemption=False, default_priority_value=None, abort_on_priority_when_disabled=False, schedule_low_priority_values_first=False, priority_scheduling_preemption_threshold=10, retraction_policy='length', schedule_conservativeness=1.0, page_size=64, swa_full_tokens_ratio=0.8, disable_hybrid_swa_memory=False, radix_eviction_policy='lru', prefill_only_disable_kv_cache=False, disable_radix_cache=True, enable_page_major_kv_layout=False, enable_unified_memory=False, disable_chunked_prefix_cache=False, disable_overlap_schedule=True, num_continuous_decode_steps=1, scheduler_recv_interval=1, enable_mixed_chunk=False, nccl_port=None, dist_timeout=3600, dist_init_addr='174.1.60.1:20000', nnodes=4, node_rank=1, tp_size=4, dcp_size=1, pp_size=8, pp_max_micro_batch_size=None, pp_async_batch_depth=0, dp_size=1, load_balance_method='follow_bootstrap_room', attn_cp_size=1, moe_dp_size=1, dcp_comm_backend='ag_rs', dcp_replicate_q_proj=None, dwdp_size=1, enable_prefill_cp=False, cp_strategy=None, enable_dsa_cache_layer_split=False, enable_dsa_prefill_context_parallel=False, dsa_prefill_cp_mode='round-robin-split', enable_prefill_context_parallel=False, prefill_cp_mode='in-seq-split', enable_cp_decode_attn_tp=False, enable_dp_attention=False, enable_dp_attention_local_control_broadcast=False, enable_dp_lm_head=False, enable_attn_tp_input_scattered=False, disable_attn_tp_gather=False, enable_p2p_check=False, device='cuda', base_gpu_id=0, gpu_id_step=1, random_seed=577755302, watchdog_timeout=300, soft_watchdog_timeout=None, sleep_on_idle=False, use_ray=False, custom_sigquit_handler=None, numa_node=None, gc_threshold=None, host='0.0.0.0', port=30000, fastapi_root_path='', smg_grpc_mode=False, grpc_mode=False, grpc_port=None, sidecar=None, sidecar_args=None, skip_server_warmup=False, warmups=None, enable_http2=False, ssl_keyfile=None, ssl_certfile=None, ssl_ca_certs=None, ssl_keyfile_password=None, enable_ssl_refresh=False, api_key=None, admin_api_key=None, served_model_name='kimi-k3', weight_version='default', chat_template=None, hf_chat_template_name=None, completion_template=None, file_storage_path='sglang_storage', enable_cache_report=False, reasoning_parser=None, default_chat_template_kwargs=None, strip_thinking_cache=False, enable_strict_thinking=False, tool_call_parser=None, tool_server=None, sampling_defaults='model', asr_max_buffer_seconds=60, asr_max_concurrent_sessions=32, preferred_sampling_params=None, allow_auto_truncate=False, stream_interval=1, batch_notify_size=16, stream_response_default_include_usage=False, incremental_streaming_output=False, enable_streaming_session=False, enable_session_radix_cache=False, log_level='info', log_level_http=None, log_requests=False, log_requests_level=2, log_requests_format='text', log_requests_target=None, uvicorn_access_log_exclude_prefixes=[], crash_dump_folder=None, show_time_cost=False, enable_metrics=False, smg_http_sidecar_port=None, enable_mfu_metrics=False, enable_metrics_for_all_schedulers=False, load_snapshot_publish_interval=15, tokenizer_metrics_custom_labels_header='x-custom-labels', tokenizer_metrics_allowed_custom_labels=None, extra_metric_labels=None, bucket_time_to_first_token=None, bucket_inter_token_latency=None, bucket_e2e_request_latency=None, prompt_tokens_buckets=None, generation_tokens_buckets=None, gc_warning_threshold_secs=0.0, decode_log_interval=40, enable_request_time_stats_logging=False, kv_events_config=None, enable_forward_pass_metrics=False, forward_pass_metrics_worker_id='', forward_pass_metrics_ipc_name=None, enable_trace=False, trace_modules='request', otlp_traces_endpoint='localhost:4317', export_metrics_to_file=False, export_metrics_to_file_dir=None, stat_loggers=None, constrained_json_whitespace_pattern=None, constrained_json_disable_any_whitespace=False, attention_backend='triton', decode_attention_backend=None, prefill_attention_backend=None, sampling_backend='flashinfer', grammar_backend='xgrammar', radix_cache_backend=None, mm_attention_backend=None, fp8_gemm_runner_backend='auto', fp4_gemm_runner_backend='auto', bf16_gemm_backend='auto', dsa_prefill_backend=None, dsa_decode_backend=None, dsa_paged_mqa_logits_backend='auto', dsa_topk_backend='sgl-kernel', disable_flashinfer_autotune=False, flashinfer_autotune_skip_ops=None, mamba_backend='triton', cuda_graph_config=CudaGraphConfig(decode=PhaseConfig(backend='disabled', max_bs=16, bs=[1, 2, 4, 8, 12, 16], tc_compiler='eager', full_prefill_max_req=None), prefill=PhaseConfig(backend='disabled', max_bs=2048, bs=[4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], tc_compiler='eager', full_prefill_max_req=None)), cuda_graph_backend_decode=None, cuda_graph_backend_prefill=None, cuda_graph_max_bs_decode=16, cuda_graph_max_bs_prefill=None, cuda_graph_bs_decode=None, cuda_graph_bs_prefill=None, cuda_graph_tc_compiler=None, disable_prefill_cuda_graph=False, disable_decode_cuda_graph=False, disable_cuda_graph=False, disable_cuda_graph_padding=False, enable_profile_cuda_graph=False, enable_cudagraph_gc=False, debug_cuda_graph=False, enable_layerwise_nvtx_marker=False, enable_nccl_nvls=False, enable_symm_mem=False, triton_attention_reduce_in_fp32=False, triton_attention_num_kv_splits=8, triton_attention_split_tile_size=None, flashinfer_mla_disable_ragged=False, enable_fused_qk_norm_rope=False, enable_precise_embedding_interpolation=False, enable_fused_moe_sum_all_reduce=False, enable_deepseek_v4_fp4_indexer=False, disable_custom_all_reduce=False, enable_mscclpp=False, enable_torch_symm_mem=False, enable_scattered_sconv=False, pre_warm_nccl=False, enable_quant_communications=False, enable_flashinfer_allreduce_fusion=False, enforce_disable_flashinfer_allreduce_fusion=False, flashinfer_allreduce_fusion_backend=None, enable_aiter_allreduce_fusion=False, enable_torch_compile=False, enable_torch_compile_debug_mode=False, torch_compile_max_bs=32, torchao_config='', speculative_algorithm=None, speculative_draft_model_path=None, speculative_draft_model_revision=None, speculative_draft_load_format=None, speculative_num_steps=None, speculative_eagle_topk=None, speculative_num_draft_tokens=None, speculative_dflash_block_size=None, speculative_dspark_block_size=None, speculative_dspark_sps_table_path=None, speculative_dspark_confidence_sts_path=None, speculative_dspark_align_verify_tokens_to_graph_tier=False, speculative_accept_threshold_single=1.0, speculative_accept_threshold_acc=1.0, speculative_use_rejection_sampling=False, speculative_token_map=None, speculative_attention_mode='prefill', speculative_draft_attention_backend=None, speculative_draft_window_size=None, speculative_moe_runner_backend='flashinfer_mxfp4', speculative_moe_a2a_backend=None, speculative_draft_model_quantization=None, speculative_skip_dp_mlp_sync=False, enable_multi_layer_eagle=False, speculative_adaptive=False, speculative_adaptive_config=None, decoupled_spec_bind_endpoint=None, decoupled_spec_connect_endpoints=None, decoupled_spec_rank=None, decoupled_spec_role='null', spec_trace_dir=None, speculative_ngram_min_bfs_breadth=1, speculative_ngram_max_bfs_breadth=10, speculative_ngram_match_type='BFS', speculative_ngram_max_trie_depth=18, speculative_ngram_capacity=10000000, speculative_ngram_external_corpus_path=None, speculative_ngram_external_sam_budget=0, speculative_ngram_external_corpus_max_tokens=10000000, ep_size=4, moe_a2a_backend='none', moe_runner_backend='flashinfer_mxfp4', flashinfer_mxfp4_moe_precision='default', deepep_mode='auto', fuseep_mode=2, deepep_dispatcher_output_dtype='auto', ep_num_redundant_experts=0, ep_dispatch_algorithm=None, init_expert_location='trivial', enable_eplb=False, eplb_algorithm='auto', eplb_rebalance_num_iterations=1000, eplb_rebalance_layers_per_chunk=None, eplb_min_rebalancing_utilization_threshold=1.0, expert_distribution_recorder_mode=None, expert_distribution_recorder_buffer_size=1000, enable_expert_distribution_metrics=False, deepep_config=None, moe_dense_tp_size=None, elastic_ep_backend=None, enable_elastic_expert_backup=False, mooncake_ib_device=None, enable_waterfill=False, ep_join_mode=None, ep_join_rank_offset=0, elastic_ep_initial_size=None, max_ep_size=None, elastic_ep_scale_timeout=600, elastic_ep_rejoin=False, disable_flashinfer_cutlass_moe_fp4_allgather=False, disable_shared_experts_fusion=False, enforce_shared_experts_fusion=False, max_mamba_cache_size=None, mamba_ssm_dtype=None, mamba_max_states_per_path=-1, enable_mamba_cache_stochastic_rounding=False, mamba_cache_philox_rounds=0, mamba_full_memory_ratio=0.36, mamba_radix_cache_strategy='extra_buffer_lazy', uses_mamba_radix_cache=False, mamba_track_interval=256, enable_int8_mamba_checkpoint=False, int8_mamba_ckpt_size=None, linear_attn_backend='triton', linear_attn_decode_backend=None, linear_attn_prefill_backend=None, linear_attn_verify_backend=None, enable_linear_replayssm=False, linear_replayssm_cache_len=16, enable_linear_replayssm_spec=False, enable_hierarchical_cache=False, hicache_ratio=2.0, hicache_size=0, hicache_write_policy='write_through', hicache_io_backend='kernel', hicache_mem_layout='page_first', hicache_storage_backend=None, hicache_storage_prefetch_policy='timeout', hicache_storage_backend_extra_config=None, enable_hisparse=False, hisparse_config=None, enable_broadcast_mm_inputs_process=False, enable_prefix_mm_cache=False, mm_enable_dp_encoder=False, mm_process_config={}, mm_processor_worker_num=0, mm_io_worker_num=0, limit_mm_data_per_request=None, enable_mm_global_cache=False, disable_fast_image_processor=False, mm_feature_transport='cpu', keep_mm_feature_on_device=False, enable_lora=None, enable_lora_overlap_loading=None, max_lora_rank=None, lora_target_modules=None, lora_paths=None, max_loaded_loras=None, max_loras_per_batch=8, lora_eviction_policy='lru', lora_backend='csgmv', max_lora_chunk_size=16, experts_shared_outer_loras=None, lora_use_virtual_experts=False, lora_strict_loading=False, lora_drain_wait_threshold=0.0, enable_two_batch_overlap=False, enable_single_batch_overlap=False, tbo_token_distribution_threshold=0.48, cpu_offload_gb=0, offload_group_size=-1, offload_num_in_group=1, offload_prefetch_step=1, offload_mode='cpu', enable_lmcache=False, lmcache_config_file=None, enable_flexkv=False, flexkv_config_file=None, kt_weight_path=None, kt_method='AMXINT4', kt_cpuinfer=None, kt_threadpool_count=2, kt_num_gpu_experts=None, kt_max_deferred_experts_per_token=None, dllm_algorithm=None, dllm_algorithm_config=None, dllm_fdfo=True, disaggregation_mode='prefill', disaggregation_transfer_backend='mooncake', disaggregation_bootstrap_port=28800, disaggregation_ib_device='mlx5_0,mlx5_1,mlx5_2,mlx5_3', disaggregation_decode_enable_radix_cache=False, disaggregation_decode_enable_offload_kvcache=False, num_reserved_decode_tokens=512, disaggregation_decode_extra_slots=None, disaggregation_decode_polling_interval=1, optimistic_prefill_attempts=0, encoder_only=False, language_only=False, encoder_transfer_backend='zmq_to_tokenizer', encoder_urls=[], encoder_bootstrap_port=8997, encoder_register_urls=[], enable_adaptive_dispatch_to_encoder=False, enable_pdmux=False, pdmux_config_path=None, sm_group_num=8, custom_weight_loader=[], weight_loader_disable_mmap=False, weight_loader_prefetch_checkpoints=False, weight_loader_prefetch_num_threads=4, weight_loader_drop_cache_after_load=False, remote_instance_weight_loader_seed_instance_ip=None, remote_instance_weight_loader_seed_instance_service_port=None, remote_instance_weight_loader_send_weights_group_ports=None, remote_instance_weight_loader_backend='nccl', remote_instance_weight_loader_start_seed_via_transfer_engine=False, engine_info_bootstrap_port=6789, modelexpress_config=None, download_dir=None, model_checksum=None, delete_ckpt_after_loading=False, decrypted_config_file=None, decrypted_draft_config_file=None, checkpoint_engine_wait_weights_before_ready=False, enable_prefill_delayer=False, prefill_delayer_max_delay_passes=30, prefill_delayer_token_usage_low_watermark=None, prefill_delayer_forward_passes_buckets=None, prefill_delayer_wait_seconds_buckets=None, prefill_delayer_queue_min_ratio=None, prefill_delayer_max_delay_ms=None, min_free_slots_delay=None, enable_deterministic_inference=False, rl_on_policy_target=None, kv_canary='none', kv_canary_real_data='none', kv_canary_sweep_interval=0, enable_dynamic_batch_tokenizer=False, dynamic_batch_tokenizer_batch_size=32, dynamic_batch_tokenizer_batch_timeout=0.002, enable_tokenizer_batch_encode=False, disable_tokenizer_batch_decode=False, debug_tensor_dump_output_folder=None, debug_tensor_dump_layers=None, debug_tensor_dump_input_file=None, enable_memory_saver=False, enable_weights_cpu_backup=False, enable_draft_weights_cpu_backup=False, enable_custom_logit_processor=False, enable_return_hidden_states=False, enable_return_routed_experts=False, enable_return_indexer_topk=False, disable_outlines_disk_cache=False, enable_mis=False, weight_cache_mode='off', weight_cache_socket=None, weight_cache_timeout=1800, forward_hooks=None, msprobe_dump_config=None)
+[2026-08-27 04:28:46] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:46] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:46] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:46] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:46] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:46] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:46] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:46] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:02 PP3 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:02 PP3 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:03 PP3 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:03 PP3 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:03 PP3 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:03 PP3 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:03 PP2 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:03 PP2 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:03 PP2 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:03 PP2 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:04 PP2 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:04 PP2 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:04 PP2 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:04 PP2 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:04 PP3 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:04 PP3 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:05 PP3 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:05 PP2 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:05 PP3 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:05 PP3 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:05 PP2 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:06.044592   241 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:06.044694   241 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:06.044710   241 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.2 port: 12001
+I0827 04:29:06.044759   241 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.2:16343
+I0827 04:29:06.044883   241 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:06.053642   242 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:06.053738   242 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:06.053754   242 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.2 port: 12001
+I0827 04:29:06.053797   242 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.2:15170
+I0827 04:29:06.053925   242 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:06.058039   237 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:06.058135   237 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:06.058149   237 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.2 port: 12001
+I0827 04:29:06.058192   237 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.2:15486
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:06.058215   236 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:06.058310   236 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:06.058316   237 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:06.058326   236 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.2 port: 12001
+I0827 04:29:06.058367   236 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.2:16140
+I0827 04:29:06.058492   236 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:06.059734   239 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:06.059828   239 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:06.059844   239 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.2 port: 12001
+I0827 04:29:06.059890   239 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.2:15755
+I0827 04:29:06.060027   239 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:06.064433   237 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:06.064513   236 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:06.065588   239 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:06.069945   237 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:06.070852   236 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:06.071113   239 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:06.071978   241 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:06.079348   236 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:06.079967   242 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:06.080070   241 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:06.087044   241 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:06.087074   242 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:06.094024   242 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:06.095021   241 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:06.095822   237 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:06.097810   241 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:06.097839   241 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:06.097889   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.099939   239 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:06.100984   242 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:06.103683   242 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:06.103708   242 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:06.103756   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.103785   236 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:06.103833   237 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:06.106014   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:06.106982   236 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:06.107007   236 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:06.107053   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.107446   241 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:02
+I0827 04:29:06.107475   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.107936   239 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:06.107971   237 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:06.107997   237 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:06.108048   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.110760   239 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:06.110790   239 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:06.110847   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.111824   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:06.113214   242 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:02
+I0827 04:29:06.113242   242 rdma_context.cpp:205] Using SIEVE endpoint store
+[2026-08-27 04:29:06 PP2 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+I0827 04:29:06.114821   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:06.115196   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+[2026-08-27 04:29:06 PP2 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+I0827 04:29:06.116189   241 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:02
+I0827 04:29:06.116214   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.116621   237 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:02
+I0827 04:29:06.116650   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.118325   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:06.119874   239 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:02
+I0827 04:29:06.119906   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.121601   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:06.122952   242 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:02
+I0827 04:29:06.122977   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.127393   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:06.128841   239 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:02
+I0827 04:29:06.128867   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.132040   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:06.133428   236 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:02
+I0827 04:29:06.133458   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.136312   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:06.137817   239 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:02
+I0827 04:29:06.137846   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.144285   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:06.144373   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:06.145742   237 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:02
+I0827 04:29:06.145776   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.145833   241 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:02
+I0827 04:29:06.145866   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.152084   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:06.153820   242 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:02
+I0827 04:29:06.153860   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.154352   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:06.155728   241 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:02
+I0827 04:29:06.155835   241 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:06 PP3 TP2 EP2] Init torch distributed begin.
+I0827 04:29:06.163965   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:06.164443   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:06.164605   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:06.165319   236 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:02
+I0827 04:29:06.165345   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.165951   239 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:02
+I0827 04:29:06.166070   239 transfer_engine_impl.cpp:388] installTransport, type=rdma
+I0827 04:29:06.166186   242 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:02
+I0827 04:29:06.166311   242 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:06 PP3 TP0 EP0] Init torch distributed begin.
+[2026-08-27 04:29:06 PP3 TP3 EP3] Init torch distributed begin.
+I0827 04:29:06.172065   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:06.173699   237 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:02
+I0827 04:29:06.173743   237 rdma_context.cpp:205] Using SIEVE endpoint store
+[2026-08-27 04:29:06 PP3 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+I0827 04:29:06.192251   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:06.193763   236 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:02
+I0827 04:29:06.193797   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.200259   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:06.201810   237 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:02
+I0827 04:29:06.201937   237 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:06 PP2 TP2 EP2] Init torch distributed begin.
+I0827 04:29:06.220070   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:06.221505   236 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:02
+I0827 04:29:06.221653   236 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:06 PP2 TP1 EP1] Init torch distributed begin.
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:06.613179   235 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:06.613286   235 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:06.613301   235 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.2 port: 12001
+I0827 04:29:06.613344   235 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.2:16382
+I0827 04:29:06.613492   235 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:06.618471   235 topology.cpp:127] Device mlx5_0 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:06.618719   238 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:06.618824   238 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:06.618839   238 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.2 port: 12001
+I0827 04:29:06.618882   238 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.2:16170
+I0827 04:29:06.619020   238 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:06.624203   238 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:06.629066   238 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:06.633975   238 topology.cpp:127] Device mlx5_2 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:06.636366   240 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:06.636454   240 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:06.636466   240 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.2 port: 12001
+I0827 04:29:06.636509   240 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.2:15840
+I0827 04:29:06.636675   240 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:06.638876   238 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:06.640780   238 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:06.640802   238 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:06.640842   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.641396   240 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:06.643196   235 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:06.645701   240 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:06.649896   235 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:06.650310   240 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:06.654819   240 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:06.656375   240 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:06.656396   240 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:06.656432   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.657156   235 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:06.659863   235 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:06.659895   235 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:06.659950   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.661671   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:06.662956   240 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:02
+I0827 04:29:06.662979   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.664074   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:06.665762   238 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:02
+I0827 04:29:06.665796   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.667351   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:06.668808   235 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:02
+I0827 04:29:06.668838   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.676294   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:06.677721   235 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:02
+I0827 04:29:06.677748   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.692129   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:06.693621   240 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:02
+I0827 04:29:06.693648   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.696089   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:06.697444   238 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:02
+I0827 04:29:06.697470   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.707720   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:06.709362   235 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:02
+I0827 04:29:06.709394   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.719987   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:06.721463   240 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:02
+I0827 04:29:06.721489   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.728108   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:06.728855   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:06.729467   238 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:02
+I0827 04:29:06.729491   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.730392   240 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:02
+I0827 04:29:06.730533   240 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:06 PP3 TP1 EP1] Init torch distributed begin.
+I0827 04:29:06.736888   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:06.738317   238 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:02
+I0827 04:29:06.738422   238 transfer_engine_impl.cpp:388] installTransport, type=rdma
+I0827 04:29:06.738466   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+[2026-08-27 04:29:06 PP2 TP3 EP3] Init torch distributed begin.
+I0827 04:29:06.739869   235 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:02
+I0827 04:29:06.739982   235 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:06 PP2 TP0 EP0] Init torch distributed begin.
+[2026-08-27 04:29:26 PP3 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:26 PP2 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:27 PP2 TP3 EP3] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP2 TP1 EP1] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP2 TP2 EP2] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP2 TP0 EP0] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP3 TP3 EP3] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP3 TP0 EP0] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP3 TP2 EP2] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP3 TP1 EP1] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP3 TP3 EP3] Init torch distributed ends. elapsed=21.61 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP3 TP2 EP2] Init torch distributed ends. elapsed=21.62 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP3 TP1 EP1] Init torch distributed ends. elapsed=21.04 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP3 TP0 EP0] Init torch distributed ends. elapsed=21.61 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP2 TP2 EP2] Init torch distributed ends. elapsed=21.57 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP2 TP1 EP1] Init torch distributed ends. elapsed=21.55 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP2 TP0 EP0] Init torch distributed ends. elapsed=21.03 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP2 TP3 EP3] Init torch distributed ends. elapsed=21.04 s, mem usage=0.54 GB
+[2026-08-27 04:29:28 PP3 TP3 EP3] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP3 TP0 EP0] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP2 TP0 EP0] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP3 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP2 TP1 EP1] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP3 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP2 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP3 TP2 EP2] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP2 TP3 EP3] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP2 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP2 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP3 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP2 TP2 EP2] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP3 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP3 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP2 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP3 TP1 EP1] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP2 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP3 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP2 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP3 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP3 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP2 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP3 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP2 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP3 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP2 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP2 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP2 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP2 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP3 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP3 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:30:31 PP2 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:31 PP2 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:31 PP2 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:31 PP2 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:31 PP2 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:31 PP2 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:31 PP2 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:31 PP2 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP2 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:32 PP2 TP0 EP0] Load weight end. elapsed=64.92 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=36.14 GB, mem usage=45.91 GB.
+[2026-08-27 04:30:33 PP3 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:33 PP3 TP1 EP1] Load weight end. elapsed=65.31 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:34 PP2 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:34 PP2 TP1 EP1] Load weight end. elapsed=66.07 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=36.14 GB, mem usage=45.91 GB.
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:35 PP3 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:35 PP3 TP0 EP0] Load weight end. elapsed=67.41 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:36 PP2 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:36 PP2 TP3 EP3] Load weight end. elapsed=68.04 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=36.14 GB, mem usage=45.91 GB.
+[2026-08-27 04:30:36 PP2 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:36 PP2 TP2 EP2] Load weight end. elapsed=68.23 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=36.14 GB, mem usage=45.91 GB.
+[2026-08-27 04:30:36 PP2 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:36 PP2 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:36 PP2 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:36 PP2 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:36 PP2 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:36 PP2 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:36 PP2 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:36 PP2 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:36 PP3 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:36 PP3 TP3 EP3] Load weight end. elapsed=68.86 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:37 PP3 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:37 PP3 TP2 EP2] Load weight end. elapsed=69.68 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:38 PP3 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:38 PP3 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:38 PP3 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:38 PP3 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:38 PP3 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:38 PP3 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:38 PP3 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:38 PP3 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:54 PP3 TP3 EP3] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP3 TP2 EP2] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP3 TP1 EP1] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP2 TP3 EP3] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP3 TP0 EP0] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP2 TP2 EP2] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP2 TP1 EP1] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP2 TP0 EP0] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP2 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:54 PP2 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:54 PP3 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP3 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP2 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:54 PP3 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP3 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP2 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:54 PP3 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP3 TP3 EP3] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP2 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP2 TP2 EP2] Memory pool end. avail mem=23.25 GB
+[2026-08-27 04:30:54 PP2 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP2 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP2 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP3 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP3 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP2 TP3 EP3] Memory pool end. avail mem=23.25 GB
+[2026-08-27 04:30:54 PP2 TP1 EP1] Memory pool end. avail mem=23.25 GB
+[2026-08-27 04:30:54 PP3 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP3 TP1 EP1] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP3 TP2 EP2] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP2 TP0 EP0] Memory pool end. avail mem=23.25 GB
+[2026-08-27 04:30:54 PP3 TP0 EP0] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP3 TP3 EP3] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp3_pp3_dp0.json
+[2026-08-27 04:30:54 PP2 TP1 EP1] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp1_pp2_dp0.json
+[2026-08-27 04:30:54 PP2 TP2 EP2] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp2_pp2_dp0.json
+[2026-08-27 04:30:54 PP3 TP1 EP1] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp1_pp3_dp0.json
+[2026-08-27 04:30:54 PP2 TP3 EP3] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp3_pp2_dp0.json
+[2026-08-27 04:30:54 PP3 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:30:54 PP3 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:30:54 PP3 TP2 EP2] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp2_pp3_dp0.json
+[2026-08-27 04:30:54 PP2 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:30:54 PP2 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO[TensorRT-LLM][INFO] Set logger level to INFO
+
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[2026-08-27 04:30:54 PP3 TP0 EP0] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp0_pp3_dp0.json
+[2026-08-27 04:30:54 PP2 TP0 EP0] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp0_pp2_dp0.json
+[TensorRT-LLM][INFO] Set logger level to INFO
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.65profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.60profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.79profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.51profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.86profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.27profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.62profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.98profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.86profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 12.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 13.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.93profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.92profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.09profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.63profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  6.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  7.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.33profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.58profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  8.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.56profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.23profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.63profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.36profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.09profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.97profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  8.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.30profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 13.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 13.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  8.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  8.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 11.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 11.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.23profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.48profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.23profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.53profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.27profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.59profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.62profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.97profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.60profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.46profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.12profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.68profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.77profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.62profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  6.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.87profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.30profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.40profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.83profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.94profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.59profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.66profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.29profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.23profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:01,  3.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.68profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.75profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  7.01profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.22profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.65profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.62profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.46profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.22profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  7.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:01<00:06,  1.03s/profile]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:01,  3.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.50profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.93profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.48profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.11profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.12profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.30profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.68profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  7.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 11.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  9.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 12.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 11.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 11.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 11.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.11profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.31profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.65profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.31profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.35profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.49profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:01<00:01,  2.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.65profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.68profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  7.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.09profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 11.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.88profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.98profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.50profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.62profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.51profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.91profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:01,  3.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.31profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.86profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.27profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.59profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.17profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.76profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.73profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  9.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.37profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.15profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.98profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.65profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.97profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.23profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.46profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  9.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.03profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.57profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.40profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.40profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.57profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.13profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.56profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  6.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.73profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.23profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  3.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.86profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.77profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.92profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.37profile/s]
+[2026-08-27 04:31:26 PP2 TP3 EP3] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.39profile/s]
+[2026-08-27 04:31:26 PP2 TP2 EP2] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.75profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.46profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.35profile/s]
+[2026-08-27 04:31:26 PP2 TP1 EP1] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.12profile/s]
+[2026-08-27 04:31:26 PP2 TP0 EP0] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.66profile/s][2026-08-27 04:31:26 PP2 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:26 PP2 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:26 PP2 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:26 PP2 TP0 EP0] max_total_num_tokens=3801408, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=54, context_len=1048576, available_gpu_mem=20.47 GB
+[2026-08-27 04:31:26 PP2 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:26 PP2 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:26 PP2 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:26 PP2 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:26 PP2 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.93profile/s]
+[2026-08-27 04:31:27 PP3 TP1 EP1] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.16profile/s]
+[2026-08-27 04:31:27 PP3 TP0 EP0] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.43profile/s]
+[2026-08-27 04:31:27 PP3 TP2 EP2] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.05profile/s]
+[2026-08-27 04:31:27 PP3 TP3 EP3] FlashInfer autotune completed.
+[2026-08-27 04:31:27 PP3 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:27 PP3 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:27 PP3 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:27 PP3 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:27 PP3 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:27 PP3 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:27 PP3 TP0 EP0] max_total_num_tokens=3801408, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=54, context_len=1048576, available_gpu_mem=16.54 GB
+[2026-08-27 04:31:27 PP3 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:27 PP3 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+W0827 04:31:30.590376  2521 memory_location.cpp:73] Failed to get NUMA node, addr: 0x513c5ec0, len: 6912: Operation not permitted [1]
+W0827 04:31:30.590432  2522 memory_location.cpp:73] Failed to get NUMA node, addr: 0x514a0900, len: 6912: Operation not permitted [1]
+W0827 04:31:30.590495  2525 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51226dc0, len: 55296: Operation not permitted [1]
+W0827 04:31:30.590492  2523 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5128db40, len: 6912: Operation not permitted [1]
+W0827 04:31:30.590533  2524 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5156ff80, len: 6912: Operation not permitted [1]
+W0827 04:31:30.590600  2526 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51320ac0, len: 55296: Operation not permitted [1]
+W0827 04:31:30.590636  2527 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51768540, len: 6912: Operation not permitted [1]
+W0827 04:31:30.590826  2528 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51483dc0, len: 13824: Operation not permitted [1]
+W0827 04:31:30.590914  2529 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51119380, len: 6912: Operation not permitted [1]
+W0827 04:31:30.590979  2530 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50fe1e00, len: 6912: Operation not permitted [1]
+W0827 04:31:30.846120  2547 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50ceebc0, len: 6912: Operation not permitted [1]
+W0827 04:31:30.846271  2548 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50c57640, len: 6912: Operation not permitted [1]
+W0827 04:31:30.846328  2549 memory_location.cpp:73] Failed to get NUMA node, addr: 0x502ed680, len: 6912: Operation not permitted [1]
+W0827 04:31:30.846354  2550 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50d1c800, len: 6912: Operation not permitted [1]
+W0827 04:31:30.846448  2551 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50510a40, len: 55296: Operation not permitted [1]
+W0827 04:31:30.846473  2552 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5051e280, len: 55296: Operation not permitted [1]
+W0827 04:31:30.846571  2553 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50d01340, len: 6912: Operation not permitted [1]
+W0827 04:31:30.846689  2554 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50ce3440, len: 13824: Operation not permitted [1]
+W0827 04:31:30.846858  2555 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5030f440, len: 6912: Operation not permitted [1]
+W0827 04:31:30.846926  2556 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50cae980, len: 6912: Operation not permitted [1]
+W0827 04:31:31.198441  2575 memory_location.cpp:73] Failed to get NUMA node, addr: 0x514f46c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.198501  2576 memory_location.cpp:73] Failed to get NUMA node, addr: 0x509c6400, len: 6912: Operation not permitted [1]
+W0827 04:31:31.198501  2577 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51513300, len: 6912: Operation not permitted [1]
+W0827 04:31:31.198679  2578 memory_location.cpp:73] Failed to get NUMA node, addr: 0x506eef40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.198787  2579 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50dc9180, len: 55296: Operation not permitted [1]
+W0827 04:31:31.198915  2580 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50904700, len: 55296: Operation not permitted [1]
+W0827 04:31:31.198982  2581 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50aa9440, len: 6912: Operation not permitted [1]
+W0827 04:31:31.199081  2582 memory_location.cpp:73] Failed to get NUMA node, addr: 0x515012c0, len: 13824: Operation not permitted [1]
+W0827 04:31:31.199148  2583 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5161ac80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.199185  2584 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50dd69c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.763885  2603 memory_location.cpp:73] Failed to get NUMA node, addr: 0x519f1500, len: 6912: Operation not permitted [1]
+W0827 04:31:31.763998  2605 memory_location.cpp:73] Failed to get NUMA node, addr: 0x517df940, len: 6912: Operation not permitted [1]
+W0827 04:31:31.763923  2604 memory_location.cpp:73] Failed to get NUMA node, addr: 0x514ab440, len: 6912: Operation not permitted [1]
+W0827 04:31:31.764139  2606 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51556180, len: 6912: Operation not permitted [1]
+W0827 04:31:31.764169  2608 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50ed8140, len: 55296: Operation not permitted [1]
+W0827 04:31:31.764184  2607 memory_location.cpp:73] Failed to get NUMA node, addr: 0x515df040, len: 55296: Operation not permitted [1]
+W0827 04:31:31.764356  2609 memory_location.cpp:73] Failed to get NUMA node, addr: 0x517e3580, len: 6912: Operation not permitted [1]
+W0827 04:31:31.764564  2610 memory_location.cpp:73] Failed to get NUMA node, addr: 0x519cdb80, len: 13824: Operation not permitted [1]
+W0827 04:31:31.764624  2611 memory_location.cpp:73] Failed to get NUMA node, addr: 0x519dbf80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.764714  2612 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50f08d80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.868036  2629 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50a7a680, len: 6912: Operation not permitted [1]
+W0827 04:31:31.868046  2630 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50b6e400, len: 6912: Operation not permitted [1]
+W0827 04:31:31.868139  2631 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50e5f580, len: 6912: Operation not permitted [1]
+W0827 04:31:31.868261  2632 memory_location.cpp:73] Failed to get NUMA node, addr: 0x503c7100, len: 6912: Operation not permitted [1]
+W0827 04:31:31.868330  2633 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50c94ac0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.868432  2635 memory_location.cpp:73] Failed to get NUMA node, addr: 0x509b2880, len: 6912: Operation not permitted [1]
+W0827 04:31:31.868502  2634 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50b14c80, len: 55296: Operation not permitted [1]
+W0827 04:31:31.868564  2636 memory_location.cpp:73] Failed to get NUMA node, addr: 0x503ee840, len: 13824: Operation not permitted [1]
+W0827 04:31:31.868708  2637 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51057d80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.868835  2638 memory_location.cpp:73] Failed to get NUMA node, addr: 0x509cd940, len: 6912: Operation not permitted [1]
+W0827 04:31:32.003983  2656 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50715b00, len: 6912: Operation not permitted [1]
+W0827 04:31:32.004096  2657 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51346e80, len: 6912: Operation not permitted [1]
+W0827 04:31:32.004005  2655 memory_location.cpp:73] Failed to get NUMA node, addr: 0x507176c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.004179  2658 memory_location.cpp:73] Failed to get NUMA node, addr: 0x507e3a40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.004302  2659 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5078aa80, len: 55296: Operation not permitted [1]
+W0827 04:31:32.004357  2660 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5135c980, len: 55296: Operation not permitted [1]
+W0827 04:31:32.004410  2661 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5099f180, len: 6912: Operation not permitted [1]
+W0827 04:31:32.004585  2662 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50b287c0, len: 13824: Operation not permitted [1]
+W0827 04:31:32.004739  2663 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5070b6c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.004812  2664 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50753540, len: 6912: Operation not permitted [1]
+W0827 04:31:32.144867  2688 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50aed700, len: 6912: Operation not permitted [1]
+W0827 04:31:32.144946  2689 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50792740, len: 6912: Operation not permitted [1]
+W0827 04:31:32.145030  2690 memory_location.cpp:73] Failed to get NUMA node, addr: 0x508fca40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.145135  2691 memory_location.cpp:73] Failed to get NUMA node, addr: 0x505d7c40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.145318  2693 memory_location.cpp:73] Failed to get NUMA node, addr: 0x505ec340, len: 55296: Operation not permitted [1]
+W0827 04:31:32.145387  2692 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50602cc0, len: 55296: Operation not permitted [1]
+W0827 04:31:32.145479  2694 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50748c00, len: 6912: Operation not permitted [1]
+W0827 04:31:32.145632  2695 memory_location.cpp:73] Failed to get NUMA node, addr: 0x507e3480, len: 13824: Operation not permitted [1]
+W0827 04:31:32.145767  2696 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50316400, len: 6912: Operation not permitted [1]
+W0827 04:31:32.145864  2697 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50788600, len: 6912: Operation not permitted [1]
+W0827 04:31:32.177534  2726 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fff3c80, len: 6912: Operation not permitted [1]
+W0827 04:31:32.177623  2728 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fad6780, len: 6912: Operation not permitted [1]
+W0827 04:31:32.177625  2727 memory_location.cpp:73] Failed to get NUMA node, addr: 0x501dc2c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.178279  2729 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50596480, len: 6912: Operation not permitted [1]
+W0827 04:31:32.178463  2730 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50222dc0, len: 55296: Operation not permitted [1]
+W0827 04:31:32.178498  2731 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5038f6c0, len: 55296: Operation not permitted [1]
+W0827 04:31:32.178581  2732 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4ffd7ec0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.178695  2733 memory_location.cpp:73] Failed to get NUMA node, addr: 0x506d84c0, len: 13824: Operation not permitted [1]
+W0827 04:31:32.178884  2734 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5078efc0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.178922  2735 memory_location.cpp:73] Failed to get NUMA node, addr: 0x502c7b00, len: 6912: Operation not permitted [1]
+[2026-08-27 04:31:32] Dummy health check server started in background thread at 0.0.0.0:30000
+[rank10]:[W827 04:32:32.605391940 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank11]:[W827 04:32:32.605815131 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank9]:[W827 04:32:32.606041691 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank8]:[W827 04:32:32.608313727 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank11]:[W827 04:33:19.698792619 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank8]:[W827 04:33:19.699392665 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank9]:[W827 04:33:19.699501937 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank10]:[W827 04:33:19.699589651 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank14]:[W827 04:33:19.728965278 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank12]:[W827 04:33:19.729026815 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank13]:[W827 04:33:19.729092243 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank15]:[W827 04:33:19.730463091 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank14]:[W827 04:33:25.763484084 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank13]:[W827 04:33:25.764518868 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank15]:[W827 04:33:25.764776298 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank12]:[W827 04:33:25.765271843 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[2026-08-27 04:35:10 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.29
+[2026-08-27 04:35:10 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.29
+[2026-08-27 04:35:29 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.39
+[2026-08-27 04:35:29 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.39
+[2026-08-27 04:35:31 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25.86
+[2026-08-27 04:35:31 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25.87
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/ready_p_node2.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/ready_p_node2.log
new file mode 100644
index 0000000..934b5d8
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/ready_p_node2.log
@@ -0,0 +1,754 @@
+/usr/local/lib/python3.12/dist-packages/torchao/quantization/quant_api.py:1731: SyntaxWarning: invalid escape sequence '\.'
+  """Configuration class for applying different quantization configs to modules or parameters based on their fully qualified names (FQNs).
+/sgl-workspace/sglang/python/sglang/launch_server.py:56: UserWarning: 'python -m sglang.launch_server' is still supported, but 'sglang serve' is the recommended entrypoint.
+  Example: sglang serve --model-path <model> [options]
+  warnings.warn(
+[2026-08-27 04:28:43] Breakable CUDA graph is incompatible with MLA attention; disabling prefill CUDA graph.
+[2026-08-27 04:28:43] Attention backend not specified. Use triton backend by default.
+[2026-08-27 04:28:43] Pipeline parallelism is incompatible with overlap schedule.
+[2026-08-27 04:28:45] server_args=ServerArgs(model_path='/data/hf_models/Kimi-K3', tokenizer_path='/data/hf_models/Kimi-K3', tokenizer_mode='auto', tokenizer_backend='huggingface', tokenizer_worker_num=1, detokenizer_worker_num=1, skip_tokenizer_init=False, load_format='auto', model_loader_extra_config='{}', trust_remote_code=True, context_length=None, is_embedding=False, enable_multimodal=None, revision=None, model_impl='auto', model_config_parser='auto', json_model_override_args='{}', dtype='auto', quantization=None, quantization_param_path=None, kv_cache_dtype='auto', enable_fp32_lm_head=False, modelopt_quant=None, modelopt_checkpoint_restore_path=None, modelopt_checkpoint_save_path=None, modelopt_export_path=None, quantize_and_serve=False, rl_quant_profile=None, enable_tf32_matmul=False, mem_fraction_static=0.88, max_running_requests=None, max_queued_requests=None, max_total_tokens=None, chunked_prefill_size=8192, enable_dynamic_chunking=False, max_prefill_tokens=16384, prefill_max_requests=None, schedule_policy='fcfs', enable_priority_scheduling=False, disable_priority_preemption=False, default_priority_value=None, abort_on_priority_when_disabled=False, schedule_low_priority_values_first=False, priority_scheduling_preemption_threshold=10, retraction_policy='length', schedule_conservativeness=1.0, page_size=64, swa_full_tokens_ratio=0.8, disable_hybrid_swa_memory=False, radix_eviction_policy='lru', prefill_only_disable_kv_cache=False, disable_radix_cache=True, enable_page_major_kv_layout=False, enable_unified_memory=False, disable_chunked_prefix_cache=False, disable_overlap_schedule=True, num_continuous_decode_steps=1, scheduler_recv_interval=1, enable_mixed_chunk=False, nccl_port=None, dist_timeout=3600, dist_init_addr='174.1.60.1:20000', nnodes=4, node_rank=2, tp_size=4, dcp_size=1, pp_size=8, pp_max_micro_batch_size=None, pp_async_batch_depth=0, dp_size=1, load_balance_method='follow_bootstrap_room', attn_cp_size=1, moe_dp_size=1, dcp_comm_backend='ag_rs', dcp_replicate_q_proj=None, dwdp_size=1, enable_prefill_cp=False, cp_strategy=None, enable_dsa_cache_layer_split=False, enable_dsa_prefill_context_parallel=False, dsa_prefill_cp_mode='round-robin-split', enable_prefill_context_parallel=False, prefill_cp_mode='in-seq-split', enable_cp_decode_attn_tp=False, enable_dp_attention=False, enable_dp_attention_local_control_broadcast=False, enable_dp_lm_head=False, enable_attn_tp_input_scattered=False, disable_attn_tp_gather=False, enable_p2p_check=False, device='cuda', base_gpu_id=0, gpu_id_step=1, random_seed=336932344, watchdog_timeout=300, soft_watchdog_timeout=None, sleep_on_idle=False, use_ray=False, custom_sigquit_handler=None, numa_node=None, gc_threshold=None, host='0.0.0.0', port=30000, fastapi_root_path='', smg_grpc_mode=False, grpc_mode=False, grpc_port=None, sidecar=None, sidecar_args=None, skip_server_warmup=False, warmups=None, enable_http2=False, ssl_keyfile=None, ssl_certfile=None, ssl_ca_certs=None, ssl_keyfile_password=None, enable_ssl_refresh=False, api_key=None, admin_api_key=None, served_model_name='kimi-k3', weight_version='default', chat_template=None, hf_chat_template_name=None, completion_template=None, file_storage_path='sglang_storage', enable_cache_report=False, reasoning_parser=None, default_chat_template_kwargs=None, strip_thinking_cache=False, enable_strict_thinking=False, tool_call_parser=None, tool_server=None, sampling_defaults='model', asr_max_buffer_seconds=60, asr_max_concurrent_sessions=32, preferred_sampling_params=None, allow_auto_truncate=False, stream_interval=1, batch_notify_size=16, stream_response_default_include_usage=False, incremental_streaming_output=False, enable_streaming_session=False, enable_session_radix_cache=False, log_level='info', log_level_http=None, log_requests=False, log_requests_level=2, log_requests_format='text', log_requests_target=None, uvicorn_access_log_exclude_prefixes=[], crash_dump_folder=None, show_time_cost=False, enable_metrics=False, smg_http_sidecar_port=None, enable_mfu_metrics=False, enable_metrics_for_all_schedulers=False, load_snapshot_publish_interval=15, tokenizer_metrics_custom_labels_header='x-custom-labels', tokenizer_metrics_allowed_custom_labels=None, extra_metric_labels=None, bucket_time_to_first_token=None, bucket_inter_token_latency=None, bucket_e2e_request_latency=None, prompt_tokens_buckets=None, generation_tokens_buckets=None, gc_warning_threshold_secs=0.0, decode_log_interval=40, enable_request_time_stats_logging=False, kv_events_config=None, enable_forward_pass_metrics=False, forward_pass_metrics_worker_id='', forward_pass_metrics_ipc_name=None, enable_trace=False, trace_modules='request', otlp_traces_endpoint='localhost:4317', export_metrics_to_file=False, export_metrics_to_file_dir=None, stat_loggers=None, constrained_json_whitespace_pattern=None, constrained_json_disable_any_whitespace=False, attention_backend='triton', decode_attention_backend=None, prefill_attention_backend=None, sampling_backend='flashinfer', grammar_backend='xgrammar', radix_cache_backend=None, mm_attention_backend=None, fp8_gemm_runner_backend='auto', fp4_gemm_runner_backend='auto', bf16_gemm_backend='auto', dsa_prefill_backend=None, dsa_decode_backend=None, dsa_paged_mqa_logits_backend='auto', dsa_topk_backend='sgl-kernel', disable_flashinfer_autotune=False, flashinfer_autotune_skip_ops=None, mamba_backend='triton', cuda_graph_config=CudaGraphConfig(decode=PhaseConfig(backend='disabled', max_bs=16, bs=[1, 2, 4, 8, 12, 16], tc_compiler='eager', full_prefill_max_req=None), prefill=PhaseConfig(backend='disabled', max_bs=2048, bs=[4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], tc_compiler='eager', full_prefill_max_req=None)), cuda_graph_backend_decode=None, cuda_graph_backend_prefill=None, cuda_graph_max_bs_decode=16, cuda_graph_max_bs_prefill=None, cuda_graph_bs_decode=None, cuda_graph_bs_prefill=None, cuda_graph_tc_compiler=None, disable_prefill_cuda_graph=False, disable_decode_cuda_graph=False, disable_cuda_graph=False, disable_cuda_graph_padding=False, enable_profile_cuda_graph=False, enable_cudagraph_gc=False, debug_cuda_graph=False, enable_layerwise_nvtx_marker=False, enable_nccl_nvls=False, enable_symm_mem=False, triton_attention_reduce_in_fp32=False, triton_attention_num_kv_splits=8, triton_attention_split_tile_size=None, flashinfer_mla_disable_ragged=False, enable_fused_qk_norm_rope=False, enable_precise_embedding_interpolation=False, enable_fused_moe_sum_all_reduce=False, enable_deepseek_v4_fp4_indexer=False, disable_custom_all_reduce=False, enable_mscclpp=False, enable_torch_symm_mem=False, enable_scattered_sconv=False, pre_warm_nccl=False, enable_quant_communications=False, enable_flashinfer_allreduce_fusion=False, enforce_disable_flashinfer_allreduce_fusion=False, flashinfer_allreduce_fusion_backend=None, enable_aiter_allreduce_fusion=False, enable_torch_compile=False, enable_torch_compile_debug_mode=False, torch_compile_max_bs=32, torchao_config='', speculative_algorithm=None, speculative_draft_model_path=None, speculative_draft_model_revision=None, speculative_draft_load_format=None, speculative_num_steps=None, speculative_eagle_topk=None, speculative_num_draft_tokens=None, speculative_dflash_block_size=None, speculative_dspark_block_size=None, speculative_dspark_sps_table_path=None, speculative_dspark_confidence_sts_path=None, speculative_dspark_align_verify_tokens_to_graph_tier=False, speculative_accept_threshold_single=1.0, speculative_accept_threshold_acc=1.0, speculative_use_rejection_sampling=False, speculative_token_map=None, speculative_attention_mode='prefill', speculative_draft_attention_backend=None, speculative_draft_window_size=None, speculative_moe_runner_backend='flashinfer_mxfp4', speculative_moe_a2a_backend=None, speculative_draft_model_quantization=None, speculative_skip_dp_mlp_sync=False, enable_multi_layer_eagle=False, speculative_adaptive=False, speculative_adaptive_config=None, decoupled_spec_bind_endpoint=None, decoupled_spec_connect_endpoints=None, decoupled_spec_rank=None, decoupled_spec_role='null', spec_trace_dir=None, speculative_ngram_min_bfs_breadth=1, speculative_ngram_max_bfs_breadth=10, speculative_ngram_match_type='BFS', speculative_ngram_max_trie_depth=18, speculative_ngram_capacity=10000000, speculative_ngram_external_corpus_path=None, speculative_ngram_external_sam_budget=0, speculative_ngram_external_corpus_max_tokens=10000000, ep_size=4, moe_a2a_backend='none', moe_runner_backend='flashinfer_mxfp4', flashinfer_mxfp4_moe_precision='default', deepep_mode='auto', fuseep_mode=2, deepep_dispatcher_output_dtype='auto', ep_num_redundant_experts=0, ep_dispatch_algorithm=None, init_expert_location='trivial', enable_eplb=False, eplb_algorithm='auto', eplb_rebalance_num_iterations=1000, eplb_rebalance_layers_per_chunk=None, eplb_min_rebalancing_utilization_threshold=1.0, expert_distribution_recorder_mode=None, expert_distribution_recorder_buffer_size=1000, enable_expert_distribution_metrics=False, deepep_config=None, moe_dense_tp_size=None, elastic_ep_backend=None, enable_elastic_expert_backup=False, mooncake_ib_device=None, enable_waterfill=False, ep_join_mode=None, ep_join_rank_offset=0, elastic_ep_initial_size=None, max_ep_size=None, elastic_ep_scale_timeout=600, elastic_ep_rejoin=False, disable_flashinfer_cutlass_moe_fp4_allgather=False, disable_shared_experts_fusion=False, enforce_shared_experts_fusion=False, max_mamba_cache_size=None, mamba_ssm_dtype=None, mamba_max_states_per_path=-1, enable_mamba_cache_stochastic_rounding=False, mamba_cache_philox_rounds=0, mamba_full_memory_ratio=0.36, mamba_radix_cache_strategy='extra_buffer_lazy', uses_mamba_radix_cache=False, mamba_track_interval=256, enable_int8_mamba_checkpoint=False, int8_mamba_ckpt_size=None, linear_attn_backend='triton', linear_attn_decode_backend=None, linear_attn_prefill_backend=None, linear_attn_verify_backend=None, enable_linear_replayssm=False, linear_replayssm_cache_len=16, enable_linear_replayssm_spec=False, enable_hierarchical_cache=False, hicache_ratio=2.0, hicache_size=0, hicache_write_policy='write_through', hicache_io_backend='kernel', hicache_mem_layout='page_first', hicache_storage_backend=None, hicache_storage_prefetch_policy='timeout', hicache_storage_backend_extra_config=None, enable_hisparse=False, hisparse_config=None, enable_broadcast_mm_inputs_process=False, enable_prefix_mm_cache=False, mm_enable_dp_encoder=False, mm_process_config={}, mm_processor_worker_num=0, mm_io_worker_num=0, limit_mm_data_per_request=None, enable_mm_global_cache=False, disable_fast_image_processor=False, mm_feature_transport='cpu', keep_mm_feature_on_device=False, enable_lora=None, enable_lora_overlap_loading=None, max_lora_rank=None, lora_target_modules=None, lora_paths=None, max_loaded_loras=None, max_loras_per_batch=8, lora_eviction_policy='lru', lora_backend='csgmv', max_lora_chunk_size=16, experts_shared_outer_loras=None, lora_use_virtual_experts=False, lora_strict_loading=False, lora_drain_wait_threshold=0.0, enable_two_batch_overlap=False, enable_single_batch_overlap=False, tbo_token_distribution_threshold=0.48, cpu_offload_gb=0, offload_group_size=-1, offload_num_in_group=1, offload_prefetch_step=1, offload_mode='cpu', enable_lmcache=False, lmcache_config_file=None, enable_flexkv=False, flexkv_config_file=None, kt_weight_path=None, kt_method='AMXINT4', kt_cpuinfer=None, kt_threadpool_count=2, kt_num_gpu_experts=None, kt_max_deferred_experts_per_token=None, dllm_algorithm=None, dllm_algorithm_config=None, dllm_fdfo=True, disaggregation_mode='prefill', disaggregation_transfer_backend='mooncake', disaggregation_bootstrap_port=28800, disaggregation_ib_device='mlx5_0,mlx5_1,mlx5_2,mlx5_3', disaggregation_decode_enable_radix_cache=False, disaggregation_decode_enable_offload_kvcache=False, num_reserved_decode_tokens=512, disaggregation_decode_extra_slots=None, disaggregation_decode_polling_interval=1, optimistic_prefill_attempts=0, encoder_only=False, language_only=False, encoder_transfer_backend='zmq_to_tokenizer', encoder_urls=[], encoder_bootstrap_port=8997, encoder_register_urls=[], enable_adaptive_dispatch_to_encoder=False, enable_pdmux=False, pdmux_config_path=None, sm_group_num=8, custom_weight_loader=[], weight_loader_disable_mmap=False, weight_loader_prefetch_checkpoints=False, weight_loader_prefetch_num_threads=4, weight_loader_drop_cache_after_load=False, remote_instance_weight_loader_seed_instance_ip=None, remote_instance_weight_loader_seed_instance_service_port=None, remote_instance_weight_loader_send_weights_group_ports=None, remote_instance_weight_loader_backend='nccl', remote_instance_weight_loader_start_seed_via_transfer_engine=False, engine_info_bootstrap_port=6789, modelexpress_config=None, download_dir=None, model_checksum=None, delete_ckpt_after_loading=False, decrypted_config_file=None, decrypted_draft_config_file=None, checkpoint_engine_wait_weights_before_ready=False, enable_prefill_delayer=False, prefill_delayer_max_delay_passes=30, prefill_delayer_token_usage_low_watermark=None, prefill_delayer_forward_passes_buckets=None, prefill_delayer_wait_seconds_buckets=None, prefill_delayer_queue_min_ratio=None, prefill_delayer_max_delay_ms=None, min_free_slots_delay=None, enable_deterministic_inference=False, rl_on_policy_target=None, kv_canary='none', kv_canary_real_data='none', kv_canary_sweep_interval=0, enable_dynamic_batch_tokenizer=False, dynamic_batch_tokenizer_batch_size=32, dynamic_batch_tokenizer_batch_timeout=0.002, enable_tokenizer_batch_encode=False, disable_tokenizer_batch_decode=False, debug_tensor_dump_output_folder=None, debug_tensor_dump_layers=None, debug_tensor_dump_input_file=None, enable_memory_saver=False, enable_weights_cpu_backup=False, enable_draft_weights_cpu_backup=False, enable_custom_logit_processor=False, enable_return_hidden_states=False, enable_return_routed_experts=False, enable_return_indexer_topk=False, disable_outlines_disk_cache=False, enable_mis=False, weight_cache_mode='off', weight_cache_socket=None, weight_cache_timeout=1800, forward_hooks=None, msprobe_dump_config=None)
+[2026-08-27 04:28:45] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:45] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:45] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:45] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:45] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:45] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:45] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:45] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:01 PP4 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:01 PP4 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:01 PP5 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:01 PP5 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:01 PP4 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:01 PP4 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:01 PP5 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:01 PP5 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:01 PP4 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:01 PP4 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:01 PP4 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:01 PP4 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:02 PP5 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:02 PP5 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:02 PP5 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:02 PP5 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:03 PP5 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:03 PP5 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:03 PP4 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:03 PP4 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:03 PP5 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:03 PP5 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:03 PP4 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:03 PP4 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:05.195727   240 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:05.195828   240 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:05.195843   240 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.3 port: 12001
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:05.195847   236 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:05.195899   240 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.3:15652
+I0827 04:29:05.195935   236 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:05.195948   236 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.3 port: 12001
+I0827 04:29:05.195989   236 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.3:15032
+I0827 04:29:05.196039   240 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:05.196110   236 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:05.206050   241 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:05.206164   241 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:05.206182   241 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.3 port: 12001
+I0827 04:29:05.206233   241 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.3:16487
+I0827 04:29:05.206414   241 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:05.206894   238 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:05.207010   238 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:05.207029   238 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.3 port: 12001
+I0827 04:29:05.207082   238 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.3:15213
+I0827 04:29:05.207248   238 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:05.213657   238 topology.cpp:127] Device mlx5_0 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:05.215921   237 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:05.216045   237 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:05.216063   237 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.3 port: 12001
+I0827 04:29:05.216125   237 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.3:16450
+I0827 04:29:05.216308   237 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:05.219630   235 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:05.219745   235 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:05.219763   235 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.3 port: 12001
+I0827 04:29:05.219815   235 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.3:15658
+I0827 04:29:05.219810   236 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:05.220000   235 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:05.222787   237 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:05.223301   240 topology.cpp:127] Device mlx5_0 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:05.223796   239 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+W0827 04:29:05.223798   242 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:05.223948   242 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:05.223948   239 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:05.223969   242 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.3 port: 12001
+I0827 04:29:05.223973   239 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.3 port: 12001
+I0827 04:29:05.224030   242 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.3:16182
+I0827 04:29:05.224030   239 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.3:15581
+I0827 04:29:05.224237   242 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:05.224242   239 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:05.226227   235 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:05.227162   236 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:05.228888   237 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:05.229261   240 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:05.232141   235 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:05.232787   239 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:05.232861   242 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:05.233434   241 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:05.233938   236 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:05.236400   237 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:05.236641   240 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:05.238986   235 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:05.241036   236 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:05.241355   238 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:05.241638   241 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:05.242061   239 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:05.242151   242 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:05.244179   236 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:05.244223   236 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:05.244287   237 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:05.244297   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.244899   240 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:05.245494   235 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:05.249099   237 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:05.249130   237 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:05.249187   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.251125   235 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:05.251137   240 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:05.251168   235 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:05.251168   240 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:05.251241   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.251241   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.251624   239 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:05.251729   242 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:05.251823   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:05.253651   236 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:03
+I0827 04:29:05.253687   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.256857   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:05.257697   239 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:05.258617   237 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:03
+I0827 04:29:05.258652   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.258757   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:05.258822   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:05.260108   239 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:05.260136   239 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:05.260191   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.260488   235 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:03
+I0827 04:29:05.260502   240 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:03
+I0827 04:29:05.260530   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.260533   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.262187   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:05.264094   236 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:03
+I0827 04:29:05.264137   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.266639   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:05.268234   237 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:03
+I0827 04:29:05.268261   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.268549   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:05.269187   241 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:05.269232   238 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:05.270207   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:05.270285   240 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:03
+I0827 04:29:05.270316   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.271827   235 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:03
+I0827 04:29:05.271884   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.275274   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:05.276836   238 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:05.276907   242 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:05.276996   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:05.277058   236 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:03
+I0827 04:29:05.277101   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.278084   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:05.278632   237 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:03
+I0827 04:29:05.278667   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.279789   240 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:03
+I0827 04:29:05.279817   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.282152   238 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:05.282193   238 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:05.282260   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.283128   242 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:05.283200   242 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:05.283263   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.286593   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:05.287469   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:05.288283   239 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:03
+I0827 04:29:05.288322   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.289064   240 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:03
+I0827 04:29:05.289194   240 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:05 PP5 TP1 EP1] Init torch distributed begin.
+I0827 04:29:05.290841   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:05.292457   242 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:03
+I0827 04:29:05.292488   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.292999   241 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:05.296203   241 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:05.296272   241 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:05.296341   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.300206   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:05.301280   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:05.301775   242 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:03
+I0827 04:29:05.301801   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.303004   235 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:03
+I0827 04:29:05.303042   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.303999   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:05.305142   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:05.305728   241 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:03
+I0827 04:29:05.305768   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.305974   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:05.306758   237 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:03
+I0827 04:29:05.306921   237 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:05 PP4 TP2 EP2] Init torch distributed begin.
+I0827 04:29:05.307497   236 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:03
+I0827 04:29:05.307627   236 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:05 PP4 TP1 EP1] Init torch distributed begin.
+I0827 04:29:05.309624   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:05.309773   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:05.311504   238 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:03
+I0827 04:29:05.311542   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.311590   242 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:03
+I0827 04:29:05.311633   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.313076   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:05.314762   239 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:03
+I0827 04:29:05.314813   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.319522   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:05.319566   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:05.321249   242 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:03
+I0827 04:29:05.321398   242 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:05 PP5 TP3 EP3] Init torch distributed begin.
+I0827 04:29:05.326727   238 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:03
+I0827 04:29:05.326759   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.329339   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:05.331007   235 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:03
+I0827 04:29:05.331142   235 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:05 PP4 TP0 EP0] Init torch distributed begin.
+I0827 04:29:05.334666   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:05.336285   238 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:03
+I0827 04:29:05.336313   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.337296   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:05.338828   241 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:03
+I0827 04:29:05.338877   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.344906   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:05.346731   239 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:03
+I0827 04:29:05.346776   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.365299   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:05.366928   238 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:03
+I0827 04:29:05.367062   238 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:05 PP4 TP3 EP3] Init torch distributed begin.
+I0827 04:29:05.368464   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:05.369644   241 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:03
+I0827 04:29:05.369663   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.377357   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:05.379019   239 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:03
+I0827 04:29:05.379146   239 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:05 PP5 TP0 EP0] Init torch distributed begin.
+I0827 04:29:05.396831   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:05.398531   241 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:03
+I0827 04:29:05.398662   241 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:05 PP5 TP2 EP2] Init torch distributed begin.
+[2026-08-27 04:29:26 PP5 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:26 PP4 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:27 PP4 TP1 EP1] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP4 TP3 EP3] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP4 TP2 EP2] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP4 TP0 EP0] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP5 TP2 EP2] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP5 TP3 EP3] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP5 TP0 EP0] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP5 TP1 EP1] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP5 TP0 EP0] Init torch distributed ends. elapsed=22.37 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP5 TP3 EP3] Init torch distributed ends. elapsed=22.42 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP5 TP2 EP2] Init torch distributed ends. elapsed=22.35 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP4 TP1 EP1] Init torch distributed ends. elapsed=22.44 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP5 TP1 EP1] Init torch distributed ends. elapsed=22.46 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP4 TP3 EP3] Init torch distributed ends. elapsed=22.38 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP4 TP2 EP2] Init torch distributed ends. elapsed=22.44 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP4 TP0 EP0] Init torch distributed ends. elapsed=22.41 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP5 TP0 EP0] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP4 TP1 EP1] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP4 TP3 EP3] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP4 TP2 EP2] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP5 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP4 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP4 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP5 TP2 EP2] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP4 TP0 EP0] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP4 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP5 TP3 EP3] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP5 TP1 EP1] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP4 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP5 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP5 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP5 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP4 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP5 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP4 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP5 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP5 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP4 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP4 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP5 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP5 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP4 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP4 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP4 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP5 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP5 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP4 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP5 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:30:33 PP4 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:33 PP4 TP2 EP2] Load weight end. elapsed=65.50 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:34 PP4 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:34 PP4 TP3 EP3] Load weight end. elapsed=66.61 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP4 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:36 PP4 TP1 EP1] Load weight end. elapsed=68.45 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:36 PP5 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:36 PP5 TP0 EP0] Load weight end. elapsed=69.00 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:37 PP5 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:37 PP5 TP3 EP3] Load weight end. elapsed=69.63 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:37 PP5 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:37 PP5 TP1 EP1] Load weight end. elapsed=69.72 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:38 PP5 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:38 PP5 TP2 EP2] Load weight end. elapsed=70.08 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:38 PP5 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:38 PP5 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:38 PP5 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:38 PP5 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:38 PP5 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:38 PP5 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:38 PP5 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:38 PP5 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:47 PP4 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:47 PP4 TP0 EP0] Load weight end. elapsed=79.31 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:47 PP4 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:47 PP4 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:47 PP4 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:47 PP4 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:47 PP4 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:47 PP4 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:47 PP4 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:47 PP4 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:54 PP5 TP3 EP3] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP5 TP1 EP1] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP5 TP2 EP2] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP5 TP0 EP0] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP4 TP3 EP3] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP4 TP2 EP2] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP4 TP1 EP1] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP4 TP0 EP0] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP5 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP5 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP4 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP4 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP4 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP5 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP4 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP5 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP5 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP5 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP4 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP5 TP3 EP3] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP4 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP4 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP4 TP1 EP1] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP5 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP5 TP2 EP2] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP4 TP3 EP3] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP4 TP0 EP0] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP5 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP5 TP1 EP1] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP5 TP0 EP0] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP4 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP4 TP2 EP2] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP4 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:30:54 PP4 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:30:54 PP4 TP0 EP0] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp0_pp4_dp0.json
+[2026-08-27 04:30:54 PP4 TP1 EP1] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp1_pp4_dp0.json
+[2026-08-27 04:30:54 PP5 TP2 EP2] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp2_pp5_dp0.json
+[2026-08-27 04:30:54 PP5 TP3 EP3] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp3_pp5_dp0.json
+[2026-08-27 04:30:54 PP5 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:30:54 PP5 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:30:54 PP5 TP0 EP0] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp0_pp5_dp0.json
+[2026-08-27 04:30:54 PP4 TP2 EP2] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp2_pp4_dp0.json
+[2026-08-27 04:30:54 PP4 TP3 EP3] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp3_pp4_dp0.json
+[2026-08-27 04:30:54 PP5 TP1 EP1] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp1_pp5_dp0.json
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO[TensorRT-LLM][INFO] Set logger level to INFO
+
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.37profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.12profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.87profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.32profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.91profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.77profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.94profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.88profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.20profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.87profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  7.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 10.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.46profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.67profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.91profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.29profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.49profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  7.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.30profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.94profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.47profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.12profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.06profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.35profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  9.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:01<00:02,  2.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.89profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.58profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.32profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.70profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.20profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.27profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  6.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  7.03profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.56profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.73profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.88profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.37profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:05,  1.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:01<00:02,  2.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.05profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.19profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.77profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.18profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.25profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  3.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.18profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.20profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.37profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  9.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.27profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  8.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.43profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.89profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.35profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.96profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.90profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.30profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.72profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.18profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.33profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.51profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  2.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.03profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.68profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.16profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.01profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.24profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 10.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:02,  2.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.06profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.83profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.35profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.87profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.19profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  3.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.06profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.86profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 13.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 13.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.13profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:01,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:01,  3.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.47profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  6.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.47profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.57profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.22profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.17profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.50profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.71profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.19profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.55profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.90profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:05,  1.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:01<00:06,  1.00s/profile]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.33profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.19profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.03profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.76profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.28profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.32profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.15profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.79profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.31profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.18profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 11.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:01,  3.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.06profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:01,  3.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  6.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  3.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.05profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.60profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.50profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.92profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.00profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.29profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 11.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.09profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.43profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.11profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.91profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  7.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:02,  2.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.18profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.78profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.75profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.89profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.64profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.05profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.47profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.47profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.65profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:05,  1.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.85profile/s]
+[2026-08-27 04:31:28 PP5 TP1 EP1] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.21profile/s]
+[2026-08-27 04:31:28 PP5 TP0 EP0] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.74profile/s]
+[2026-08-27 04:31:28 PP5 TP2 EP2] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.77profile/s]
+[2026-08-27 04:31:28 PP5 TP3 EP3] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.19profile/s][2026-08-27 04:31:28 PP5 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:28 PP5 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:28 PP5 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:28 PP5 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:28 PP5 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:28 PP5 TP0 EP0] max_total_num_tokens=3801408, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=54, context_len=1048576, available_gpu_mem=16.30 GB
+[2026-08-27 04:31:28 PP5 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:28 PP5 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:28 PP5 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.87profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.90profile/s]
+[2026-08-27 04:31:29 PP4 TP3 EP3] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.87profile/s]
+[2026-08-27 04:31:29 PP4 TP2 EP2] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.93profile/s]
+[2026-08-27 04:31:29 PP4 TP0 EP0] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.43profile/s]
+[2026-08-27 04:31:29 PP4 TP1 EP1] FlashInfer autotune completed.
+[2026-08-27 04:31:29 PP4 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:29 PP4 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:29 PP4 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:29 PP4 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:29 PP4 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:29 PP4 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:29 PP4 TP0 EP0] max_total_num_tokens=3801408, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=54, context_len=1048576, available_gpu_mem=16.30 GB
+[2026-08-27 04:31:29 PP4 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:29 PP4 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+W0827 04:31:30.513926  2527 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50a93b80, len: 6912: Operation not permitted [1]
+W0827 04:31:30.514082  2528 memory_location.cpp:73] Failed to get NUMA node, addr: 0x507df700, len: 6912: Operation not permitted [1]
+W0827 04:31:30.514106  2529 memory_location.cpp:73] Failed to get NUMA node, addr: 0x508aa940, len: 6912: Operation not permitted [1]
+W0827 04:31:30.514214  2530 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50f67bc0, len: 6912: Operation not permitted [1]
+W0827 04:31:30.514272  2531 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50a2f680, len: 55296: Operation not permitted [1]
+W0827 04:31:30.514393  2532 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50afecc0, len: 55296: Operation not permitted [1]
+W0827 04:31:30.514506  2533 memory_location.cpp:73] Failed to get NUMA node, addr: 0x502be580, len: 6912: Operation not permitted [1]
+W0827 04:31:30.514607  2534 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5054d3c0, len: 13824: Operation not permitted [1]
+W0827 04:31:30.514652  2535 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50ad4580, len: 6912: Operation not permitted [1]
+W0827 04:31:30.514743  2536 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50eb9880, len: 6912: Operation not permitted [1]
+W0827 04:31:31.680995  2556 memory_location.cpp:73] Failed to get NUMA node, addr: 0x501c3300, len: 6912: Operation not permitted [1]
+W0827 04:31:31.680985  2557 memory_location.cpp:73] Failed to get NUMA node, addr: 0x504ecec0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.680985  2555 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd4bac0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681169  2558 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50706900, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681272  2559 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5006d700, len: 55296: Operation not permitted [1]
+W0827 04:31:31.681360  2560 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50041cc0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.681468  2561 memory_location.cpp:73] Failed to get NUMA node, addr: 0x506f5a40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681562  2562 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fdfdf40, len: 13824: Operation not permitted [1]
+W0827 04:31:31.681607  2563 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd16080, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681685  2564 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fcfa300, len: 6912: Operation not permitted [1]
+W0827 04:31:31.733497  2583 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51554040, len: 6912: Operation not permitted [1]
+W0827 04:31:31.733552  2584 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5110a8c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.733618  2585 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51ca8e80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.733795  2586 memory_location.cpp:73] Failed to get NUMA node, addr: 0x52017680, len: 6912: Operation not permitted [1]
+W0827 04:31:31.733909  2587 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5174af80, len: 55296: Operation not permitted [1]
+W0827 04:31:31.734004  2588 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51770440, len: 55296: Operation not permitted [1]
+W0827 04:31:31.734061  2589 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51118000, len: 6912: Operation not permitted [1]
+W0827 04:31:31.734179  2590 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51564f40, len: 13824: Operation not permitted [1]
+W0827 04:31:31.734277  2591 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51f69480, len: 6912: Operation not permitted [1]
+W0827 04:31:31.734400  2592 memory_location.cpp:73] Failed to get NUMA node, addr: 0x517c2940, len: 6912: Operation not permitted [1]
+W0827 04:31:31.741586  2611 memory_location.cpp:73] Failed to get NUMA node, addr: 0x519520c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.741609  2612 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50ff8c40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.741679  2613 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51a2ad80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.741792  2614 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51b5b300, len: 6912: Operation not permitted [1]
+W0827 04:31:31.741938  2615 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50fe5b80, len: 55296: Operation not permitted [1]
+W0827 04:31:31.742043  2616 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51625140, len: 55296: Operation not permitted [1]
+W0827 04:31:31.742168  2617 memory_location.cpp:73] Failed to get NUMA node, addr: 0x516bf640, len: 6912: Operation not permitted [1]
+W0827 04:31:31.742278  2618 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51b73e80, len: 13824: Operation not permitted [1]
+W0827 04:31:31.742414  2619 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51ef60c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.742429  2620 memory_location.cpp:73] Failed to get NUMA node, addr: 0x513e95c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.900663  2641 memory_location.cpp:73] Failed to get NUMA node, addr: 0x517e8c80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.900679  2640 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5170b840, len: 6912: Operation not permitted [1]
+W0827 04:31:31.900738  2642 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5131d0c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.900669  2639 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51372400, len: 6912: Operation not permitted [1]
+W0827 04:31:31.900923  2643 memory_location.cpp:73] Failed to get NUMA node, addr: 0x512b7e00, len: 55296: Operation not permitted [1]
+W0827 04:31:31.900976  2644 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51380200, len: 55296: Operation not permitted [1]
+W0827 04:31:31.901000  2646 memory_location.cpp:73] Failed to get NUMA node, addr: 0x513151c0, len: 13824: Operation not permitted [1]
+W0827 04:31:31.901011  2645 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51194f40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.901038  2647 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51520740, len: 6912: Operation not permitted [1]
+W0827 04:31:31.901073  2648 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51724980, len: 6912: Operation not permitted [1]
+W0827 04:31:31.931138  2668 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51061000, len: 6912: Operation not permitted [1]
+W0827 04:31:31.931178  2669 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50d74c00, len: 6912: Operation not permitted [1]
+W0827 04:31:31.931098  2667 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50b2fb40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.931272  2670 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50de9000, len: 6912: Operation not permitted [1]
+W0827 04:31:31.931349  2671 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5097dc80, len: 55296: Operation not permitted [1]
+W0827 04:31:31.931519  2672 memory_location.cpp:73] Failed to get NUMA node, addr: 0x515aa000, len: 55296: Operation not permitted [1]
+W0827 04:31:31.931557  2675 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51269700, len: 6912: Operation not permitted [1]
+W0827 04:31:31.931556  2673 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5097bd00, len: 6912: Operation not permitted [1]
+W0827 04:31:31.931576  2674 memory_location.cpp:73] Failed to get NUMA node, addr: 0x514ee800, len: 13824: Operation not permitted [1]
+W0827 04:31:31.931759  2676 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50deac80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.983201  2696 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50819380, len: 6912: Operation not permitted [1]
+W0827 04:31:31.983201  2695 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50600a80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.983378  2697 memory_location.cpp:73] Failed to get NUMA node, addr: 0x506ddf40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.983438  2698 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50716800, len: 6912: Operation not permitted [1]
+W0827 04:31:31.983490  2699 memory_location.cpp:73] Failed to get NUMA node, addr: 0x501575c0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.983624  2700 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50863340, len: 55296: Operation not permitted [1]
+W0827 04:31:31.983737  2701 memory_location.cpp:73] Failed to get NUMA node, addr: 0x508d7a40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.983930  2702 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50668c40, len: 13824: Operation not permitted [1]
+W0827 04:31:31.984035  2703 memory_location.cpp:73] Failed to get NUMA node, addr: 0x506841c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.984107  2704 memory_location.cpp:73] Failed to get NUMA node, addr: 0x508f16c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.001853  2725 memory_location.cpp:73] Failed to get NUMA node, addr: 0x507c8d40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.001852  2724 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50b39240, len: 6912: Operation not permitted [1]
+W0827 04:31:32.002048  2727 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50a8ecc0, len: 55296: Operation not permitted [1]
+W0827 04:31:32.001888  2726 memory_location.cpp:73] Failed to get NUMA node, addr: 0x512a8140, len: 6912: Operation not permitted [1]
+W0827 04:31:32.002044  2728 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50a68880, len: 55296: Operation not permitted [1]
+W0827 04:31:32.002257  2729 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5087b880, len: 6912: Operation not permitted [1]
+W0827 04:31:32.002264  2731 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50fc6980, len: 6912: Operation not permitted [1]
+W0827 04:31:32.002266  2730 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51331100, len: 13824: Operation not permitted [1]
+W0827 04:31:32.002470  2732 memory_location.cpp:73] Failed to get NUMA node, addr: 0x513384c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.002780  2723 memory_location.cpp:73] Failed to get NUMA node, addr: 0x512c6940, len: 6912: Operation not permitted [1]
+[2026-08-27 04:31:32] Dummy health check server started in background thread at 0.0.0.0:30000
+[rank16]:[W827 04:33:25.972222579 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank17]:[W827 04:33:25.980800100 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank19]:[W827 04:33:25.981413562 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank18]:[W827 04:33:26.644294178 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank16]:[W827 04:34:10.467297472 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank18]:[W827 04:34:10.467837454 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank17]:[W827 04:34:10.468151853 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank19]:[W827 04:34:10.468427857 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank22]:[W827 04:34:10.495396864 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank23]:[W827 04:34:10.495456053 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank21]:[W827 04:34:10.495795969 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank20]:[W827 04:34:10.495814123 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank22]:[W827 04:34:16.703081720 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank20]:[W827 04:34:16.703915664 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank21]:[W827 04:34:16.704033167 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank23]:[W827 04:34:16.704314444 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[2026-08-27 04:35:10 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.29
+[2026-08-27 04:35:10 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.29
+[2026-08-27 04:35:29 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.39
+[2026-08-27 04:35:29 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.39
+[2026-08-27 04:35:31 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25.89
+[2026-08-27 04:35:31 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25.91
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/ready_p_node3.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/ready_p_node3.log
new file mode 100644
index 0000000..be32943
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/ready_p_node3.log
@@ -0,0 +1,757 @@
+/usr/local/lib/python3.12/dist-packages/torchao/quantization/quant_api.py:1731: SyntaxWarning: invalid escape sequence '\.'
+  """Configuration class for applying different quantization configs to modules or parameters based on their fully qualified names (FQNs).
+/sgl-workspace/sglang/python/sglang/launch_server.py:56: UserWarning: 'python -m sglang.launch_server' is still supported, but 'sglang serve' is the recommended entrypoint.
+  Example: sglang serve --model-path <model> [options]
+  warnings.warn(
+[2026-08-27 04:28:46] Breakable CUDA graph is incompatible with MLA attention; disabling prefill CUDA graph.
+[2026-08-27 04:28:47] Attention backend not specified. Use triton backend by default.
+[2026-08-27 04:28:47] Pipeline parallelism is incompatible with overlap schedule.
+[2026-08-27 04:28:49] server_args=ServerArgs(model_path='/data/hf_models/Kimi-K3', tokenizer_path='/data/hf_models/Kimi-K3', tokenizer_mode='auto', tokenizer_backend='huggingface', tokenizer_worker_num=1, detokenizer_worker_num=1, skip_tokenizer_init=False, load_format='auto', model_loader_extra_config='{}', trust_remote_code=True, context_length=None, is_embedding=False, enable_multimodal=None, revision=None, model_impl='auto', model_config_parser='auto', json_model_override_args='{}', dtype='auto', quantization=None, quantization_param_path=None, kv_cache_dtype='auto', enable_fp32_lm_head=False, modelopt_quant=None, modelopt_checkpoint_restore_path=None, modelopt_checkpoint_save_path=None, modelopt_export_path=None, quantize_and_serve=False, rl_quant_profile=None, enable_tf32_matmul=False, mem_fraction_static=0.88, max_running_requests=None, max_queued_requests=None, max_total_tokens=None, chunked_prefill_size=8192, enable_dynamic_chunking=False, max_prefill_tokens=16384, prefill_max_requests=None, schedule_policy='fcfs', enable_priority_scheduling=False, disable_priority_preemption=False, default_priority_value=None, abort_on_priority_when_disabled=False, schedule_low_priority_values_first=False, priority_scheduling_preemption_threshold=10, retraction_policy='length', schedule_conservativeness=1.0, page_size=64, swa_full_tokens_ratio=0.8, disable_hybrid_swa_memory=False, radix_eviction_policy='lru', prefill_only_disable_kv_cache=False, disable_radix_cache=True, enable_page_major_kv_layout=False, enable_unified_memory=False, disable_chunked_prefix_cache=False, disable_overlap_schedule=True, num_continuous_decode_steps=1, scheduler_recv_interval=1, enable_mixed_chunk=False, nccl_port=None, dist_timeout=3600, dist_init_addr='174.1.60.1:20000', nnodes=4, node_rank=3, tp_size=4, dcp_size=1, pp_size=8, pp_max_micro_batch_size=None, pp_async_batch_depth=0, dp_size=1, load_balance_method='follow_bootstrap_room', attn_cp_size=1, moe_dp_size=1, dcp_comm_backend='ag_rs', dcp_replicate_q_proj=None, dwdp_size=1, enable_prefill_cp=False, cp_strategy=None, enable_dsa_cache_layer_split=False, enable_dsa_prefill_context_parallel=False, dsa_prefill_cp_mode='round-robin-split', enable_prefill_context_parallel=False, prefill_cp_mode='in-seq-split', enable_cp_decode_attn_tp=False, enable_dp_attention=False, enable_dp_attention_local_control_broadcast=False, enable_dp_lm_head=False, enable_attn_tp_input_scattered=False, disable_attn_tp_gather=False, enable_p2p_check=False, device='cuda', base_gpu_id=0, gpu_id_step=1, random_seed=927276655, watchdog_timeout=300, soft_watchdog_timeout=None, sleep_on_idle=False, use_ray=False, custom_sigquit_handler=None, numa_node=None, gc_threshold=None, host='0.0.0.0', port=30000, fastapi_root_path='', smg_grpc_mode=False, grpc_mode=False, grpc_port=None, sidecar=None, sidecar_args=None, skip_server_warmup=False, warmups=None, enable_http2=False, ssl_keyfile=None, ssl_certfile=None, ssl_ca_certs=None, ssl_keyfile_password=None, enable_ssl_refresh=False, api_key=None, admin_api_key=None, served_model_name='kimi-k3', weight_version='default', chat_template=None, hf_chat_template_name=None, completion_template=None, file_storage_path='sglang_storage', enable_cache_report=False, reasoning_parser=None, default_chat_template_kwargs=None, strip_thinking_cache=False, enable_strict_thinking=False, tool_call_parser=None, tool_server=None, sampling_defaults='model', asr_max_buffer_seconds=60, asr_max_concurrent_sessions=32, preferred_sampling_params=None, allow_auto_truncate=False, stream_interval=1, batch_notify_size=16, stream_response_default_include_usage=False, incremental_streaming_output=False, enable_streaming_session=False, enable_session_radix_cache=False, log_level='info', log_level_http=None, log_requests=False, log_requests_level=2, log_requests_format='text', log_requests_target=None, uvicorn_access_log_exclude_prefixes=[], crash_dump_folder=None, show_time_cost=False, enable_metrics=False, smg_http_sidecar_port=None, enable_mfu_metrics=False, enable_metrics_for_all_schedulers=False, load_snapshot_publish_interval=15, tokenizer_metrics_custom_labels_header='x-custom-labels', tokenizer_metrics_allowed_custom_labels=None, extra_metric_labels=None, bucket_time_to_first_token=None, bucket_inter_token_latency=None, bucket_e2e_request_latency=None, prompt_tokens_buckets=None, generation_tokens_buckets=None, gc_warning_threshold_secs=0.0, decode_log_interval=40, enable_request_time_stats_logging=False, kv_events_config=None, enable_forward_pass_metrics=False, forward_pass_metrics_worker_id='', forward_pass_metrics_ipc_name=None, enable_trace=False, trace_modules='request', otlp_traces_endpoint='localhost:4317', export_metrics_to_file=False, export_metrics_to_file_dir=None, stat_loggers=None, constrained_json_whitespace_pattern=None, constrained_json_disable_any_whitespace=False, attention_backend='triton', decode_attention_backend=None, prefill_attention_backend=None, sampling_backend='flashinfer', grammar_backend='xgrammar', radix_cache_backend=None, mm_attention_backend=None, fp8_gemm_runner_backend='auto', fp4_gemm_runner_backend='auto', bf16_gemm_backend='auto', dsa_prefill_backend=None, dsa_decode_backend=None, dsa_paged_mqa_logits_backend='auto', dsa_topk_backend='sgl-kernel', disable_flashinfer_autotune=False, flashinfer_autotune_skip_ops=None, mamba_backend='triton', cuda_graph_config=CudaGraphConfig(decode=PhaseConfig(backend='disabled', max_bs=16, bs=[1, 2, 4, 8, 12, 16], tc_compiler='eager', full_prefill_max_req=None), prefill=PhaseConfig(backend='disabled', max_bs=2048, bs=[4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], tc_compiler='eager', full_prefill_max_req=None)), cuda_graph_backend_decode=None, cuda_graph_backend_prefill=None, cuda_graph_max_bs_decode=16, cuda_graph_max_bs_prefill=None, cuda_graph_bs_decode=None, cuda_graph_bs_prefill=None, cuda_graph_tc_compiler=None, disable_prefill_cuda_graph=False, disable_decode_cuda_graph=False, disable_cuda_graph=False, disable_cuda_graph_padding=False, enable_profile_cuda_graph=False, enable_cudagraph_gc=False, debug_cuda_graph=False, enable_layerwise_nvtx_marker=False, enable_nccl_nvls=False, enable_symm_mem=False, triton_attention_reduce_in_fp32=False, triton_attention_num_kv_splits=8, triton_attention_split_tile_size=None, flashinfer_mla_disable_ragged=False, enable_fused_qk_norm_rope=False, enable_precise_embedding_interpolation=False, enable_fused_moe_sum_all_reduce=False, enable_deepseek_v4_fp4_indexer=False, disable_custom_all_reduce=False, enable_mscclpp=False, enable_torch_symm_mem=False, enable_scattered_sconv=False, pre_warm_nccl=False, enable_quant_communications=False, enable_flashinfer_allreduce_fusion=False, enforce_disable_flashinfer_allreduce_fusion=False, flashinfer_allreduce_fusion_backend=None, enable_aiter_allreduce_fusion=False, enable_torch_compile=False, enable_torch_compile_debug_mode=False, torch_compile_max_bs=32, torchao_config='', speculative_algorithm=None, speculative_draft_model_path=None, speculative_draft_model_revision=None, speculative_draft_load_format=None, speculative_num_steps=None, speculative_eagle_topk=None, speculative_num_draft_tokens=None, speculative_dflash_block_size=None, speculative_dspark_block_size=None, speculative_dspark_sps_table_path=None, speculative_dspark_confidence_sts_path=None, speculative_dspark_align_verify_tokens_to_graph_tier=False, speculative_accept_threshold_single=1.0, speculative_accept_threshold_acc=1.0, speculative_use_rejection_sampling=False, speculative_token_map=None, speculative_attention_mode='prefill', speculative_draft_attention_backend=None, speculative_draft_window_size=None, speculative_moe_runner_backend='flashinfer_mxfp4', speculative_moe_a2a_backend=None, speculative_draft_model_quantization=None, speculative_skip_dp_mlp_sync=False, enable_multi_layer_eagle=False, speculative_adaptive=False, speculative_adaptive_config=None, decoupled_spec_bind_endpoint=None, decoupled_spec_connect_endpoints=None, decoupled_spec_rank=None, decoupled_spec_role='null', spec_trace_dir=None, speculative_ngram_min_bfs_breadth=1, speculative_ngram_max_bfs_breadth=10, speculative_ngram_match_type='BFS', speculative_ngram_max_trie_depth=18, speculative_ngram_capacity=10000000, speculative_ngram_external_corpus_path=None, speculative_ngram_external_sam_budget=0, speculative_ngram_external_corpus_max_tokens=10000000, ep_size=4, moe_a2a_backend='none', moe_runner_backend='flashinfer_mxfp4', flashinfer_mxfp4_moe_precision='default', deepep_mode='auto', fuseep_mode=2, deepep_dispatcher_output_dtype='auto', ep_num_redundant_experts=0, ep_dispatch_algorithm=None, init_expert_location='trivial', enable_eplb=False, eplb_algorithm='auto', eplb_rebalance_num_iterations=1000, eplb_rebalance_layers_per_chunk=None, eplb_min_rebalancing_utilization_threshold=1.0, expert_distribution_recorder_mode=None, expert_distribution_recorder_buffer_size=1000, enable_expert_distribution_metrics=False, deepep_config=None, moe_dense_tp_size=None, elastic_ep_backend=None, enable_elastic_expert_backup=False, mooncake_ib_device=None, enable_waterfill=False, ep_join_mode=None, ep_join_rank_offset=0, elastic_ep_initial_size=None, max_ep_size=None, elastic_ep_scale_timeout=600, elastic_ep_rejoin=False, disable_flashinfer_cutlass_moe_fp4_allgather=False, disable_shared_experts_fusion=False, enforce_shared_experts_fusion=False, max_mamba_cache_size=None, mamba_ssm_dtype=None, mamba_max_states_per_path=-1, enable_mamba_cache_stochastic_rounding=False, mamba_cache_philox_rounds=0, mamba_full_memory_ratio=0.36, mamba_radix_cache_strategy='extra_buffer_lazy', uses_mamba_radix_cache=False, mamba_track_interval=256, enable_int8_mamba_checkpoint=False, int8_mamba_ckpt_size=None, linear_attn_backend='triton', linear_attn_decode_backend=None, linear_attn_prefill_backend=None, linear_attn_verify_backend=None, enable_linear_replayssm=False, linear_replayssm_cache_len=16, enable_linear_replayssm_spec=False, enable_hierarchical_cache=False, hicache_ratio=2.0, hicache_size=0, hicache_write_policy='write_through', hicache_io_backend='kernel', hicache_mem_layout='page_first', hicache_storage_backend=None, hicache_storage_prefetch_policy='timeout', hicache_storage_backend_extra_config=None, enable_hisparse=False, hisparse_config=None, enable_broadcast_mm_inputs_process=False, enable_prefix_mm_cache=False, mm_enable_dp_encoder=False, mm_process_config={}, mm_processor_worker_num=0, mm_io_worker_num=0, limit_mm_data_per_request=None, enable_mm_global_cache=False, disable_fast_image_processor=False, mm_feature_transport='cpu', keep_mm_feature_on_device=False, enable_lora=None, enable_lora_overlap_loading=None, max_lora_rank=None, lora_target_modules=None, lora_paths=None, max_loaded_loras=None, max_loras_per_batch=8, lora_eviction_policy='lru', lora_backend='csgmv', max_lora_chunk_size=16, experts_shared_outer_loras=None, lora_use_virtual_experts=False, lora_strict_loading=False, lora_drain_wait_threshold=0.0, enable_two_batch_overlap=False, enable_single_batch_overlap=False, tbo_token_distribution_threshold=0.48, cpu_offload_gb=0, offload_group_size=-1, offload_num_in_group=1, offload_prefetch_step=1, offload_mode='cpu', enable_lmcache=False, lmcache_config_file=None, enable_flexkv=False, flexkv_config_file=None, kt_weight_path=None, kt_method='AMXINT4', kt_cpuinfer=None, kt_threadpool_count=2, kt_num_gpu_experts=None, kt_max_deferred_experts_per_token=None, dllm_algorithm=None, dllm_algorithm_config=None, dllm_fdfo=True, disaggregation_mode='prefill', disaggregation_transfer_backend='mooncake', disaggregation_bootstrap_port=28800, disaggregation_ib_device='mlx5_0,mlx5_1,mlx5_2,mlx5_3', disaggregation_decode_enable_radix_cache=False, disaggregation_decode_enable_offload_kvcache=False, num_reserved_decode_tokens=512, disaggregation_decode_extra_slots=None, disaggregation_decode_polling_interval=1, optimistic_prefill_attempts=0, encoder_only=False, language_only=False, encoder_transfer_backend='zmq_to_tokenizer', encoder_urls=[], encoder_bootstrap_port=8997, encoder_register_urls=[], enable_adaptive_dispatch_to_encoder=False, enable_pdmux=False, pdmux_config_path=None, sm_group_num=8, custom_weight_loader=[], weight_loader_disable_mmap=False, weight_loader_prefetch_checkpoints=False, weight_loader_prefetch_num_threads=4, weight_loader_drop_cache_after_load=False, remote_instance_weight_loader_seed_instance_ip=None, remote_instance_weight_loader_seed_instance_service_port=None, remote_instance_weight_loader_send_weights_group_ports=None, remote_instance_weight_loader_backend='nccl', remote_instance_weight_loader_start_seed_via_transfer_engine=False, engine_info_bootstrap_port=6789, modelexpress_config=None, download_dir=None, model_checksum=None, delete_ckpt_after_loading=False, decrypted_config_file=None, decrypted_draft_config_file=None, checkpoint_engine_wait_weights_before_ready=False, enable_prefill_delayer=False, prefill_delayer_max_delay_passes=30, prefill_delayer_token_usage_low_watermark=None, prefill_delayer_forward_passes_buckets=None, prefill_delayer_wait_seconds_buckets=None, prefill_delayer_queue_min_ratio=None, prefill_delayer_max_delay_ms=None, min_free_slots_delay=None, enable_deterministic_inference=False, rl_on_policy_target=None, kv_canary='none', kv_canary_real_data='none', kv_canary_sweep_interval=0, enable_dynamic_batch_tokenizer=False, dynamic_batch_tokenizer_batch_size=32, dynamic_batch_tokenizer_batch_timeout=0.002, enable_tokenizer_batch_encode=False, disable_tokenizer_batch_decode=False, debug_tensor_dump_output_folder=None, debug_tensor_dump_layers=None, debug_tensor_dump_input_file=None, enable_memory_saver=False, enable_weights_cpu_backup=False, enable_draft_weights_cpu_backup=False, enable_custom_logit_processor=False, enable_return_hidden_states=False, enable_return_routed_experts=False, enable_return_indexer_topk=False, disable_outlines_disk_cache=False, enable_mis=False, weight_cache_mode='off', weight_cache_socket=None, weight_cache_timeout=1800, forward_hooks=None, msprobe_dump_config=None)
+[2026-08-27 04:28:49] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:49] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:49] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:49] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:49] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:49] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:49] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:49] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:05 PP6 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:05 PP6 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:07 PP6 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:07 PP6 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:07 PP6 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:07 PP7 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:07 PP7 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:07 PP7 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:07 PP7 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:07 PP6 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:07 PP6 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:07.662256   238 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:07.662379   238 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:07.662392   238 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.4 port: 12001
+I0827 04:29:07.662429   238 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.4:15914
+I0827 04:29:07.662514   238 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:07.685189   238 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:07.689950   238 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:07.695539   238 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:07.701294   238 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:07.703691   238 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:07.703715   238 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:07.703760   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:07.710492   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:07.711953   238 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:04
+I0827 04:29:07.711989   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:07.735716   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:07.737195   238 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:04
+I0827 04:29:07.737223   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:07.763653   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:07.765146   238 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:04
+I0827 04:29:07.765172   238 rdma_context.cpp:205] Using SIEVE endpoint store
+[2026-08-27 04:29:07 PP7 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:07 PP7 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:07 PP7 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:07 PP7 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+I0827 04:29:07.791589   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:07.793025   238 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:04
+I0827 04:29:07.793139   238 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:07 PP6 TP3 EP3] Init torch distributed begin.
+[2026-08-27 04:29:07 PP6 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:07 PP6 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:08 PP6 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:08.916008   235 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:08.916072   235 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:08.916082   235 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.4 port: 12001
+I0827 04:29:08.916121   235 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.4:15737
+I0827 04:29:08.916215   235 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:08.941840   235 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:08.950320   235 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:08.979278   235 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:08.988204   235 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:08.991855   235 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:08.991879   235 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:08.991928   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:08.999985   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:09.001416   235 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:04
+I0827 04:29:09.001446   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:09.027477   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:09.028954   235 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:04
+I0827 04:29:09.028980   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:09.055524   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:09.056981   235 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:04
+I0827 04:29:09.057020   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:09.083544   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:09.084911   235 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:04
+I0827 04:29:09.085031   235 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:09 PP6 TP0 EP0] Init torch distributed begin.
+[2026-08-27 04:29:09 PP7 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:09 PP6 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:09 PP7 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:09 PP6 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:09 PP7 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:09 PP7 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:10.000486   242 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:10.000609   242 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:10.000628   242 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.4 port: 12001
+I0827 04:29:10.000689   242 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.4:16890
+I0827 04:29:10.000851   242 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:10.006592   236 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:10.006711   236 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:10.006728   236 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.4 port: 12001
+I0827 04:29:10.006786   236 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.4:15893
+I0827 04:29:10.006963   236 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:10.022238   241 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:10.022354   241 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:10.022370   241 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.4 port: 12001
+I0827 04:29:10.022423   241 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.4:16277
+I0827 04:29:10.022593   241 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:10.032052   242 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:10.032101   236 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:10.039772   236 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:10.039858   242 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:10.047292   236 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:10.047341   242 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:10.047785   241 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:10.054806   242 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:10.054838   236 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:10.055280   241 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:10.059140   236 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:10.059171   236 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:10.059175   242 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:10.059201   242 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:10.059227   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.059260   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.062610   241 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:10.066787   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:10.066826   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:10.068500   242 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:04
+I0827 04:29:10.068532   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.068558   236 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:04
+I0827 04:29:10.068590   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.069866   241 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:10.072566   241 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:10.072592   241 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:10.072649   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.080149   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:10.081809   241 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:04
+I0827 04:29:10.081843   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.095701   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:10.097368   236 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:04
+I0827 04:29:10.097400   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.099974   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:10.101672   242 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:04
+I0827 04:29:10.101717   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.107705   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:10.109316   241 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:04
+I0827 04:29:10.109347   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.123868   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:10.125504   236 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:04
+I0827 04:29:10.125536   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.127957   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:10.129673   242 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:04
+I0827 04:29:10.129703   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.139813   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:10.141404   241 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:04
+I0827 04:29:10.141431   241 rdma_context.cpp:205] Using SIEVE endpoint store
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:10.150050   237 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:10.150159   237 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:10.150173   237 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.4 port: 12001
+I0827 04:29:10.150219   237 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.4:15413
+I0827 04:29:10.150358   237 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:10.151880   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:10.153475   236 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:04
+I0827 04:29:10.153601   236 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:10 PP6 TP1 EP1] Init torch distributed begin.
+I0827 04:29:10.155849   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:10.155864   237 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:10.157462   242 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:04
+I0827 04:29:10.157585   242 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:10 PP7 TP3 EP3] Init torch distributed begin.
+I0827 04:29:10.171923   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:10.172516   240 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:10.172667   240 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:10.172688   240 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.4 port: 12001
+I0827 04:29:10.172762   240 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.4:15705
+I0827 04:29:10.172977   240 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:10.173573   241 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:04
+I0827 04:29:10.173692   241 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:10 PP7 TP2 EP2] Init torch distributed begin.
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:10.177717   239 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:10.177865   239 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:10.177886   239 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.4 port: 12001
+I0827 04:29:10.177956   239 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.4:16192
+I0827 04:29:10.178146   239 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:10.183638   237 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:10.187090   239 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:10.191001   237 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:10.195258   239 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:10.198271   237 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:10.200984   237 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:10.201025   237 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:10.201079   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.203014   239 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:10.203156   240 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:10.209589   240 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:10.210305   239 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:10.212793   239 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:10.212821   239 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:10.212878   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.216092   240 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:10.219946   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:10.221297   239 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:04
+I0827 04:29:10.221323   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.222187   240 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:10.224284   240 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:10.224304   240 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:10.224344   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.230640   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:10.231778   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:10.231937   240 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:04
+I0827 04:29:10.231962   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.233362   237 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:04
+I0827 04:29:10.233397   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.236478   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:10.237651   240 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:04
+I0827 04:29:10.237670   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.250197   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:10.251711   239 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:04
+I0827 04:29:10.251737   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.263743   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:10.265311   237 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:04
+I0827 04:29:10.265339   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.267139   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:10.268607   240 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:04
+I0827 04:29:10.268632   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.273075   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:10.274682   237 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:04
+I0827 04:29:10.274710   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.276602   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:10.278275   239 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:04
+I0827 04:29:10.278314   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.295709   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:10.297385   240 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:04
+I0827 04:29:10.297510   240 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:10 PP7 TP1 EP1] Init torch distributed begin.
+I0827 04:29:10.299893   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:10.301563   237 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:04
+I0827 04:29:10.301689   237 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:10 PP6 TP2 EP2] Init torch distributed begin.
+I0827 04:29:10.303737   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:10.305307   239 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:04
+I0827 04:29:10.305433   239 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:10 PP7 TP0 EP0] Init torch distributed begin.
+[2026-08-27 04:29:26 PP7 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:26 PP6 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:27 PP6 TP0 EP0] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP6 TP3 EP3] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP6 TP2 EP2] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP6 TP1 EP1] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP7 TP1 EP1] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP7 TP0 EP0] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP7 TP3 EP3] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP7 TP2 EP2] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP7 TP1 EP1] Init torch distributed ends. elapsed=17.45 s, mem usage=0.46 GB
+[2026-08-27 04:29:27 PP7 TP3 EP3] Init torch distributed ends. elapsed=17.59 s, mem usage=0.46 GB
+[2026-08-27 04:29:27 PP7 TP2 EP2] Init torch distributed ends. elapsed=17.57 s, mem usage=0.46 GB
+[2026-08-27 04:29:27 PP7 TP0 EP0] Init torch distributed ends. elapsed=17.44 s, mem usage=0.46 GB
+[2026-08-27 04:29:27 PP6 TP1 EP1] Init torch distributed ends. elapsed=17.59 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP6 TP3 EP3] Init torch distributed ends. elapsed=19.95 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP6 TP0 EP0] Init torch distributed ends. elapsed=18.66 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP6 TP2 EP2] Init torch distributed ends. elapsed=17.44 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP6 TP1 EP1] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP7 TP1 EP1] Load weight begin. avail mem=82.12 GB
+[2026-08-27 04:29:27 PP6 TP0 EP0] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP6 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP7 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP7 TP3 EP3] Load weight begin. avail mem=82.12 GB
+[2026-08-27 04:29:27 PP6 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP7 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP6 TP2 EP2] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP6 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP7 TP2 EP2] Load weight begin. avail mem=82.12 GB
+[2026-08-27 04:29:28 PP7 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP6 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP7 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP6 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP7 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP7 TP0 EP0] Load weight begin. avail mem=82.12 GB
+[2026-08-27 04:29:28 PP6 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP7 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP7 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP6 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP7 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP6 TP3 EP3] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP6 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP6 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP6 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP7 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP7 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP7 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP6 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP7 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP6 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:30:40 PP7 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:40 PP7 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:40 PP7 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:40 PP7 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:40 PP7 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:40 PP7 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:40 PP7 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:40 PP7 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:41 PP7 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:41 PP7 TP1 EP1] Load weight end. elapsed=73.17 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=31.91 GB, mem usage=50.21 GB.
+[2026-08-27 04:30:41 PP7 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:41 PP7 TP0 EP0] Load weight end. elapsed=73.80 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=31.91 GB, mem usage=50.21 GB.
+[2026-08-27 04:30:42 PP6 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:42 PP6 TP1 EP1] Load weight end. elapsed=74.47 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:45 PP6 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:45 PP6 TP2 EP2] Load weight end. elapsed=77.20 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:50 PP7 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:50 PP7 TP3 EP3] Load weight end. elapsed=82.79 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=31.91 GB, mem usage=50.21 GB.
+[2026-08-27 04:30:51 PP6 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:51 PP6 TP3 EP3] Load weight end. elapsed=83.19 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:51 PP7 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:51 PP7 TP2 EP2] Load weight end. elapsed=83.68 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=31.91 GB, mem usage=50.21 GB.
+[2026-08-27 04:30:51 PP7 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:51 PP7 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:52 PP7 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:52 PP7 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:52 PP7 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:52 PP7 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:52 PP7 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:52 PP7 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:53 PP6 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:53 PP6 TP0 EP0] Load weight end. elapsed=85.47 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:53 PP6 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:53 PP6 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:53 PP6 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:53 PP6 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:53 PP6 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:53 PP6 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:53 PP6 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:53 PP6 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:54 PP7 TP3 EP3] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP7 TP2 EP2] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP7 TP1 EP1] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP7 TP0 EP0] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP6 TP3 EP3] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP6 TP2 EP2] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP6 TP1 EP1] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP6 TP0 EP0] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP6 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP6 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP7 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:54 PP6 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP7 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:54 PP7 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:54 PP7 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:54 PP6 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP6 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP6 TP3 EP3] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP6 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP6 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP6 TP2 EP2] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP6 TP1 EP1] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP7 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 16.31 GB
+[2026-08-27 04:30:54 PP7 TP2 EP2] Memory pool end. avail mem=14.95 GB
+[2026-08-27 04:30:54 PP6 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP6 TP0 EP0] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP7 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 16.31 GB
+[2026-08-27 04:30:54 PP7 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 16.31 GB
+[2026-08-27 04:30:54 PP7 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 16.31 GB
+[2026-08-27 04:30:54 PP7 TP3 EP3] Memory pool end. avail mem=14.95 GB
+[2026-08-27 04:30:54 PP7 TP0 EP0] Memory pool end. avail mem=14.95 GB
+[2026-08-27 04:30:54 PP7 TP1 EP1] Memory pool end. avail mem=14.95 GB
+[2026-08-27 04:30:54 PP7 TP2 EP2] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp2_pp7_dp0.json
+[2026-08-27 04:30:54 PP6 TP3 EP3] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp3_pp6_dp0.json
+[2026-08-27 04:30:54 PP6 TP1 EP1] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp1_pp6_dp0.json
+[2026-08-27 04:30:54 PP7 TP3 EP3] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp3_pp7_dp0.json
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[2026-08-27 04:30:54 PP6 TP2 EP2] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp2_pp6_dp0.json
+[2026-08-27 04:30:54 PP6 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:30:54 PP6 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:30:54 PP6 TP0 EP0] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp0_pp6_dp0.json
+[2026-08-27 04:30:54 PP7 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:30:54 PP7 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:30:54 PP7 TP0 EP0] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp0_pp7_dp0.json
+[2026-08-27 04:30:54 PP7 TP1 EP1] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp1_pp7_dp0.json
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 11.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 13.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.76profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.62profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.66profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.72profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.68profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.55profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.12profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  6.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  7.16profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.92profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.15profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.59profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.56profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 19.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  8.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.67profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.40profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.88profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  8.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  6.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.83profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.50profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.32profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.92profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.77profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.94profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.57profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 12.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 12.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 13.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.36profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.66profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  6.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.72profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  6.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.22profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.65profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.13profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.03profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.46profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.59profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.16profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  6.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.96profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.76profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.67profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  8.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.27profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.88profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.01profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.57profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.12profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 11.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  2.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.68profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.90profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.43profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.92profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  8.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.66profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  6.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  7.13profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.32profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.71profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:05,  1.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.16profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.31profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.77profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.28profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.48profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.87profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.40profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 13.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.35profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.98profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.88profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.73profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  3.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.88profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.09profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.20profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.97profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.83profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.47profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.40profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.60profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  8.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  7.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.97profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.05profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.05profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  6.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.49profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.72profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.19profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 10.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.91profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.64profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.58profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  3.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.35profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  9.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 12.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.09profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:02,  2.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 12.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.15profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.76profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.49profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.03profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.79profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.96profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.47profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.65profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.55profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.03profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  9.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.63profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.98profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:05,  1.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:01,  3.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.50profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  6.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.94profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.13profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.78profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.53profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.35profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.56profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.36profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:01<00:06,  1.06s/profile]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  8.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.91profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.66profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  8.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  7.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:02,  2.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.83profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.01profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.66profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.13profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.59profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  8.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  8.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.71profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.29profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.97profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.88profile/s]
+[2026-08-27 04:31:27 PP6 TP3 EP3] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.86profile/s]
+[2026-08-27 04:31:27 PP6 TP1 EP1] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.24profile/s]
+[2026-08-27 04:31:27 PP6 TP2 EP2] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.33profile/s]
+[2026-08-27 04:31:27 PP6 TP0 EP0] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.56profile/s][2026-08-27 04:31:27 PP6 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:27 PP6 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:27 PP6 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:27 PP6 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:27 PP6 TP0 EP0] max_total_num_tokens=3801408, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=54, context_len=1048576, available_gpu_mem=16.52 GB
+[2026-08-27 04:31:27 PP6 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:27 PP6 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:27 PP6 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:27 PP6 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
+[2026-08-27 04:31:28 PP7 TP2 EP2] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.32profile/s]
+[2026-08-27 04:31:28 PP7 TP3 EP3] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.60profile/s]
+[2026-08-27 04:31:28 PP7 TP0 EP0] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.60profile/s]
+[2026-08-27 04:31:28 PP7 TP1 EP1] FlashInfer autotune completed.
+[2026-08-27 04:31:28 PP7 TP2 EP2] multimem all-gather disabled (no multicast for world_size=4)
+[2026-08-27 04:31:28 PP7 TP3 EP3] multimem all-gather disabled (no multicast for world_size=4)
+[2026-08-27 04:31:28 PP7 TP1 EP1] multimem all-gather disabled (no multicast for world_size=4)
+[2026-08-27 04:31:28 PP7 TP0 EP0] multimem all-gather disabled (no multicast for world_size=4)
+[2026-08-27 04:31:28 PP7 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:28 PP7 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:28 PP7 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:28 PP7 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:28 PP7 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:28 PP7 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:28 PP7 TP0 EP0] max_total_num_tokens=3801408, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=54, context_len=1048576, available_gpu_mem=12.14 GB
+[2026-08-27 04:31:28 PP7 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:28 PP7 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+W0827 04:31:31.959518  2596 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5efc8ec0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.959637  2598 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ee654c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.959640  2597 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5eda3c80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.959542  2595 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5efc6f80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.959741  2599 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5eeaf500, len: 55296: Operation not permitted [1]
+W0827 04:31:31.959794  2600 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ef4e380, len: 55296: Operation not permitted [1]
+W0827 04:31:31.959827  2601 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5f548500, len: 6912: Operation not permitted [1]
+W0827 04:31:31.959918  2602 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5eec3bc0, len: 13824: Operation not permitted [1]
+W0827 04:31:31.960085  2603 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5f01bcc0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.960222  2604 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5f4c8ac0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.980162  2623 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60ba35c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.980152  2624 memory_location.cpp:73] Failed to get NUMA node, addr: 0x609ccf40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.980242  2625 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60d38600, len: 6912: Operation not permitted [1]
+W0827 04:31:31.980361  2626 memory_location.cpp:73] Failed to get NUMA node, addr: 0x609546c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.980516  2628 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60c49a00, len: 55296: Operation not permitted [1]
+W0827 04:31:31.980537  2627 memory_location.cpp:73] Failed to get NUMA node, addr: 0x607c6800, len: 55296: Operation not permitted [1]
+W0827 04:31:31.980547  2629 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60d11800, len: 6912: Operation not permitted [1]
+W0827 04:31:31.980834  2630 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60a1b740, len: 13824: Operation not permitted [1]
+W0827 04:31:31.981053  2632 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60ce5000, len: 6912: Operation not permitted [1]
+W0827 04:31:31.981062  2631 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60aef300, len: 6912: Operation not permitted [1]
+W0827 04:31:32.117022  2651 memory_location.cpp:73] Failed to get NUMA node, addr: 0x74c79d40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.117132  2652 memory_location.cpp:73] Failed to get NUMA node, addr: 0x74f99700, len: 6912: Operation not permitted [1]
+W0827 04:31:32.117317  2653 memory_location.cpp:73] Failed to get NUMA node, addr: 0x7546c740, len: 6912: Operation not permitted [1]
+W0827 04:31:32.117331  2654 memory_location.cpp:73] Failed to get NUMA node, addr: 0x74f4f140, len: 6912: Operation not permitted [1]
+W0827 04:31:32.117414  2656 memory_location.cpp:73] Failed to get NUMA node, addr: 0x750037c0, len: 55296: Operation not permitted [1]
+W0827 04:31:32.117445  2655 memory_location.cpp:73] Failed to get NUMA node, addr: 0x74bf1200, len: 55296: Operation not permitted [1]
+W0827 04:31:32.117705  2657 memory_location.cpp:73] Failed to get NUMA node, addr: 0x74f80080, len: 6912: Operation not permitted [1]
+W0827 04:31:32.117738  2658 memory_location.cpp:73] Failed to get NUMA node, addr: 0x74fbd2c0, len: 13824: Operation not permitted [1]
+W0827 04:31:32.117790  2659 memory_location.cpp:73] Failed to get NUMA node, addr: 0x74fa5e40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.118012  2660 memory_location.cpp:73] Failed to get NUMA node, addr: 0x74c24880, len: 6912: Operation not permitted [1]
+W0827 04:31:32.137427  2677 memory_location.cpp:73] Failed to get NUMA node, addr: 0x685bea40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.137485  2678 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6865c240, len: 6912: Operation not permitted [1]
+W0827 04:31:32.137612  2680 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6832f280, len: 6912: Operation not permitted [1]
+W0827 04:31:32.137683  2679 memory_location.cpp:73] Failed to get NUMA node, addr: 0x68511ec0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.137791  2681 memory_location.cpp:73] Failed to get NUMA node, addr: 0x687ae280, len: 55296: Operation not permitted [1]
+W0827 04:31:32.137800  2682 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6863aec0, len: 55296: Operation not permitted [1]
+W0827 04:31:32.137928  2683 memory_location.cpp:73] Failed to get NUMA node, addr: 0x685f4540, len: 6912: Operation not permitted [1]
+W0827 04:31:32.138145  2685 memory_location.cpp:73] Failed to get NUMA node, addr: 0x68b70f80, len: 6912: Operation not permitted [1]
+W0827 04:31:32.138144  2684 memory_location.cpp:73] Failed to get NUMA node, addr: 0x686e87c0, len: 13824: Operation not permitted [1]
+W0827 04:31:32.138219  2686 memory_location.cpp:73] Failed to get NUMA node, addr: 0x685491c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.187127  2706 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8eaeff40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.187187  2707 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8eaa1740, len: 6912: Operation not permitted [1]
+W0827 04:31:32.187136  2705 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8e5f2600, len: 6912: Operation not permitted [1]
+W0827 04:31:32.187413  2709 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ec4eb00, len: 55296: Operation not permitted [1]
+W0827 04:31:32.187405  2708 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ea3b800, len: 6912: Operation not permitted [1]
+W0827 04:31:32.187651  2710 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ed92ec0, len: 55296: Operation not permitted [1]
+W0827 04:31:32.187799  2711 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8eb3fbc0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.188133  2712 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ed0e1c0, len: 13824: Operation not permitted [1]
+W0827 04:31:32.188231  2713 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8e9d20c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.188272  2714 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8f2c7a80, len: 6912: Operation not permitted [1]
+W0827 04:31:32.252096  2732 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72b6b180, len: 6912: Operation not permitted [1]
+W0827 04:31:32.252050  2731 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72be0600, len: 6912: Operation not permitted [1]
+W0827 04:31:32.252147  2733 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72716c00, len: 6912: Operation not permitted [1]
+W0827 04:31:32.252276  2734 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72739940, len: 6912: Operation not permitted [1]
+W0827 04:31:32.252379  2735 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72da8a80, len: 55296: Operation not permitted [1]
+W0827 04:31:32.252439  2736 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72d15800, len: 55296: Operation not permitted [1]
+W0827 04:31:32.252678  2737 memory_location.cpp:73] Failed to get NUMA node, addr: 0x7315db40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.252701  2738 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72e0e700, len: 13824: Operation not permitted [1]
+W0827 04:31:32.252745  2739 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72f5f480, len: 6912: Operation not permitted [1]
+W0827 04:31:32.252832  2740 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72b295c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.268508  2760 memory_location.cpp:73] Failed to get NUMA node, addr: 0x69eefc00, len: 6912: Operation not permitted [1]
+W0827 04:31:32.268532  2759 memory_location.cpp:73] Failed to get NUMA node, addr: 0x69ec8680, len: 6912: Operation not permitted [1]
+W0827 04:31:32.268577  2761 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6a02e280, len: 6912: Operation not permitted [1]
+W0827 04:31:32.268708  2762 memory_location.cpp:73] Failed to get NUMA node, addr: 0x69d75dc0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.268766  2763 memory_location.cpp:73] Failed to get NUMA node, addr: 0x69fc4880, len: 55296: Operation not permitted [1]
+W0827 04:31:32.268886  2764 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6a0c8540, len: 55296: Operation not permitted [1]
+W0827 04:31:32.268975  2765 memory_location.cpp:73] Failed to get NUMA node, addr: 0x69c75940, len: 6912: Operation not permitted [1]
+W0827 04:31:32.269125  2766 memory_location.cpp:73] Failed to get NUMA node, addr: 0x698ae1c0, len: 13824: Operation not permitted [1]
+W0827 04:31:32.269232  2767 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6a5457c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.269354  2768 memory_location.cpp:73] Failed to get NUMA node, addr: 0x69c8fe80, len: 6912: Operation not permitted [1]
+W0827 04:31:32.281724  2785 memory_location.cpp:73] Failed to get NUMA node, addr: 0x678e47c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.281751  2786 memory_location.cpp:73] Failed to get NUMA node, addr: 0x67a39b80, len: 6912: Operation not permitted [1]
+W0827 04:31:32.281852  2787 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6798d9c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.281994  2788 memory_location.cpp:73] Failed to get NUMA node, addr: 0x67981dc0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.282085  2789 memory_location.cpp:73] Failed to get NUMA node, addr: 0x67823300, len: 55296: Operation not permitted [1]
+W0827 04:31:32.282146  2790 memory_location.cpp:73] Failed to get NUMA node, addr: 0x67a3bb00, len: 55296: Operation not permitted [1]
+W0827 04:31:32.282214  2791 memory_location.cpp:73] Failed to get NUMA node, addr: 0x67cdad40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.282285  2792 memory_location.cpp:73] Failed to get NUMA node, addr: 0x68028e40, len: 13824: Operation not permitted [1]
+W0827 04:31:32.282544  2794 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6796cf40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.282545  2793 memory_location.cpp:73] Failed to get NUMA node, addr: 0x67fa6600, len: 6912: Operation not permitted [1]
+[2026-08-27 04:31:32] Dummy health check server started in background thread at 0.0.0.0:30000
+[rank27]:[W827 04:34:16.127713860 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank25]:[W827 04:34:16.128748517 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank24]:[W827 04:34:16.129831915 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank26]:[W827 04:34:16.136239904 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank26]:[W827 04:35:04.964905528 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank25]:[W827 04:35:04.966225881 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank27]:[W827 04:35:04.966239242 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank24]:[W827 04:35:04.966301664 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank31]:[W827 04:35:04.990701389 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank28]:[W827 04:35:04.991914270 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank29]:[W827 04:35:04.994546241 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank30]:[W827 04:35:04.999020079 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank31]:[W827 04:35:10.873647776 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank30]:[W827 04:35:10.873901815 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank28]:[W827 04:35:10.874227564 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank29]:[W827 04:35:10.885318244 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[2026-08-27 04:35:10 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.29
+[2026-08-27 04:35:10 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.29
+[2026-08-27 04:35:29 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.39
+[2026-08-27 04:35:29 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.39
+[2026-08-27 04:35:31 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25.93
+[2026-08-27 04:35:31 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25.95
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/ready_router.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/ready_router.log
new file mode 100644
index 0000000..41c1bfd
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/logs/ready_router.log
@@ -0,0 +1,7 @@
+MiniLB is only for debugging purposes, it only supports random policy!
+INFO:     174.1.60.1:51774 - "GET /health HTTP/1.1" 200 OK
+[MiniLB] Overriding policy to random
+INFO:     Started server process [1]
+INFO:     Waiting for application startup.
+INFO:     Application startup complete.
+INFO:     Uvicorn running on http://0.0.0.0:31000 (Press CTRL+C to quit)
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/metadata/pd_image_versions.txt b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/metadata/pd_image_versions.txt
new file mode 100644
index 0000000..e15c3c4
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/metadata/pd_image_versions.txt
@@ -0,0 +1,12 @@
+39a17c2aa63dfe853018fb3d3f535d25b2f2397a79a9aa525ab812e481bcf3a5  /sgl-workspace/sglang/python/sglang/srt/disaggregation/prefill.py
+e9927f508ad50eefddbd54c57b6a532c5dfb724be7c8912d84707261c3f5a829  /sgl-workspace/sglang/python/sglang/srt/disaggregation/decode.py
+392ecc2d56902b931637cf9fe6abe0fdd7d4296b8950e43eaeba84495bd9f2dc  /sgl-workspace/sglang/python/sglang/srt/disaggregation/utils.py
+85d1083561ad7cfa2cffbe9b0d96afde501e2e245660117a316d40313a373c66  /sgl-workspace/sglang/python/sglang/srt/managers/scheduler_pp_mixin.py
+99264aa174c121d771c7cb52f398faefc29a24aa672ed9da470d9e23e0f949bf  /sgl-workspace/sglang/python/sglang/srt/managers/scheduler.py
+flashinfer-cubin                         0.6.15.post1
+flashinfer-jit-cache                     0.6.15.post1+cu130
+flashinfer-python                        0.6.18
+mooncake-transfer-engine-cuda13          0.3.12.post1
+sglang                                   0.5.16              /sgl-workspace/sglang/python
+sglang-kernel                            0.4.5
+sglang-router                            0.3.2
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/metadata/preflight.tsv b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/metadata/preflight.tsv
new file mode 100644
index 0000000..bdc2c94
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/metadata/preflight.tsv
@@ -0,0 +1,9 @@
+node	image	model	rdma	gpu_processes	ports
+1	ok	ok	ok	0	free
+2	ok	ok	ok	0	free
+3	ok	ok	ok	0	free
+4	ok	ok	ok	0	free
+5	ok	ok	ok	0	free
+6	ok	ok	ok	0	free
+7	ok	ok	ok	0	free
+8	ok	ok	ok	0	free
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/metadata/status_ready.txt b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/metadata/status_ready.txt
new file mode 100644
index 0000000..cbb5a5c
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/metadata/status_ready.txt
@@ -0,0 +1,85 @@
+===== node 1 =====
+kimi3_pro6000_pd_pp8_standard_prefill_node0|local/sglang:kimi-k3-sm120-flashinfer-mxfp4-pd1|Up 14 minutes
+kimi3_pro6000_pd_pp8_standard_mc_master|local/sglang:kimi-k3-sm120-flashinfer-mxfp4-pd1|Up 14 minutes
+0, 57245 MiB, 0 %
+1, 57245 MiB, 0 %
+2, 57245 MiB, 0 %
+3, 57245 MiB, 0 %
+4, 64177 MiB, 0 %
+5, 64185 MiB, 0 %
+6, 64181 MiB, 0 %
+7, 64187 MiB, 0 %
+===== node 2 =====
+kimi3_pro6000_pd_pp8_standard_prefill_node1|local/sglang:kimi-k3-sm120-flashinfer-mxfp4-pd1|Up 14 minutes
+0, 64291 MiB, 0 %
+1, 64297 MiB, 0 %
+2, 64297 MiB, 0 %
+3, 64297 MiB, 0 %
+4, 68181 MiB, 0 %
+5, 68181 MiB, 0 %
+6, 68181 MiB, 0 %
+7, 68175 MiB, 0 %
+===== node 3 =====
+kimi3_pro6000_pd_pp8_standard_prefill_node2|local/sglang:kimi-k3-sm120-flashinfer-mxfp4-pd1|Up 14 minutes
+0, 68565 MiB, 0 %
+1, 68565 MiB, 0 %
+2, 68565 MiB, 0 %
+3, 68565 MiB, 0 %
+4, 68429 MiB, 0 %
+5, 68429 MiB, 0 %
+6, 68429 MiB, 0 %
+7, 68429 MiB, 0 %
+===== node 4 =====
+kimi3_pro6000_pd_pp8_standard_prefill_node3|local/sglang:kimi-k3-sm120-flashinfer-mxfp4-pd1|Up 14 minutes
+0, 68339 MiB, 0 %
+1, 68339 MiB, 0 %
+2, 68339 MiB, 0 %
+3, 68339 MiB, 0 %
+4, 72687 MiB, 0 %
+5, 72687 MiB, 0 %
+6, 72687 MiB, 0 %
+7, 72687 MiB, 0 %
+===== node 5 =====
+kimi3_pro6000_pd_pp8_standard_router|local/sglang:kimi-k3-sm120-flashinfer-mxfp4-pd1|Up 8 seconds
+kimi3_pro6000_pd_pp8_standard_decode_node0|local/sglang:kimi-k3-sm120-flashinfer-mxfp4-pd1|Up 7 minutes
+0, 76277 MiB, 0 %
+1, 76261 MiB, 0 %
+2, 76245 MiB, 0 %
+3, 76293 MiB, 0 %
+4, 76317 MiB, 0 %
+5, 76261 MiB, 0 %
+6, 76245 MiB, 0 %
+7, 76333 MiB, 0 %
+===== node 6 =====
+kimi3_pro6000_pd_pp8_standard_decode_node1|local/sglang:kimi-k3-sm120-flashinfer-mxfp4-pd1|Up 7 minutes
+0, 76317 MiB, 0 %
+1, 76261 MiB, 0 %
+2, 76245 MiB, 0 %
+3, 76333 MiB, 0 %
+4, 76317 MiB, 0 %
+5, 76261 MiB, 0 %
+6, 76245 MiB, 0 %
+7, 76333 MiB, 0 %
+===== node 7 =====
+kimi3_pro6000_pd_pp8_standard_decode_node2|local/sglang:kimi-k3-sm120-flashinfer-mxfp4-pd1|Up 7 minutes
+0, 76317 MiB, 0 %
+1, 76261 MiB, 0 %
+2, 76245 MiB, 0 %
+3, 76333 MiB, 0 %
+4, 76317 MiB, 0 %
+5, 76261 MiB, 0 %
+6, 76245 MiB, 0 %
+7, 76333 MiB, 0 %
+===== node 8 =====
+kimi3_pro6000_pd_pp8_standard_decode_node3|local/sglang:kimi-k3-sm120-flashinfer-mxfp4-pd1|Up 7 minutes
+0, 76317 MiB, 0 %
+1, 76261 MiB, 0 %
+2, 76245 MiB, 0 %
+3, 76333 MiB, 0 %
+4, 76277 MiB, 0 %
+5, 76261 MiB, 0 %
+6, 76245 MiB, 0 %
+7, 76293 MiB, 0 %
+p=healthy
+d=healthy
+router=healthy
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/orchestrator.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/orchestrator.log
new file mode 100644
index 0000000..8d182e9
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-standard-page64-20260827-1230/orchestrator.log
@@ -0,0 +1,44 @@
+[2026-08-27 12:27:56] preflight node 1
+[2026-08-27 12:27:57] preflight node 2
+[2026-08-27 12:27:59] preflight node 3
+[2026-08-27 12:28:01] preflight node 4
+[2026-08-27 12:28:03] preflight node 5
+[2026-08-27 12:28:04] preflight node 6
+[2026-08-27 12:28:07] preflight node 7
+[2026-08-27 12:28:09] preflight node 8
+[2026-08-27 12:28:15] preflight passed on all 8 nodes
+[2026-08-27 12:28:17] Mooncake master listening on 50051
+[2026-08-27 12:28:19] started p rank 1 on node 2
+[2026-08-27 12:28:21] started p rank 2 on node 3
+[2026-08-27 12:28:22] started p rank 3 on node 4
+[2026-08-27 12:28:30] started p rank 0 on node 1
+[2026-08-27 12:29:00] waiting for p, elapsed=30s
+[2026-08-27 12:29:39] waiting for p, elapsed=60s
+[2026-08-27 12:30:13] waiting for p, elapsed=90s
+[2026-08-27 12:30:52] waiting for p, elapsed=120s
+[2026-08-27 12:31:27] waiting for p, elapsed=150s
+[2026-08-27 12:32:06] waiting for p, elapsed=180s
+[2026-08-27 12:32:41] waiting for p, elapsed=210s
+[2026-08-27 12:33:20] waiting for p, elapsed=240s
+[2026-08-27 12:33:55] waiting for p, elapsed=270s
+[2026-08-27 12:34:34] waiting for p, elapsed=300s
+[2026-08-27 12:35:09] waiting for p, elapsed=330s
+[2026-08-27 12:35:32] p healthy after 333s
+[2026-08-27 12:35:38] started d rank 1 on node 6
+[2026-08-27 12:35:39] started d rank 2 on node 7
+[2026-08-27 12:35:41] started d rank 3 on node 8
+[2026-08-27 12:35:47] started d rank 0 on node 5
+[2026-08-27 12:36:17] waiting for d, elapsed=30s
+[2026-08-27 12:36:56] waiting for d, elapsed=60s
+[2026-08-27 12:37:31] waiting for d, elapsed=90s
+[2026-08-27 12:38:11] waiting for d, elapsed=120s
+[2026-08-27 12:38:46] waiting for d, elapsed=150s
+[2026-08-27 12:39:25] waiting for d, elapsed=180s
+[2026-08-27 12:40:00] waiting for d, elapsed=210s
+[2026-08-27 12:40:39] waiting for d, elapsed=240s
+[2026-08-27 12:41:14] waiting for d, elapsed=270s
+[2026-08-27 12:41:52] waiting for d, elapsed=300s
+[2026-08-27 12:42:27] waiting for d, elapsed=330s
+[2026-08-27 12:43:04] d healthy after 357s
+[2026-08-27 12:43:11] router healthy after 2s
+[2026-08-27 12:43:24] standard PD service ready at http://174.1.60.5:31000
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/bench/prefill_16k_1_c1.cmd.txt b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/bench/prefill_16k_1_c1.cmd.txt
new file mode 100644
index 0000000..ac2b0f6
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/bench/prefill_16k_1_c1.cmd.txt
@@ -0,0 +1 @@
+docker run --rm --network host -v /data/hf_models/Kimi-K3:/data/hf_models/Kimi-K3:ro -v /data/hzy/sskj/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310:/data/hzy/sskj/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310 --entrypoint python3 local/sglang:kimi-k3-sm120-flashinfer-mxfp4-pd1 -m sglang.benchmark.serving --backend sglang --host 174.1.60.5 --port 31000 --tokenizer /data/hf_models/Kimi-K3 --dataset-name random-ids --random-input-len 16384 --random-output-len 1 --random-range-ratio 1.0 --num-prompts 8 --max-concurrency 1 --request-rate 10000 --warmup-requests 0 --output-file /data/hzy/sskj/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/bench/prefill_16k_1_c1.jsonl --output-details --disable-tqdm 
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/bench/prefill_16k_1_c1.jsonl b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/bench/prefill_16k_1_c1.jsonl
new file mode 100644
index 0000000..46ef76e
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/bench/prefill_16k_1_c1.jsonl
@@ -0,0 +1 @@
+{"tag": null, "backend": "sglang", "dataset_name": "random-ids", "request_rate": 10000.0, "max_concurrency": 1, "sharegpt_output_len": null, "random_input_len": 16384, "random_output_len": 1, "random_range_ratio": 1.0, "server_info": {"internal_states": [{"model_path": "/data/hf_models/Kimi-K3", "tokenizer_path": "/data/hf_models/Kimi-K3", "tokenizer_mode": "auto", "tokenizer_backend": "huggingface", "tokenizer_worker_num": 1, "detokenizer_worker_num": 1, "skip_tokenizer_init": false, "load_format": "auto", "model_loader_extra_config": "{}", "trust_remote_code": true, "context_length": null, "is_embedding": false, "enable_multimodal": null, "revision": null, "model_impl": "auto", "model_config_parser": "auto", "json_model_override_args": "{}", "dtype": "auto", "quantization": null, "quantization_param_path": null, "kv_cache_dtype": "auto", "enable_fp32_lm_head": false, "modelopt_quant": null, "modelopt_checkpoint_restore_path": null, "modelopt_checkpoint_save_path": null, "modelopt_export_path": null, "quantize_and_serve": false, "rl_quant_profile": null, "enable_tf32_matmul": false, "mem_fraction_static": 0.88, "max_running_requests": null, "max_queued_requests": null, "max_total_tokens": null, "chunked_prefill_size": 8192, "enable_dynamic_chunking": false, "max_prefill_tokens": 16384, "prefill_max_requests": null, "schedule_policy": "fcfs", "enable_priority_scheduling": false, "disable_priority_preemption": false, "default_priority_value": null, "abort_on_priority_when_disabled": false, "schedule_low_priority_values_first": false, "priority_scheduling_preemption_threshold": 10, "retraction_policy": "length", "schedule_conservativeness": 1.0, "page_size": 64, "swa_full_tokens_ratio": 0.8, "disable_hybrid_swa_memory": false, "radix_eviction_policy": "lru", "prefill_only_disable_kv_cache": false, "disable_radix_cache": true, "enable_page_major_kv_layout": false, "enable_unified_memory": false, "disable_chunked_prefix_cache": true, "disable_overlap_schedule": false, "num_continuous_decode_steps": 1, "scheduler_recv_interval": 1, "enable_mixed_chunk": false, "nccl_port": null, "dist_timeout": 3600, "dist_init_addr": "174.1.60.5:20000", "nnodes": 4, "node_rank": 0, "tp_size": 32, "dcp_size": 1, "pp_size": 1, "pp_max_micro_batch_size": 181, "pp_async_batch_depth": 0, "dp_size": 1, "load_balance_method": "round_robin", "attn_cp_size": 1, "moe_dp_size": 1, "dcp_comm_backend": "ag_rs", "dcp_replicate_q_proj": null, "dwdp_size": 1, "enable_prefill_cp": false, "cp_strategy": null, "enable_dsa_cache_layer_split": false, "enable_dsa_prefill_context_parallel": false, "dsa_prefill_cp_mode": "round-robin-split", "enable_prefill_context_parallel": false, "prefill_cp_mode": "in-seq-split", "enable_cp_decode_attn_tp": false, "enable_dp_attention": false, "enable_dp_attention_local_control_broadcast": false, "enable_dp_lm_head": false, "enable_attn_tp_input_scattered": false, "disable_attn_tp_gather": false, "enable_p2p_check": false, "device": "cuda", "base_gpu_id": 0, "gpu_id_step": 1, "random_seed": 369235288, "watchdog_timeout": 300, "soft_watchdog_timeout": null, "sleep_on_idle": false, "use_ray": false, "numa_node": null, "gc_threshold": null, "host": "0.0.0.0", "port": 30000, "fastapi_root_path": "", "smg_grpc_mode": false, "grpc_mode": false, "grpc_port": null, "sidecar": null, "sidecar_args": null, "skip_server_warmup": false, "warmups": null, "enable_http2": false, "ssl_keyfile": null, "ssl_certfile": null, "ssl_ca_certs": null, "ssl_keyfile_password": null, "enable_ssl_refresh": false, "api_key": null, "admin_api_key": null, "served_model_name": "kimi-k3", "weight_version": "default", "chat_template": null, "hf_chat_template_name": null, "completion_template": null, "file_storage_path": "sglang_storage", "enable_cache_report": false, "reasoning_parser": null, "default_chat_template_kwargs": null, "strip_thinking_cache": false, "enable_strict_thinking": false, "tool_call_parser": null, "tool_server": null, "sampling_defaults": "model", "asr_max_buffer_seconds": 60, "asr_max_concurrent_sessions": 32, "preferred_sampling_params": null, "allow_auto_truncate": false, "stream_interval": 1, "batch_notify_size": 16, "stream_response_default_include_usage": false, "incremental_streaming_output": false, "enable_streaming_session": false, "enable_session_radix_cache": false, "log_level": "info", "log_level_http": null, "log_requests": false, "log_requests_level": 2, "log_requests_format": "text", "log_requests_target": null, "uvicorn_access_log_exclude_prefixes": [], "crash_dump_folder": null, "show_time_cost": false, "enable_metrics": false, "smg_http_sidecar_port": null, "enable_mfu_metrics": false, "enable_metrics_for_all_schedulers": false, "load_snapshot_publish_interval": 15, "tokenizer_metrics_custom_labels_header": "x-custom-labels", "tokenizer_metrics_allowed_custom_labels": null, "extra_metric_labels": null, "bucket_time_to_first_token": null, "bucket_inter_token_latency": null, "bucket_e2e_request_latency": null, "prompt_tokens_buckets": null, "generation_tokens_buckets": null, "gc_warning_threshold_secs": 0.0, "decode_log_interval": 40, "enable_request_time_stats_logging": false, "kv_events_config": null, "enable_forward_pass_metrics": false, "forward_pass_metrics_worker_id": "", "forward_pass_metrics_ipc_name": null, "enable_trace": false, "trace_modules": "request", "otlp_traces_endpoint": "localhost:4317", "export_metrics_to_file": false, "export_metrics_to_file_dir": null, "stat_loggers": null, "constrained_json_whitespace_pattern": null, "constrained_json_disable_any_whitespace": false, "attention_backend": "triton", "decode_attention_backend": null, "prefill_attention_backend": null, "sampling_backend": "flashinfer", "grammar_backend": "xgrammar", "radix_cache_backend": null, "mm_attention_backend": null, "fp8_gemm_runner_backend": "auto", "fp4_gemm_runner_backend": "auto", "bf16_gemm_backend": "auto", "dsa_prefill_backend": null, "dsa_decode_backend": null, "dsa_paged_mqa_logits_backend": "auto", "dsa_topk_backend": "sgl-kernel", "disable_flashinfer_autotune": false, "flashinfer_autotune_skip_ops": null, "mamba_backend": "triton", "cuda_graph_config": {"decode": {"backend": "full", "max_bs": 16, "bs": [1, 2, 4, 8, 12, 16], "tc_compiler": "eager", "full_prefill_max_req": null}, "prefill": {"backend": "disabled", "max_bs": 2048, "bs": [4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], "tc_compiler": "eager", "full_prefill_max_req": null}}, "cuda_graph_backend_decode": null, "cuda_graph_backend_prefill": null, "cuda_graph_max_bs_decode": 16, "cuda_graph_max_bs_prefill": null, "cuda_graph_bs_decode": null, "cuda_graph_bs_prefill": null, "cuda_graph_tc_compiler": null, "disable_prefill_cuda_graph": false, "disable_decode_cuda_graph": false, "disable_cuda_graph": false, "disable_cuda_graph_padding": false, "enable_profile_cuda_graph": false, "enable_cudagraph_gc": false, "debug_cuda_graph": false, "enable_layerwise_nvtx_marker": false, "enable_nccl_nvls": false, "enable_symm_mem": false, "triton_attention_reduce_in_fp32": false, "triton_attention_num_kv_splits": 8, "triton_attention_split_tile_size": null, "flashinfer_mla_disable_ragged": false, "enable_fused_qk_norm_rope": false, "enable_precise_embedding_interpolation": false, "enable_fused_moe_sum_all_reduce": false, "enable_deepseek_v4_fp4_indexer": false, "disable_custom_all_reduce": false, "enable_mscclpp": false, "enable_torch_symm_mem": false, "enable_scattered_sconv": false, "pre_warm_nccl": false, "enable_quant_communications": false, "enable_flashinfer_allreduce_fusion": false, "enforce_disable_flashinfer_allreduce_fusion": false, "flashinfer_allreduce_fusion_backend": null, "enable_aiter_allreduce_fusion": false, "enable_torch_compile": false, "enable_torch_compile_debug_mode": false, "torch_compile_max_bs": 32, "torchao_config": "", "speculative_algorithm": null, "speculative_draft_model_path": null, "speculative_draft_model_revision": null, "speculative_draft_load_format": null, "speculative_num_steps": null, "speculative_eagle_topk": null, "speculative_num_draft_tokens": null, "speculative_dflash_block_size": null, "speculative_dspark_block_size": null, "speculative_dspark_sps_table_path": null, "speculative_dspark_confidence_sts_path": null, "speculative_dspark_align_verify_tokens_to_graph_tier": false, "speculative_accept_threshold_single": 1.0, "speculative_accept_threshold_acc": 1.0, "speculative_use_rejection_sampling": false, "speculative_token_map": null, "speculative_attention_mode": "prefill", "speculative_draft_attention_backend": null, "speculative_draft_window_size": null, "speculative_moe_runner_backend": "marlin", "speculative_moe_a2a_backend": null, "speculative_draft_model_quantization": null, "speculative_skip_dp_mlp_sync": false, "enable_multi_layer_eagle": false, "speculative_adaptive": false, "speculative_adaptive_config": null, "decoupled_spec_bind_endpoint": null, "decoupled_spec_connect_endpoints": null, "decoupled_spec_rank": null, "decoupled_spec_role": "null", "spec_trace_dir": null, "speculative_ngram_min_bfs_breadth": 1, "speculative_ngram_max_bfs_breadth": 10, "speculative_ngram_match_type": "BFS", "speculative_ngram_max_trie_depth": 18, "speculative_ngram_capacity": 10000000, "speculative_ngram_external_corpus_path": null, "speculative_ngram_external_sam_budget": 0, "speculative_ngram_external_corpus_max_tokens": 10000000, "ep_size": 32, "moe_a2a_backend": "none", "moe_runner_backend": "marlin", "flashinfer_mxfp4_moe_precision": "default", "deepep_mode": "auto", "fuseep_mode": 2, "deepep_dispatcher_output_dtype": "auto", "ep_num_redundant_experts": 0, "ep_dispatch_algorithm": null, "init_expert_location": "trivial", "enable_eplb": false, "eplb_algorithm": "auto", "eplb_rebalance_num_iterations": 1000, "eplb_rebalance_layers_per_chunk": null, "eplb_min_rebalancing_utilization_threshold": 1.0, "expert_distribution_recorder_mode": null, "expert_distribution_recorder_buffer_size": 1000, "enable_expert_distribution_metrics": false, "deepep_config": null, "moe_dense_tp_size": null, "elastic_ep_backend": null, "enable_elastic_expert_backup": false, "mooncake_ib_device": null, "enable_waterfill": false, "ep_join_mode": null, "ep_join_rank_offset": 0, "elastic_ep_initial_size": null, "max_ep_size": null, "elastic_ep_scale_timeout": 600, "elastic_ep_rejoin": false, "disable_flashinfer_cutlass_moe_fp4_allgather": false, "disable_shared_experts_fusion": false, "enforce_shared_experts_fusion": false, "max_mamba_cache_size": 181, "mamba_ssm_dtype": null, "mamba_max_states_per_path": -1, "enable_mamba_cache_stochastic_rounding": false, "mamba_cache_philox_rounds": 0, "mamba_full_memory_ratio": 0.21, "mamba_radix_cache_strategy": "extra_buffer_lazy", "uses_mamba_radix_cache": false, "mamba_track_interval": 256, "enable_int8_mamba_checkpoint": false, "int8_mamba_ckpt_size": null, "linear_attn_backend": "triton", "linear_attn_decode_backend": null, "linear_attn_prefill_backend": null, "linear_attn_verify_backend": null, "enable_linear_replayssm": false, "linear_replayssm_cache_len": 16, "enable_linear_replayssm_spec": false, "enable_hierarchical_cache": false, "hicache_ratio": 2.0, "hicache_size": 0, "hicache_write_policy": "write_through", "hicache_io_backend": "kernel", "hicache_mem_layout": "page_first", "hicache_storage_backend": null, "hicache_storage_prefetch_policy": "timeout", "hicache_storage_backend_extra_config": null, "enable_hisparse": false, "hisparse_config": null, "enable_broadcast_mm_inputs_process": false, "enable_prefix_mm_cache": false, "mm_enable_dp_encoder": false, "mm_process_config": {}, "mm_processor_worker_num": 0, "mm_io_worker_num": 0, "limit_mm_data_per_request": null, "enable_mm_global_cache": false, "disable_fast_image_processor": false, "mm_feature_transport": "cpu", "keep_mm_feature_on_device": false, "enable_lora": null, "enable_lora_overlap_loading": null, "max_lora_rank": null, "lora_target_modules": null, "lora_paths": null, "max_loaded_loras": null, "max_loras_per_batch": 8, "lora_eviction_policy": "lru", "lora_backend": "csgmv", "max_lora_chunk_size": 16, "experts_shared_outer_loras": null, "lora_use_virtual_experts": false, "lora_strict_loading": false, "lora_drain_wait_threshold": 0.0, "enable_two_batch_overlap": false, "enable_single_batch_overlap": false, "tbo_token_distribution_threshold": 0.48, "cpu_offload_gb": 0, "offload_group_size": -1, "offload_num_in_group": 1, "offload_prefetch_step": 1, "offload_mode": "cpu", "enable_lmcache": false, "lmcache_config_file": null, "enable_flexkv": false, "flexkv_config_file": null, "kt_weight_path": null, "kt_method": "AMXINT4", "kt_cpuinfer": null, "kt_threadpool_count": 2, "kt_num_gpu_experts": null, "kt_max_deferred_experts_per_token": null, "dllm_algorithm": null, "dllm_algorithm_config": null, "dllm_fdfo": true, "disaggregation_mode": "decode", "disaggregation_transfer_backend": "mooncake", "disaggregation_bootstrap_port": 28800, "disaggregation_ib_device": "mlx5_0,mlx5_1,mlx5_2,mlx5_3", "disaggregation_decode_enable_radix_cache": false, "disaggregation_decode_enable_offload_kvcache": false, "num_reserved_decode_tokens": 512, "disaggregation_decode_extra_slots": 0, "disaggregation_decode_polling_interval": 1, "optimistic_prefill_attempts": 0, "encoder_only": false, "language_only": false, "encoder_transfer_backend": "zmq_to_tokenizer", "encoder_urls": [], "encoder_bootstrap_port": 8997, "encoder_register_urls": [], "enable_adaptive_dispatch_to_encoder": false, "enable_pdmux": false, "pdmux_config_path": null, "sm_group_num": 8, "custom_weight_loader": [], "weight_loader_disable_mmap": false, "weight_loader_prefetch_checkpoints": false, "weight_loader_prefetch_num_threads": 4, "weight_loader_drop_cache_after_load": false, "remote_instance_weight_loader_seed_instance_ip": null, "remote_instance_weight_loader_seed_instance_service_port": null, "remote_instance_weight_loader_send_weights_group_ports": null, "remote_instance_weight_loader_backend": "nccl", "remote_instance_weight_loader_start_seed_via_transfer_engine": false, "engine_info_bootstrap_port": 6789, "modelexpress_config": null, "download_dir": null, "model_checksum": null, "delete_ckpt_after_loading": false, "decrypted_config_file": null, "decrypted_draft_config_file": null, "checkpoint_engine_wait_weights_before_ready": false, "enable_prefill_delayer": false, "prefill_delayer_max_delay_passes": 30, "prefill_delayer_token_usage_low_watermark": null, "prefill_delayer_forward_passes_buckets": null, "prefill_delayer_wait_seconds_buckets": null, "prefill_delayer_queue_min_ratio": null, "prefill_delayer_max_delay_ms": null, "min_free_slots_delay": null, "enable_deterministic_inference": false, "rl_on_policy_target": null, "kv_canary": "none", "kv_canary_real_data": "none", "kv_canary_sweep_interval": 0, "enable_dynamic_batch_tokenizer": false, "dynamic_batch_tokenizer_batch_size": 32, "dynamic_batch_tokenizer_batch_timeout": 0.002, "enable_tokenizer_batch_encode": false, "disable_tokenizer_batch_decode": false, "debug_tensor_dump_output_folder": null, "debug_tensor_dump_layers": null, "debug_tensor_dump_input_file": null, "enable_memory_saver": false, "enable_weights_cpu_backup": false, "enable_draft_weights_cpu_backup": false, "enable_custom_logit_processor": false, "enable_return_hidden_states": false, "enable_return_routed_experts": false, "enable_return_indexer_topk": false, "disable_outlines_disk_cache": false, "enable_mis": false, "weight_cache_mode": "off", "weight_cache_socket": null, "weight_cache_timeout": 1800, "forward_hooks": null, "msprobe_dump_config": null, "_resolved_overrides": [["_sampling_backend_default", {"sampling_backend": "flashinfer"}], ["_attention_backend_default", {"attention_backend": "triton"}], ["_data_parallelism_defaults", {"enable_dp_attention": false, "enable_dp_lm_head": false}], ["_speculative_moe_runner_default", {"speculative_moe_runner_backend": "marlin"}]], "grpc_worker_threads": 4, "_quantization_explicitly_unset": false, "_cuda_graph_config_locked": [["decode", "max_bs"]], "_declarations_materialized": true, "_runtime_mutations": [["model_runner.chunked_prefix_cache_gate", {"disable_chunked_prefix_cache": true}], ["mamba_pool.memory_budget", {"max_mamba_cache_size": 181}]], "_in_override": false, "_mx_config_cache": {}, "max_speculative_num_draft_tokens": null, "last_gen_throughput": 0.09006957280512767, "memory_usage": {"weight": 58.57, "kvcache": 11.34, "token_capacity": 440256, "graph": 0.6}, "effective_max_running_requests_per_dp": 181}], "prefill": [{"model_path": "/data/hf_models/Kimi-K3", "tokenizer_path": "/data/hf_models/Kimi-K3", "tokenizer_mode": "auto", "tokenizer_backend": "huggingface", "tokenizer_worker_num": 1, "detokenizer_worker_num": 1, "skip_tokenizer_init": false, "load_format": "auto", "model_loader_extra_config": "{}", "trust_remote_code": true, "context_length": null, "is_embedding": false, "enable_multimodal": null, "revision": null, "model_impl": "auto", "model_config_parser": "auto", "json_model_override_args": "{}", "dtype": "auto", "quantization": null, "quantization_param_path": null, "kv_cache_dtype": "auto", "enable_fp32_lm_head": false, "modelopt_quant": null, "modelopt_checkpoint_restore_path": null, "modelopt_checkpoint_save_path": null, "modelopt_export_path": null, "quantize_and_serve": false, "rl_quant_profile": null, "enable_tf32_matmul": false, "mem_fraction_static": 0.88, "max_running_requests": null, "max_queued_requests": null, "max_total_tokens": null, "chunked_prefill_size": 8192, "enable_dynamic_chunking": false, "max_prefill_tokens": 16384, "prefill_max_requests": null, "schedule_policy": "fcfs", "enable_priority_scheduling": false, "disable_priority_preemption": false, "default_priority_value": null, "abort_on_priority_when_disabled": false, "schedule_low_priority_values_first": false, "priority_scheduling_preemption_threshold": 10, "retraction_policy": "length", "schedule_conservativeness": 1.0, "page_size": 64, "swa_full_tokens_ratio": 0.8, "disable_hybrid_swa_memory": false, "radix_eviction_policy": "lru", "prefill_only_disable_kv_cache": false, "disable_radix_cache": true, "enable_page_major_kv_layout": false, "enable_unified_memory": false, "disable_chunked_prefix_cache": false, "disable_overlap_schedule": true, "num_continuous_decode_steps": 1, "scheduler_recv_interval": 1, "enable_mixed_chunk": false, "nccl_port": null, "dist_timeout": 3600, "dist_init_addr": "174.1.60.1:20000", "nnodes": 4, "node_rank": 0, "tp_size": 4, "dcp_size": 1, "pp_size": 8, "pp_max_micro_batch_size": null, "pp_async_batch_depth": 0, "dp_size": 1, "load_balance_method": "follow_bootstrap_room", "attn_cp_size": 1, "moe_dp_size": 1, "dcp_comm_backend": "ag_rs", "dcp_replicate_q_proj": null, "dwdp_size": 1, "enable_prefill_cp": false, "cp_strategy": null, "enable_dsa_cache_layer_split": false, "enable_dsa_prefill_context_parallel": false, "dsa_prefill_cp_mode": "round-robin-split", "enable_prefill_context_parallel": false, "prefill_cp_mode": "in-seq-split", "enable_cp_decode_attn_tp": false, "enable_dp_attention": false, "enable_dp_attention_local_control_broadcast": false, "enable_dp_lm_head": false, "enable_attn_tp_input_scattered": false, "disable_attn_tp_gather": false, "enable_p2p_check": false, "device": "cuda", "base_gpu_id": 0, "gpu_id_step": 1, "random_seed": 984607766, "watchdog_timeout": 300, "soft_watchdog_timeout": null, "sleep_on_idle": false, "use_ray": false, "custom_sigquit_handler": null, "numa_node": null, "gc_threshold": null, "host": "0.0.0.0", "port": 30000, "fastapi_root_path": "", "smg_grpc_mode": false, "grpc_mode": false, "grpc_port": null, "sidecar": null, "sidecar_args": null, "skip_server_warmup": false, "warmups": null, "enable_http2": false, "ssl_keyfile": null, "ssl_certfile": null, "ssl_ca_certs": null, "ssl_keyfile_password": null, "enable_ssl_refresh": false, "api_key": null, "admin_api_key": null, "served_model_name": "kimi-k3", "weight_version": "default", "chat_template": null, "hf_chat_template_name": null, "completion_template": null, "file_storage_path": "sglang_storage", "enable_cache_report": false, "reasoning_parser": null, "default_chat_template_kwargs": null, "strip_thinking_cache": false, "enable_strict_thinking": false, "tool_call_parser": null, "tool_server": null, "sampling_defaults": "model", "asr_max_buffer_seconds": 60, "asr_max_concurrent_sessions": 32, "preferred_sampling_params": null, "allow_auto_truncate": false, "stream_interval": 1, "batch_notify_size": 16, "stream_response_default_include_usage": false, "incremental_streaming_output": false, "enable_streaming_session": false, "enable_session_radix_cache": false, "log_level": "info", "log_level_http": null, "log_requests": false, "log_requests_level": 2, "log_requests_format": "text", "log_requests_target": null, "uvicorn_access_log_exclude_prefixes": [], "crash_dump_folder": null, "show_time_cost": false, "enable_metrics": false, "smg_http_sidecar_port": null, "enable_mfu_metrics": false, "enable_metrics_for_all_schedulers": false, "load_snapshot_publish_interval": 15, "tokenizer_metrics_custom_labels_header": "x-custom-labels", "tokenizer_metrics_allowed_custom_labels": null, "extra_metric_labels": null, "bucket_time_to_first_token": null, "bucket_inter_token_latency": null, "bucket_e2e_request_latency": null, "prompt_tokens_buckets": null, "generation_tokens_buckets": null, "gc_warning_threshold_secs": 0.0, "decode_log_interval": 40, "enable_request_time_stats_logging": false, "kv_events_config": null, "enable_forward_pass_metrics": false, "forward_pass_metrics_worker_id": "", "forward_pass_metrics_ipc_name": null, "enable_trace": false, "trace_modules": "request", "otlp_traces_endpoint": "localhost:4317", "export_metrics_to_file": false, "export_metrics_to_file_dir": null, "stat_loggers": null, "constrained_json_whitespace_pattern": null, "constrained_json_disable_any_whitespace": false, "attention_backend": "triton", "decode_attention_backend": null, "prefill_attention_backend": null, "sampling_backend": "flashinfer", "grammar_backend": "xgrammar", "radix_cache_backend": null, "mm_attention_backend": null, "fp8_gemm_runner_backend": "auto", "fp4_gemm_runner_backend": "auto", "bf16_gemm_backend": "auto", "dsa_prefill_backend": null, "dsa_decode_backend": null, "dsa_paged_mqa_logits_backend": "auto", "dsa_topk_backend": "sgl-kernel", "disable_flashinfer_autotune": false, "flashinfer_autotune_skip_ops": null, "mamba_backend": "triton", "cuda_graph_config": {"decode": {"backend": "disabled", "max_bs": 16, "bs": [1, 2, 4, 8, 12, 16], "tc_compiler": "eager", "full_prefill_max_req": null}, "prefill": {"backend": "disabled", "max_bs": 2048, "bs": [4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], "tc_compiler": "eager", "full_prefill_max_req": null}}, "cuda_graph_backend_decode": null, "cuda_graph_backend_prefill": null, "cuda_graph_max_bs_decode": 16, "cuda_graph_max_bs_prefill": null, "cuda_graph_bs_decode": null, "cuda_graph_bs_prefill": null, "cuda_graph_tc_compiler": null, "disable_prefill_cuda_graph": false, "disable_decode_cuda_graph": false, "disable_cuda_graph": false, "disable_cuda_graph_padding": false, "enable_profile_cuda_graph": false, "enable_cudagraph_gc": false, "debug_cuda_graph": false, "enable_layerwise_nvtx_marker": false, "enable_nccl_nvls": false, "enable_symm_mem": false, "triton_attention_reduce_in_fp32": false, "triton_attention_num_kv_splits": 8, "triton_attention_split_tile_size": null, "flashinfer_mla_disable_ragged": false, "enable_fused_qk_norm_rope": false, "enable_precise_embedding_interpolation": false, "enable_fused_moe_sum_all_reduce": false, "enable_deepseek_v4_fp4_indexer": false, "disable_custom_all_reduce": false, "enable_mscclpp": false, "enable_torch_symm_mem": false, "enable_scattered_sconv": false, "pre_warm_nccl": false, "enable_quant_communications": false, "enable_flashinfer_allreduce_fusion": false, "enforce_disable_flashinfer_allreduce_fusion": false, "flashinfer_allreduce_fusion_backend": null, "enable_aiter_allreduce_fusion": false, "enable_torch_compile": false, "enable_torch_compile_debug_mode": false, "torch_compile_max_bs": 32, "torchao_config": "", "speculative_algorithm": null, "speculative_draft_model_path": null, "speculative_draft_model_revision": null, "speculative_draft_load_format": null, "speculative_num_steps": null, "speculative_eagle_topk": null, "speculative_num_draft_tokens": null, "speculative_dflash_block_size": null, "speculative_dspark_block_size": null, "speculative_dspark_sps_table_path": null, "speculative_dspark_confidence_sts_path": null, "speculative_dspark_align_verify_tokens_to_graph_tier": false, "speculative_accept_threshold_single": 1.0, "speculative_accept_threshold_acc": 1.0, "speculative_use_rejection_sampling": false, "speculative_token_map": null, "speculative_attention_mode": "prefill", "speculative_draft_attention_backend": null, "speculative_draft_window_size": null, "speculative_moe_runner_backend": "flashinfer_mxfp4", "speculative_moe_a2a_backend": null, "speculative_draft_model_quantization": null, "speculative_skip_dp_mlp_sync": false, "enable_multi_layer_eagle": false, "speculative_adaptive": false, "speculative_adaptive_config": null, "decoupled_spec_bind_endpoint": null, "decoupled_spec_connect_endpoints": null, "decoupled_spec_rank": null, "decoupled_spec_role": "null", "spec_trace_dir": null, "speculative_ngram_min_bfs_breadth": 1, "speculative_ngram_max_bfs_breadth": 10, "speculative_ngram_match_type": "BFS", "speculative_ngram_max_trie_depth": 18, "speculative_ngram_capacity": 10000000, "speculative_ngram_external_corpus_path": null, "speculative_ngram_external_sam_budget": 0, "speculative_ngram_external_corpus_max_tokens": 10000000, "ep_size": 4, "moe_a2a_backend": "none", "moe_runner_backend": "flashinfer_mxfp4", "flashinfer_mxfp4_moe_precision": "default", "deepep_mode": "auto", "fuseep_mode": 2, "deepep_dispatcher_output_dtype": "auto", "ep_num_redundant_experts": 0, "ep_dispatch_algorithm": null, "init_expert_location": "trivial", "enable_eplb": false, "eplb_algorithm": "auto", "eplb_rebalance_num_iterations": 1000, "eplb_rebalance_layers_per_chunk": null, "eplb_min_rebalancing_utilization_threshold": 1.0, "expert_distribution_recorder_mode": null, "expert_distribution_recorder_buffer_size": 1000, "enable_expert_distribution_metrics": false, "deepep_config": null, "moe_dense_tp_size": null, "elastic_ep_backend": null, "enable_elastic_expert_backup": false, "mooncake_ib_device": null, "enable_waterfill": false, "ep_join_mode": null, "ep_join_rank_offset": 0, "elastic_ep_initial_size": null, "max_ep_size": null, "elastic_ep_scale_timeout": 600, "elastic_ep_rejoin": false, "disable_flashinfer_cutlass_moe_fp4_allgather": false, "disable_shared_experts_fusion": false, "enforce_shared_experts_fusion": false, "max_mamba_cache_size": null, "mamba_ssm_dtype": null, "mamba_max_states_per_path": -1, "enable_mamba_cache_stochastic_rounding": false, "mamba_cache_philox_rounds": 0, "mamba_full_memory_ratio": 0.36, "mamba_radix_cache_strategy": "extra_buffer_lazy", "uses_mamba_radix_cache": false, "mamba_track_interval": 256, "enable_int8_mamba_checkpoint": false, "int8_mamba_ckpt_size": null, "linear_attn_backend": "triton", "linear_attn_decode_backend": null, "linear_attn_prefill_backend": null, "linear_attn_verify_backend": null, "enable_linear_replayssm": false, "linear_replayssm_cache_len": 16, "enable_linear_replayssm_spec": false, "enable_hierarchical_cache": false, "hicache_ratio": 2.0, "hicache_size": 0, "hicache_write_policy": "write_through", "hicache_io_backend": "kernel", "hicache_mem_layout": "page_first", "hicache_storage_backend": null, "hicache_storage_prefetch_policy": "timeout", "hicache_storage_backend_extra_config": null, "enable_hisparse": false, "hisparse_config": null, "enable_broadcast_mm_inputs_process": false, "enable_prefix_mm_cache": false, "mm_enable_dp_encoder": false, "mm_process_config": {}, "mm_processor_worker_num": 0, "mm_io_worker_num": 0, "limit_mm_data_per_request": null, "enable_mm_global_cache": false, "disable_fast_image_processor": false, "mm_feature_transport": "cpu", "keep_mm_feature_on_device": false, "enable_lora": null, "enable_lora_overlap_loading": null, "max_lora_rank": null, "lora_target_modules": null, "lora_paths": null, "max_loaded_loras": null, "max_loras_per_batch": 8, "lora_eviction_policy": "lru", "lora_backend": "csgmv", "max_lora_chunk_size": 16, "experts_shared_outer_loras": null, "lora_use_virtual_experts": false, "lora_strict_loading": false, "lora_drain_wait_threshold": 0.0, "enable_two_batch_overlap": false, "enable_single_batch_overlap": false, "tbo_token_distribution_threshold": 0.48, "cpu_offload_gb": 0, "offload_group_size": -1, "offload_num_in_group": 1, "offload_prefetch_step": 1, "offload_mode": "cpu", "enable_lmcache": false, "lmcache_config_file": null, "enable_flexkv": false, "flexkv_config_file": null, "kt_weight_path": null, "kt_method": "AMXINT4", "kt_cpuinfer": null, "kt_threadpool_count": 2, "kt_num_gpu_experts": null, "kt_max_deferred_experts_per_token": null, "dllm_algorithm": null, "dllm_algorithm_config": null, "dllm_fdfo": true, "disaggregation_mode": "prefill", "disaggregation_transfer_backend": "mooncake", "disaggregation_bootstrap_port": 28800, "disaggregation_ib_device": "mlx5_0,mlx5_1,mlx5_2,mlx5_3", "disaggregation_decode_enable_radix_cache": false, "disaggregation_decode_enable_offload_kvcache": false, "num_reserved_decode_tokens": 512, "disaggregation_decode_extra_slots": null, "disaggregation_decode_polling_interval": 1, "optimistic_prefill_attempts": 0, "encoder_only": false, "language_only": false, "encoder_transfer_backend": "zmq_to_tokenizer", "encoder_urls": [], "encoder_bootstrap_port": 8997, "encoder_register_urls": [], "enable_adaptive_dispatch_to_encoder": false, "enable_pdmux": false, "pdmux_config_path": null, "sm_group_num": 8, "custom_weight_loader": [], "weight_loader_disable_mmap": false, "weight_loader_prefetch_checkpoints": false, "weight_loader_prefetch_num_threads": 4, "weight_loader_drop_cache_after_load": false, "remote_instance_weight_loader_seed_instance_ip": null, "remote_instance_weight_loader_seed_instance_service_port": null, "remote_instance_weight_loader_send_weights_group_ports": null, "remote_instance_weight_loader_backend": "nccl", "remote_instance_weight_loader_start_seed_via_transfer_engine": false, "engine_info_bootstrap_port": 6789, "modelexpress_config": null, "download_dir": null, "model_checksum": null, "delete_ckpt_after_loading": false, "decrypted_config_file": null, "decrypted_draft_config_file": null, "checkpoint_engine_wait_weights_before_ready": false, "enable_prefill_delayer": false, "prefill_delayer_max_delay_passes": 30, "prefill_delayer_token_usage_low_watermark": null, "prefill_delayer_forward_passes_buckets": null, "prefill_delayer_wait_seconds_buckets": null, "prefill_delayer_queue_min_ratio": null, "prefill_delayer_max_delay_ms": null, "min_free_slots_delay": null, "enable_deterministic_inference": false, "rl_on_policy_target": null, "kv_canary": "none", "kv_canary_real_data": "none", "kv_canary_sweep_interval": 0, "enable_dynamic_batch_tokenizer": false, "dynamic_batch_tokenizer_batch_size": 32, "dynamic_batch_tokenizer_batch_timeout": 0.002, "enable_tokenizer_batch_encode": false, "disable_tokenizer_batch_decode": false, "debug_tensor_dump_output_folder": null, "debug_tensor_dump_layers": null, "debug_tensor_dump_input_file": null, "enable_memory_saver": false, "enable_weights_cpu_backup": false, "enable_draft_weights_cpu_backup": false, "enable_custom_logit_processor": false, "enable_return_hidden_states": false, "enable_return_routed_experts": false, "enable_return_indexer_topk": false, "disable_outlines_disk_cache": false, "enable_mis": false, "weight_cache_mode": "off", "weight_cache_socket": null, "weight_cache_timeout": 1800, "forward_hooks": null, "msprobe_dump_config": null, "status": "ready", "max_total_num_tokens": 3801408, "max_req_input_len": 1048570, "internal_states": [{"model_path": "/data/hf_models/Kimi-K3", "tokenizer_path": "/data/hf_models/Kimi-K3", "tokenizer_mode": "auto", "tokenizer_backend": "huggingface", "tokenizer_worker_num": 1, "detokenizer_worker_num": 1, "skip_tokenizer_init": false, "load_format": "auto", "model_loader_extra_config": "{}", "trust_remote_code": true, "context_length": null, "is_embedding": false, "enable_multimodal": null, "revision": null, "model_impl": "auto", "model_config_parser": "auto", "json_model_override_args": "{}", "dtype": "auto", "quantization": null, "quantization_param_path": null, "kv_cache_dtype": "auto", "enable_fp32_lm_head": false, "modelopt_quant": null, "modelopt_checkpoint_restore_path": null, "modelopt_checkpoint_save_path": null, "modelopt_export_path": null, "quantize_and_serve": false, "rl_quant_profile": null, "enable_tf32_matmul": false, "mem_fraction_static": 0.88, "max_running_requests": null, "max_queued_requests": null, "max_total_tokens": null, "chunked_prefill_size": 8192, "enable_dynamic_chunking": false, "max_prefill_tokens": 16384, "prefill_max_requests": null, "schedule_policy": "fcfs", "enable_priority_scheduling": false, "disable_priority_preemption": false, "default_priority_value": null, "abort_on_priority_when_disabled": false, "schedule_low_priority_values_first": false, "priority_scheduling_preemption_threshold": 10, "retraction_policy": "length", "schedule_conservativeness": 1.0, "page_size": 64, "swa_full_tokens_ratio": 0.8, "disable_hybrid_swa_memory": false, "radix_eviction_policy": "lru", "prefill_only_disable_kv_cache": false, "disable_radix_cache": true, "enable_page_major_kv_layout": false, "enable_unified_memory": false, "disable_chunked_prefix_cache": true, "disable_overlap_schedule": true, "num_continuous_decode_steps": 1, "scheduler_recv_interval": 1, "enable_mixed_chunk": false, "nccl_port": null, "dist_timeout": 3600, "dist_init_addr": "174.1.60.1:20000", "nnodes": 4, "node_rank": 0, "tp_size": 4, "dcp_size": 1, "pp_size": 8, "pp_max_micro_batch_size": 6, "pp_async_batch_depth": 0, "dp_size": 1, "load_balance_method": "follow_bootstrap_room", "attn_cp_size": 1, "moe_dp_size": 1, "dcp_comm_backend": "ag_rs", "dcp_replicate_q_proj": null, "dwdp_size": 1, "enable_prefill_cp": false, "cp_strategy": null, "enable_dsa_cache_layer_split": false, "enable_dsa_prefill_context_parallel": false, "dsa_prefill_cp_mode": "round-robin-split", "enable_prefill_context_parallel": false, "prefill_cp_mode": "in-seq-split", "enable_cp_decode_attn_tp": false, "enable_dp_attention": false, "enable_dp_attention_local_control_broadcast": false, "enable_dp_lm_head": false, "enable_attn_tp_input_scattered": false, "disable_attn_tp_gather": false, "enable_p2p_check": false, "device": "cuda", "base_gpu_id": 0, "gpu_id_step": 1, "random_seed": 984607766, "watchdog_timeout": 300, "soft_watchdog_timeout": null, "sleep_on_idle": false, "use_ray": false, "numa_node": null, "gc_threshold": null, "host": "0.0.0.0", "port": 30000, "fastapi_root_path": "", "smg_grpc_mode": false, "grpc_mode": false, "grpc_port": null, "sidecar": null, "sidecar_args": null, "skip_server_warmup": false, "warmups": null, "enable_http2": false, "ssl_keyfile": null, "ssl_certfile": null, "ssl_ca_certs": null, "ssl_keyfile_password": null, "enable_ssl_refresh": false, "api_key": null, "admin_api_key": null, "served_model_name": "kimi-k3", "weight_version": "default", "chat_template": null, "hf_chat_template_name": null, "completion_template": null, "file_storage_path": "sglang_storage", "enable_cache_report": false, "reasoning_parser": null, "default_chat_template_kwargs": null, "strip_thinking_cache": false, "enable_strict_thinking": false, "tool_call_parser": null, "tool_server": null, "sampling_defaults": "model", "asr_max_buffer_seconds": 60, "asr_max_concurrent_sessions": 32, "preferred_sampling_params": null, "allow_auto_truncate": false, "stream_interval": 1, "batch_notify_size": 16, "stream_response_default_include_usage": false, "incremental_streaming_output": false, "enable_streaming_session": false, "enable_session_radix_cache": false, "log_level": "info", "log_level_http": null, "log_requests": false, "log_requests_level": 2, "log_requests_format": "text", "log_requests_target": null, "uvicorn_access_log_exclude_prefixes": [], "crash_dump_folder": null, "show_time_cost": false, "enable_metrics": false, "smg_http_sidecar_port": null, "enable_mfu_metrics": false, "enable_metrics_for_all_schedulers": false, "load_snapshot_publish_interval": 15, "tokenizer_metrics_custom_labels_header": "x-custom-labels", "tokenizer_metrics_allowed_custom_labels": null, "extra_metric_labels": null, "bucket_time_to_first_token": null, "bucket_inter_token_latency": null, "bucket_e2e_request_latency": null, "prompt_tokens_buckets": null, "generation_tokens_buckets": null, "gc_warning_threshold_secs": 0.0, "decode_log_interval": 40, "enable_request_time_stats_logging": false, "kv_events_config": null, "enable_forward_pass_metrics": false, "forward_pass_metrics_worker_id": "", "forward_pass_metrics_ipc_name": null, "enable_trace": false, "trace_modules": "request", "otlp_traces_endpoint": "localhost:4317", "export_metrics_to_file": false, "export_metrics_to_file_dir": null, "stat_loggers": null, "constrained_json_whitespace_pattern": null, "constrained_json_disable_any_whitespace": false, "attention_backend": "triton", "decode_attention_backend": null, "prefill_attention_backend": null, "sampling_backend": "flashinfer", "grammar_backend": "xgrammar", "radix_cache_backend": null, "mm_attention_backend": null, "fp8_gemm_runner_backend": "auto", "fp4_gemm_runner_backend": "auto", "bf16_gemm_backend": "auto", "dsa_prefill_backend": null, "dsa_decode_backend": null, "dsa_paged_mqa_logits_backend": "auto", "dsa_topk_backend": "sgl-kernel", "disable_flashinfer_autotune": false, "flashinfer_autotune_skip_ops": null, "mamba_backend": "triton", "cuda_graph_config": {"decode": {"backend": "disabled", "max_bs": 16, "bs": [1, 2, 4, 8, 12, 16], "tc_compiler": "eager", "full_prefill_max_req": null}, "prefill": {"backend": "disabled", "max_bs": 2048, "bs": [4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], "tc_compiler": "eager", "full_prefill_max_req": null}}, "cuda_graph_backend_decode": null, "cuda_graph_backend_prefill": null, "cuda_graph_max_bs_decode": 16, "cuda_graph_max_bs_prefill": null, "cuda_graph_bs_decode": null, "cuda_graph_bs_prefill": null, "cuda_graph_tc_compiler": null, "disable_prefill_cuda_graph": false, "disable_decode_cuda_graph": false, "disable_cuda_graph": false, "disable_cuda_graph_padding": false, "enable_profile_cuda_graph": false, "enable_cudagraph_gc": false, "debug_cuda_graph": false, "enable_layerwise_nvtx_marker": false, "enable_nccl_nvls": false, "enable_symm_mem": false, "triton_attention_reduce_in_fp32": false, "triton_attention_num_kv_splits": 8, "triton_attention_split_tile_size": null, "flashinfer_mla_disable_ragged": false, "enable_fused_qk_norm_rope": false, "enable_precise_embedding_interpolation": false, "enable_fused_moe_sum_all_reduce": false, "enable_deepseek_v4_fp4_indexer": false, "disable_custom_all_reduce": false, "enable_mscclpp": false, "enable_torch_symm_mem": false, "enable_scattered_sconv": false, "pre_warm_nccl": false, "enable_quant_communications": false, "enable_flashinfer_allreduce_fusion": false, "enforce_disable_flashinfer_allreduce_fusion": false, "flashinfer_allreduce_fusion_backend": null, "enable_aiter_allreduce_fusion": false, "enable_torch_compile": false, "enable_torch_compile_debug_mode": false, "torch_compile_max_bs": 32, "torchao_config": "", "speculative_algorithm": null, "speculative_draft_model_path": null, "speculative_draft_model_revision": null, "speculative_draft_load_format": null, "speculative_num_steps": null, "speculative_eagle_topk": null, "speculative_num_draft_tokens": null, "speculative_dflash_block_size": null, "speculative_dspark_block_size": null, "speculative_dspark_sps_table_path": null, "speculative_dspark_confidence_sts_path": null, "speculative_dspark_align_verify_tokens_to_graph_tier": false, "speculative_accept_threshold_single": 1.0, "speculative_accept_threshold_acc": 1.0, "speculative_use_rejection_sampling": false, "speculative_token_map": null, "speculative_attention_mode": "prefill", "speculative_draft_attention_backend": null, "speculative_draft_window_size": null, "speculative_moe_runner_backend": "flashinfer_mxfp4", "speculative_moe_a2a_backend": null, "speculative_draft_model_quantization": null, "speculative_skip_dp_mlp_sync": false, "enable_multi_layer_eagle": false, "speculative_adaptive": false, "speculative_adaptive_config": null, "decoupled_spec_bind_endpoint": null, "decoupled_spec_connect_endpoints": null, "decoupled_spec_rank": null, "decoupled_spec_role": "null", "spec_trace_dir": null, "speculative_ngram_min_bfs_breadth": 1, "speculative_ngram_max_bfs_breadth": 10, "speculative_ngram_match_type": "BFS", "speculative_ngram_max_trie_depth": 18, "speculative_ngram_capacity": 10000000, "speculative_ngram_external_corpus_path": null, "speculative_ngram_external_sam_budget": 0, "speculative_ngram_external_corpus_max_tokens": 10000000, "ep_size": 4, "moe_a2a_backend": "none", "moe_runner_backend": "flashinfer_mxfp4", "flashinfer_mxfp4_moe_precision": "default", "deepep_mode": "auto", "fuseep_mode": 2, "deepep_dispatcher_output_dtype": "auto", "ep_num_redundant_experts": 0, "ep_dispatch_algorithm": null, "init_expert_location": "trivial", "enable_eplb": false, "eplb_algorithm": "auto", "eplb_rebalance_num_iterations": 1000, "eplb_rebalance_layers_per_chunk": null, "eplb_min_rebalancing_utilization_threshold": 1.0, "expert_distribution_recorder_mode": null, "expert_distribution_recorder_buffer_size": 1000, "enable_expert_distribution_metrics": false, "deepep_config": null, "moe_dense_tp_size": null, "elastic_ep_backend": null, "enable_elastic_expert_backup": false, "mooncake_ib_device": null, "enable_waterfill": false, "ep_join_mode": null, "ep_join_rank_offset": 0, "elastic_ep_initial_size": null, "max_ep_size": null, "elastic_ep_scale_timeout": 600, "elastic_ep_rejoin": false, "disable_flashinfer_cutlass_moe_fp4_allgather": false, "disable_shared_experts_fusion": false, "enforce_shared_experts_fusion": false, "max_mamba_cache_size": 54, "mamba_ssm_dtype": null, "mamba_max_states_per_path": -1, "enable_mamba_cache_stochastic_rounding": false, "mamba_cache_philox_rounds": 0, "mamba_full_memory_ratio": 0.36, "mamba_radix_cache_strategy": "extra_buffer_lazy", "uses_mamba_radix_cache": false, "mamba_track_interval": 256, "enable_int8_mamba_checkpoint": false, "int8_mamba_ckpt_size": null, "linear_attn_backend": "triton", "linear_attn_decode_backend": null, "linear_attn_prefill_backend": null, "linear_attn_verify_backend": null, "enable_linear_replayssm": false, "linear_replayssm_cache_len": 16, "enable_linear_replayssm_spec": false, "enable_hierarchical_cache": false, "hicache_ratio": 2.0, "hicache_size": 0, "hicache_write_policy": "write_through", "hicache_io_backend": "kernel", "hicache_mem_layout": "page_first", "hicache_storage_backend": null, "hicache_storage_prefetch_policy": "timeout", "hicache_storage_backend_extra_config": null, "enable_hisparse": false, "hisparse_config": null, "enable_broadcast_mm_inputs_process": false, "enable_prefix_mm_cache": false, "mm_enable_dp_encoder": false, "mm_process_config": {}, "mm_processor_worker_num": 0, "mm_io_worker_num": 0, "limit_mm_data_per_request": null, "enable_mm_global_cache": false, "disable_fast_image_processor": false, "mm_feature_transport": "cpu", "keep_mm_feature_on_device": false, "enable_lora": null, "enable_lora_overlap_loading": null, "max_lora_rank": null, "lora_target_modules": null, "lora_paths": null, "max_loaded_loras": null, "max_loras_per_batch": 8, "lora_eviction_policy": "lru", "lora_backend": "csgmv", "max_lora_chunk_size": 16, "experts_shared_outer_loras": null, "lora_use_virtual_experts": false, "lora_strict_loading": false, "lora_drain_wait_threshold": 0.0, "enable_two_batch_overlap": false, "enable_single_batch_overlap": false, "tbo_token_distribution_threshold": 0.48, "cpu_offload_gb": 0, "offload_group_size": -1, "offload_num_in_group": 1, "offload_prefetch_step": 1, "offload_mode": "cpu", "enable_lmcache": false, "lmcache_config_file": null, "enable_flexkv": false, "flexkv_config_file": null, "kt_weight_path": null, "kt_method": "AMXINT4", "kt_cpuinfer": null, "kt_threadpool_count": 2, "kt_num_gpu_experts": null, "kt_max_deferred_experts_per_token": null, "dllm_algorithm": null, "dllm_algorithm_config": null, "dllm_fdfo": true, "disaggregation_mode": "prefill", "disaggregation_transfer_backend": "mooncake", "disaggregation_bootstrap_port": 28800, "disaggregation_ib_device": "mlx5_0,mlx5_1,mlx5_2,mlx5_3", "disaggregation_decode_enable_radix_cache": false, "disaggregation_decode_enable_offload_kvcache": false, "num_reserved_decode_tokens": 512, "disaggregation_decode_extra_slots": null, "disaggregation_decode_polling_interval": 1, "optimistic_prefill_attempts": 0, "encoder_only": false, "language_only": false, "encoder_transfer_backend": "zmq_to_tokenizer", "encoder_urls": [], "encoder_bootstrap_port": 8997, "encoder_register_urls": [], "enable_adaptive_dispatch_to_encoder": false, "enable_pdmux": false, "pdmux_config_path": null, "sm_group_num": 8, "custom_weight_loader": [], "weight_loader_disable_mmap": false, "weight_loader_prefetch_checkpoints": false, "weight_loader_prefetch_num_threads": 4, "weight_loader_drop_cache_after_load": false, "remote_instance_weight_loader_seed_instance_ip": null, "remote_instance_weight_loader_seed_instance_service_port": null, "remote_instance_weight_loader_send_weights_group_ports": null, "remote_instance_weight_loader_backend": "nccl", "remote_instance_weight_loader_start_seed_via_transfer_engine": false, "engine_info_bootstrap_port": 6789, "modelexpress_config": null, "download_dir": null, "model_checksum": null, "delete_ckpt_after_loading": false, "decrypted_config_file": null, "decrypted_draft_config_file": null, "checkpoint_engine_wait_weights_before_ready": false, "enable_prefill_delayer": false, "prefill_delayer_max_delay_passes": 30, "prefill_delayer_token_usage_low_watermark": null, "prefill_delayer_forward_passes_buckets": null, "prefill_delayer_wait_seconds_buckets": null, "prefill_delayer_queue_min_ratio": null, "prefill_delayer_max_delay_ms": null, "min_free_slots_delay": null, "enable_deterministic_inference": false, "rl_on_policy_target": null, "kv_canary": "none", "kv_canary_real_data": "none", "kv_canary_sweep_interval": 0, "enable_dynamic_batch_tokenizer": false, "dynamic_batch_tokenizer_batch_size": 32, "dynamic_batch_tokenizer_batch_timeout": 0.002, "enable_tokenizer_batch_encode": false, "disable_tokenizer_batch_decode": false, "debug_tensor_dump_output_folder": null, "debug_tensor_dump_layers": null, "debug_tensor_dump_input_file": null, "enable_memory_saver": false, "enable_weights_cpu_backup": false, "enable_draft_weights_cpu_backup": false, "enable_custom_logit_processor": false, "enable_return_hidden_states": false, "enable_return_routed_experts": false, "enable_return_indexer_topk": false, "disable_outlines_disk_cache": false, "enable_mis": false, "weight_cache_mode": "off", "weight_cache_socket": null, "weight_cache_timeout": 1800, "forward_hooks": null, "msprobe_dump_config": null, "_resolved_overrides": [["_sampling_backend_default", {"sampling_backend": "flashinfer"}], ["_attention_backend_default", {"attention_backend": "triton"}], ["_data_parallelism_defaults", {"enable_dp_attention": false, "enable_dp_lm_head": false}], ["_pipeline_parallel_overlap_disable", {"disable_overlap_schedule": true}], ["_speculative_moe_runner_default", {"speculative_moe_runner_backend": "flashinfer_mxfp4"}]], "grpc_worker_threads": 4, "_quantization_explicitly_unset": false, "_cuda_graph_config_locked": [["decode", "max_bs"]], "_declarations_materialized": true, "_runtime_mutations": [["model_runner.chunked_prefix_cache_gate", {"disable_chunked_prefix_cache": true}], ["mamba_pool.memory_budget", {"max_mamba_cache_size": 54}]], "_in_override": false, "_mx_config_cache": {}, "max_speculative_num_draft_tokens": null, "last_gen_throughput": 0.0, "memory_usage": {"weight": 43.09, "kvcache": 8.16, "token_capacity": 3801408, "graph": 0}, "effective_max_running_requests_per_dp": 54}], "version": "0.5.16", "kv_events": null}], "decode": [{"model_path": "/data/hf_models/Kimi-K3", "tokenizer_path": "/data/hf_models/Kimi-K3", "tokenizer_mode": "auto", "tokenizer_backend": "huggingface", "tokenizer_worker_num": 1, "detokenizer_worker_num": 1, "skip_tokenizer_init": false, "load_format": "auto", "model_loader_extra_config": "{}", "trust_remote_code": true, "context_length": null, "is_embedding": false, "enable_multimodal": null, "revision": null, "model_impl": "auto", "model_config_parser": "auto", "json_model_override_args": "{}", "dtype": "auto", "quantization": null, "quantization_param_path": null, "kv_cache_dtype": "auto", "enable_fp32_lm_head": false, "modelopt_quant": null, "modelopt_checkpoint_restore_path": null, "modelopt_checkpoint_save_path": null, "modelopt_export_path": null, "quantize_and_serve": false, "rl_quant_profile": null, "enable_tf32_matmul": false, "mem_fraction_static": 0.88, "max_running_requests": null, "max_queued_requests": null, "max_total_tokens": null, "chunked_prefill_size": 8192, "enable_dynamic_chunking": false, "max_prefill_tokens": 16384, "prefill_max_requests": null, "schedule_policy": "fcfs", "enable_priority_scheduling": false, "disable_priority_preemption": false, "default_priority_value": null, "abort_on_priority_when_disabled": false, "schedule_low_priority_values_first": false, "priority_scheduling_preemption_threshold": 10, "retraction_policy": "length", "schedule_conservativeness": 1.0, "page_size": 64, "swa_full_tokens_ratio": 0.8, "disable_hybrid_swa_memory": false, "radix_eviction_policy": "lru", "prefill_only_disable_kv_cache": false, "disable_radix_cache": true, "enable_page_major_kv_layout": false, "enable_unified_memory": false, "disable_chunked_prefix_cache": false, "disable_overlap_schedule": false, "num_continuous_decode_steps": 1, "scheduler_recv_interval": 1, "enable_mixed_chunk": false, "nccl_port": null, "dist_timeout": 3600, "dist_init_addr": "174.1.60.5:20000", "nnodes": 4, "node_rank": 0, "tp_size": 32, "dcp_size": 1, "pp_size": 1, "pp_max_micro_batch_size": null, "pp_async_batch_depth": 0, "dp_size": 1, "load_balance_method": "round_robin", "attn_cp_size": 1, "moe_dp_size": 1, "dcp_comm_backend": "ag_rs", "dcp_replicate_q_proj": null, "dwdp_size": 1, "enable_prefill_cp": false, "cp_strategy": null, "enable_dsa_cache_layer_split": false, "enable_dsa_prefill_context_parallel": false, "dsa_prefill_cp_mode": "round-robin-split", "enable_prefill_context_parallel": false, "prefill_cp_mode": "in-seq-split", "enable_cp_decode_attn_tp": false, "enable_dp_attention": false, "enable_dp_attention_local_control_broadcast": false, "enable_dp_lm_head": false, "enable_attn_tp_input_scattered": false, "disable_attn_tp_gather": false, "enable_p2p_check": false, "device": "cuda", "base_gpu_id": 0, "gpu_id_step": 1, "random_seed": 369235288, "watchdog_timeout": 300, "soft_watchdog_timeout": null, "sleep_on_idle": false, "use_ray": false, "custom_sigquit_handler": null, "numa_node": null, "gc_threshold": null, "host": "0.0.0.0", "port": 30000, "fastapi_root_path": "", "smg_grpc_mode": false, "grpc_mode": false, "grpc_port": null, "sidecar": null, "sidecar_args": null, "skip_server_warmup": false, "warmups": null, "enable_http2": false, "ssl_keyfile": null, "ssl_certfile": null, "ssl_ca_certs": null, "ssl_keyfile_password": null, "enable_ssl_refresh": false, "api_key": null, "admin_api_key": null, "served_model_name": "kimi-k3", "weight_version": "default", "chat_template": null, "hf_chat_template_name": null, "completion_template": null, "file_storage_path": "sglang_storage", "enable_cache_report": false, "reasoning_parser": null, "default_chat_template_kwargs": null, "strip_thinking_cache": false, "enable_strict_thinking": false, "tool_call_parser": null, "tool_server": null, "sampling_defaults": "model", "asr_max_buffer_seconds": 60, "asr_max_concurrent_sessions": 32, "preferred_sampling_params": null, "allow_auto_truncate": false, "stream_interval": 1, "batch_notify_size": 16, "stream_response_default_include_usage": false, "incremental_streaming_output": false, "enable_streaming_session": false, "enable_session_radix_cache": false, "log_level": "info", "log_level_http": null, "log_requests": false, "log_requests_level": 2, "log_requests_format": "text", "log_requests_target": null, "uvicorn_access_log_exclude_prefixes": [], "crash_dump_folder": null, "show_time_cost": false, "enable_metrics": false, "smg_http_sidecar_port": null, "enable_mfu_metrics": false, "enable_metrics_for_all_schedulers": false, "load_snapshot_publish_interval": 15, "tokenizer_metrics_custom_labels_header": "x-custom-labels", "tokenizer_metrics_allowed_custom_labels": null, "extra_metric_labels": null, "bucket_time_to_first_token": null, "bucket_inter_token_latency": null, "bucket_e2e_request_latency": null, "prompt_tokens_buckets": null, "generation_tokens_buckets": null, "gc_warning_threshold_secs": 0.0, "decode_log_interval": 40, "enable_request_time_stats_logging": false, "kv_events_config": null, "enable_forward_pass_metrics": false, "forward_pass_metrics_worker_id": "", "forward_pass_metrics_ipc_name": null, "enable_trace": false, "trace_modules": "request", "otlp_traces_endpoint": "localhost:4317", "export_metrics_to_file": false, "export_metrics_to_file_dir": null, "stat_loggers": null, "constrained_json_whitespace_pattern": null, "constrained_json_disable_any_whitespace": false, "attention_backend": "triton", "decode_attention_backend": null, "prefill_attention_backend": null, "sampling_backend": "flashinfer", "grammar_backend": "xgrammar", "radix_cache_backend": null, "mm_attention_backend": null, "fp8_gemm_runner_backend": "auto", "fp4_gemm_runner_backend": "auto", "bf16_gemm_backend": "auto", "dsa_prefill_backend": null, "dsa_decode_backend": null, "dsa_paged_mqa_logits_backend": "auto", "dsa_topk_backend": "sgl-kernel", "disable_flashinfer_autotune": false, "flashinfer_autotune_skip_ops": null, "mamba_backend": "triton", "cuda_graph_config": {"decode": {"backend": "full", "max_bs": 16, "bs": [1, 2, 4, 8, 12, 16], "tc_compiler": "eager", "full_prefill_max_req": null}, "prefill": {"backend": "disabled", "max_bs": 2048, "bs": [4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], "tc_compiler": "eager", "full_prefill_max_req": null}}, "cuda_graph_backend_decode": null, "cuda_graph_backend_prefill": null, "cuda_graph_max_bs_decode": 16, "cuda_graph_max_bs_prefill": null, "cuda_graph_bs_decode": null, "cuda_graph_bs_prefill": null, "cuda_graph_tc_compiler": null, "disable_prefill_cuda_graph": false, "disable_decode_cuda_graph": false, "disable_cuda_graph": false, "disable_cuda_graph_padding": false, "enable_profile_cuda_graph": false, "enable_cudagraph_gc": false, "debug_cuda_graph": false, "enable_layerwise_nvtx_marker": false, "enable_nccl_nvls": false, "enable_symm_mem": false, "triton_attention_reduce_in_fp32": false, "triton_attention_num_kv_splits": 8, "triton_attention_split_tile_size": null, "flashinfer_mla_disable_ragged": false, "enable_fused_qk_norm_rope": false, "enable_precise_embedding_interpolation": false, "enable_fused_moe_sum_all_reduce": false, "enable_deepseek_v4_fp4_indexer": false, "disable_custom_all_reduce": false, "enable_mscclpp": false, "enable_torch_symm_mem": false, "enable_scattered_sconv": false, "pre_warm_nccl": false, "enable_quant_communications": false, "enable_flashinfer_allreduce_fusion": false, "enforce_disable_flashinfer_allreduce_fusion": false, "flashinfer_allreduce_fusion_backend": null, "enable_aiter_allreduce_fusion": false, "enable_torch_compile": false, "enable_torch_compile_debug_mode": false, "torch_compile_max_bs": 32, "torchao_config": "", "speculative_algorithm": null, "speculative_draft_model_path": null, "speculative_draft_model_revision": null, "speculative_draft_load_format": null, "speculative_num_steps": null, "speculative_eagle_topk": null, "speculative_num_draft_tokens": null, "speculative_dflash_block_size": null, "speculative_dspark_block_size": null, "speculative_dspark_sps_table_path": null, "speculative_dspark_confidence_sts_path": null, "speculative_dspark_align_verify_tokens_to_graph_tier": false, "speculative_accept_threshold_single": 1.0, "speculative_accept_threshold_acc": 1.0, "speculative_use_rejection_sampling": false, "speculative_token_map": null, "speculative_attention_mode": "prefill", "speculative_draft_attention_backend": null, "speculative_draft_window_size": null, "speculative_moe_runner_backend": "marlin", "speculative_moe_a2a_backend": null, "speculative_draft_model_quantization": null, "speculative_skip_dp_mlp_sync": false, "enable_multi_layer_eagle": false, "speculative_adaptive": false, "speculative_adaptive_config": null, "decoupled_spec_bind_endpoint": null, "decoupled_spec_connect_endpoints": null, "decoupled_spec_rank": null, "decoupled_spec_role": "null", "spec_trace_dir": null, "speculative_ngram_min_bfs_breadth": 1, "speculative_ngram_max_bfs_breadth": 10, "speculative_ngram_match_type": "BFS", "speculative_ngram_max_trie_depth": 18, "speculative_ngram_capacity": 10000000, "speculative_ngram_external_corpus_path": null, "speculative_ngram_external_sam_budget": 0, "speculative_ngram_external_corpus_max_tokens": 10000000, "ep_size": 32, "moe_a2a_backend": "none", "moe_runner_backend": "marlin", "flashinfer_mxfp4_moe_precision": "default", "deepep_mode": "auto", "fuseep_mode": 2, "deepep_dispatcher_output_dtype": "auto", "ep_num_redundant_experts": 0, "ep_dispatch_algorithm": null, "init_expert_location": "trivial", "enable_eplb": false, "eplb_algorithm": "auto", "eplb_rebalance_num_iterations": 1000, "eplb_rebalance_layers_per_chunk": null, "eplb_min_rebalancing_utilization_threshold": 1.0, "expert_distribution_recorder_mode": null, "expert_distribution_recorder_buffer_size": 1000, "enable_expert_distribution_metrics": false, "deepep_config": null, "moe_dense_tp_size": null, "elastic_ep_backend": null, "enable_elastic_expert_backup": false, "mooncake_ib_device": null, "enable_waterfill": false, "ep_join_mode": null, "ep_join_rank_offset": 0, "elastic_ep_initial_size": null, "max_ep_size": null, "elastic_ep_scale_timeout": 600, "elastic_ep_rejoin": false, "disable_flashinfer_cutlass_moe_fp4_allgather": false, "disable_shared_experts_fusion": false, "enforce_shared_experts_fusion": false, "max_mamba_cache_size": null, "mamba_ssm_dtype": null, "mamba_max_states_per_path": -1, "enable_mamba_cache_stochastic_rounding": false, "mamba_cache_philox_rounds": 0, "mamba_full_memory_ratio": 0.21, "mamba_radix_cache_strategy": "extra_buffer_lazy", "uses_mamba_radix_cache": false, "mamba_track_interval": 256, "enable_int8_mamba_checkpoint": false, "int8_mamba_ckpt_size": null, "linear_attn_backend": "triton", "linear_attn_decode_backend": null, "linear_attn_prefill_backend": null, "linear_attn_verify_backend": null, "enable_linear_replayssm": false, "linear_replayssm_cache_len": 16, "enable_linear_replayssm_spec": false, "enable_hierarchical_cache": false, "hicache_ratio": 2.0, "hicache_size": 0, "hicache_write_policy": "write_through", "hicache_io_backend": "kernel", "hicache_mem_layout": "page_first", "hicache_storage_backend": null, "hicache_storage_prefetch_policy": "timeout", "hicache_storage_backend_extra_config": null, "enable_hisparse": false, "hisparse_config": null, "enable_broadcast_mm_inputs_process": false, "enable_prefix_mm_cache": false, "mm_enable_dp_encoder": false, "mm_process_config": {}, "mm_processor_worker_num": 0, "mm_io_worker_num": 0, "limit_mm_data_per_request": null, "enable_mm_global_cache": false, "disable_fast_image_processor": false, "mm_feature_transport": "cpu", "keep_mm_feature_on_device": false, "enable_lora": null, "enable_lora_overlap_loading": null, "max_lora_rank": null, "lora_target_modules": null, "lora_paths": null, "max_loaded_loras": null, "max_loras_per_batch": 8, "lora_eviction_policy": "lru", "lora_backend": "csgmv", "max_lora_chunk_size": 16, "experts_shared_outer_loras": null, "lora_use_virtual_experts": false, "lora_strict_loading": false, "lora_drain_wait_threshold": 0.0, "enable_two_batch_overlap": false, "enable_single_batch_overlap": false, "tbo_token_distribution_threshold": 0.48, "cpu_offload_gb": 0, "offload_group_size": -1, "offload_num_in_group": 1, "offload_prefetch_step": 1, "offload_mode": "cpu", "enable_lmcache": false, "lmcache_config_file": null, "enable_flexkv": false, "flexkv_config_file": null, "kt_weight_path": null, "kt_method": "AMXINT4", "kt_cpuinfer": null, "kt_threadpool_count": 2, "kt_num_gpu_experts": null, "kt_max_deferred_experts_per_token": null, "dllm_algorithm": null, "dllm_algorithm_config": null, "dllm_fdfo": true, "disaggregation_mode": "decode", "disaggregation_transfer_backend": "mooncake", "disaggregation_bootstrap_port": 28800, "disaggregation_ib_device": "mlx5_0,mlx5_1,mlx5_2,mlx5_3", "disaggregation_decode_enable_radix_cache": false, "disaggregation_decode_enable_offload_kvcache": false, "num_reserved_decode_tokens": 512, "disaggregation_decode_extra_slots": 0, "disaggregation_decode_polling_interval": 1, "optimistic_prefill_attempts": 0, "encoder_only": false, "language_only": false, "encoder_transfer_backend": "zmq_to_tokenizer", "encoder_urls": [], "encoder_bootstrap_port": 8997, "encoder_register_urls": [], "enable_adaptive_dispatch_to_encoder": false, "enable_pdmux": false, "pdmux_config_path": null, "sm_group_num": 8, "custom_weight_loader": [], "weight_loader_disable_mmap": false, "weight_loader_prefetch_checkpoints": false, "weight_loader_prefetch_num_threads": 4, "weight_loader_drop_cache_after_load": false, "remote_instance_weight_loader_seed_instance_ip": null, "remote_instance_weight_loader_seed_instance_service_port": null, "remote_instance_weight_loader_send_weights_group_ports": null, "remote_instance_weight_loader_backend": "nccl", "remote_instance_weight_loader_start_seed_via_transfer_engine": false, "engine_info_bootstrap_port": 6789, "modelexpress_config": null, "download_dir": null, "model_checksum": null, "delete_ckpt_after_loading": false, "decrypted_config_file": null, "decrypted_draft_config_file": null, "checkpoint_engine_wait_weights_before_ready": false, "enable_prefill_delayer": false, "prefill_delayer_max_delay_passes": 30, "prefill_delayer_token_usage_low_watermark": null, "prefill_delayer_forward_passes_buckets": null, "prefill_delayer_wait_seconds_buckets": null, "prefill_delayer_queue_min_ratio": null, "prefill_delayer_max_delay_ms": null, "min_free_slots_delay": null, "enable_deterministic_inference": false, "rl_on_policy_target": null, "kv_canary": "none", "kv_canary_real_data": "none", "kv_canary_sweep_interval": 0, "enable_dynamic_batch_tokenizer": false, "dynamic_batch_tokenizer_batch_size": 32, "dynamic_batch_tokenizer_batch_timeout": 0.002, "enable_tokenizer_batch_encode": false, "disable_tokenizer_batch_decode": false, "debug_tensor_dump_output_folder": null, "debug_tensor_dump_layers": null, "debug_tensor_dump_input_file": null, "enable_memory_saver": false, "enable_weights_cpu_backup": false, "enable_draft_weights_cpu_backup": false, "enable_custom_logit_processor": false, "enable_return_hidden_states": false, "enable_return_routed_experts": false, "enable_return_indexer_topk": false, "disable_outlines_disk_cache": false, "enable_mis": false, "weight_cache_mode": "off", "weight_cache_socket": null, "weight_cache_timeout": 1800, "forward_hooks": null, "msprobe_dump_config": null, "status": "ready", "max_total_num_tokens": 440256, "max_req_input_len": 440250, "internal_states": [{"model_path": "/data/hf_models/Kimi-K3", "tokenizer_path": "/data/hf_models/Kimi-K3", "tokenizer_mode": "auto", "tokenizer_backend": "huggingface", "tokenizer_worker_num": 1, "detokenizer_worker_num": 1, "skip_tokenizer_init": false, "load_format": "auto", "model_loader_extra_config": "{}", "trust_remote_code": true, "context_length": null, "is_embedding": false, "enable_multimodal": null, "revision": null, "model_impl": "auto", "model_config_parser": "auto", "json_model_override_args": "{}", "dtype": "auto", "quantization": null, "quantization_param_path": null, "kv_cache_dtype": "auto", "enable_fp32_lm_head": false, "modelopt_quant": null, "modelopt_checkpoint_restore_path": null, "modelopt_checkpoint_save_path": null, "modelopt_export_path": null, "quantize_and_serve": false, "rl_quant_profile": null, "enable_tf32_matmul": false, "mem_fraction_static": 0.88, "max_running_requests": null, "max_queued_requests": null, "max_total_tokens": null, "chunked_prefill_size": 8192, "enable_dynamic_chunking": false, "max_prefill_tokens": 16384, "prefill_max_requests": null, "schedule_policy": "fcfs", "enable_priority_scheduling": false, "disable_priority_preemption": false, "default_priority_value": null, "abort_on_priority_when_disabled": false, "schedule_low_priority_values_first": false, "priority_scheduling_preemption_threshold": 10, "retraction_policy": "length", "schedule_conservativeness": 1.0, "page_size": 64, "swa_full_tokens_ratio": 0.8, "disable_hybrid_swa_memory": false, "radix_eviction_policy": "lru", "prefill_only_disable_kv_cache": false, "disable_radix_cache": true, "enable_page_major_kv_layout": false, "enable_unified_memory": false, "disable_chunked_prefix_cache": true, "disable_overlap_schedule": false, "num_continuous_decode_steps": 1, "scheduler_recv_interval": 1, "enable_mixed_chunk": false, "nccl_port": null, "dist_timeout": 3600, "dist_init_addr": "174.1.60.5:20000", "nnodes": 4, "node_rank": 0, "tp_size": 32, "dcp_size": 1, "pp_size": 1, "pp_max_micro_batch_size": 181, "pp_async_batch_depth": 0, "dp_size": 1, "load_balance_method": "round_robin", "attn_cp_size": 1, "moe_dp_size": 1, "dcp_comm_backend": "ag_rs", "dcp_replicate_q_proj": null, "dwdp_size": 1, "enable_prefill_cp": false, "cp_strategy": null, "enable_dsa_cache_layer_split": false, "enable_dsa_prefill_context_parallel": false, "dsa_prefill_cp_mode": "round-robin-split", "enable_prefill_context_parallel": false, "prefill_cp_mode": "in-seq-split", "enable_cp_decode_attn_tp": false, "enable_dp_attention": false, "enable_dp_attention_local_control_broadcast": false, "enable_dp_lm_head": false, "enable_attn_tp_input_scattered": false, "disable_attn_tp_gather": false, "enable_p2p_check": false, "device": "cuda", "base_gpu_id": 0, "gpu_id_step": 1, "random_seed": 369235288, "watchdog_timeout": 300, "soft_watchdog_timeout": null, "sleep_on_idle": false, "use_ray": false, "numa_node": null, "gc_threshold": null, "host": "0.0.0.0", "port": 30000, "fastapi_root_path": "", "smg_grpc_mode": false, "grpc_mode": false, "grpc_port": null, "sidecar": null, "sidecar_args": null, "skip_server_warmup": false, "warmups": null, "enable_http2": false, "ssl_keyfile": null, "ssl_certfile": null, "ssl_ca_certs": null, "ssl_keyfile_password": null, "enable_ssl_refresh": false, "api_key": null, "admin_api_key": null, "served_model_name": "kimi-k3", "weight_version": "default", "chat_template": null, "hf_chat_template_name": null, "completion_template": null, "file_storage_path": "sglang_storage", "enable_cache_report": false, "reasoning_parser": null, "default_chat_template_kwargs": null, "strip_thinking_cache": false, "enable_strict_thinking": false, "tool_call_parser": null, "tool_server": null, "sampling_defaults": "model", "asr_max_buffer_seconds": 60, "asr_max_concurrent_sessions": 32, "preferred_sampling_params": null, "allow_auto_truncate": false, "stream_interval": 1, "batch_notify_size": 16, "stream_response_default_include_usage": false, "incremental_streaming_output": false, "enable_streaming_session": false, "enable_session_radix_cache": false, "log_level": "info", "log_level_http": null, "log_requests": false, "log_requests_level": 2, "log_requests_format": "text", "log_requests_target": null, "uvicorn_access_log_exclude_prefixes": [], "crash_dump_folder": null, "show_time_cost": false, "enable_metrics": false, "smg_http_sidecar_port": null, "enable_mfu_metrics": false, "enable_metrics_for_all_schedulers": false, "load_snapshot_publish_interval": 15, "tokenizer_metrics_custom_labels_header": "x-custom-labels", "tokenizer_metrics_allowed_custom_labels": null, "extra_metric_labels": null, "bucket_time_to_first_token": null, "bucket_inter_token_latency": null, "bucket_e2e_request_latency": null, "prompt_tokens_buckets": null, "generation_tokens_buckets": null, "gc_warning_threshold_secs": 0.0, "decode_log_interval": 40, "enable_request_time_stats_logging": false, "kv_events_config": null, "enable_forward_pass_metrics": false, "forward_pass_metrics_worker_id": "", "forward_pass_metrics_ipc_name": null, "enable_trace": false, "trace_modules": "request", "otlp_traces_endpoint": "localhost:4317", "export_metrics_to_file": false, "export_metrics_to_file_dir": null, "stat_loggers": null, "constrained_json_whitespace_pattern": null, "constrained_json_disable_any_whitespace": false, "attention_backend": "triton", "decode_attention_backend": null, "prefill_attention_backend": null, "sampling_backend": "flashinfer", "grammar_backend": "xgrammar", "radix_cache_backend": null, "mm_attention_backend": null, "fp8_gemm_runner_backend": "auto", "fp4_gemm_runner_backend": "auto", "bf16_gemm_backend": "auto", "dsa_prefill_backend": null, "dsa_decode_backend": null, "dsa_paged_mqa_logits_backend": "auto", "dsa_topk_backend": "sgl-kernel", "disable_flashinfer_autotune": false, "flashinfer_autotune_skip_ops": null, "mamba_backend": "triton", "cuda_graph_config": {"decode": {"backend": "full", "max_bs": 16, "bs": [1, 2, 4, 8, 12, 16], "tc_compiler": "eager", "full_prefill_max_req": null}, "prefill": {"backend": "disabled", "max_bs": 2048, "bs": [4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], "tc_compiler": "eager", "full_prefill_max_req": null}}, "cuda_graph_backend_decode": null, "cuda_graph_backend_prefill": null, "cuda_graph_max_bs_decode": 16, "cuda_graph_max_bs_prefill": null, "cuda_graph_bs_decode": null, "cuda_graph_bs_prefill": null, "cuda_graph_tc_compiler": null, "disable_prefill_cuda_graph": false, "disable_decode_cuda_graph": false, "disable_cuda_graph": false, "disable_cuda_graph_padding": false, "enable_profile_cuda_graph": false, "enable_cudagraph_gc": false, "debug_cuda_graph": false, "enable_layerwise_nvtx_marker": false, "enable_nccl_nvls": false, "enable_symm_mem": false, "triton_attention_reduce_in_fp32": false, "triton_attention_num_kv_splits": 8, "triton_attention_split_tile_size": null, "flashinfer_mla_disable_ragged": false, "enable_fused_qk_norm_rope": false, "enable_precise_embedding_interpolation": false, "enable_fused_moe_sum_all_reduce": false, "enable_deepseek_v4_fp4_indexer": false, "disable_custom_all_reduce": false, "enable_mscclpp": false, "enable_torch_symm_mem": false, "enable_scattered_sconv": false, "pre_warm_nccl": false, "enable_quant_communications": false, "enable_flashinfer_allreduce_fusion": false, "enforce_disable_flashinfer_allreduce_fusion": false, "flashinfer_allreduce_fusion_backend": null, "enable_aiter_allreduce_fusion": false, "enable_torch_compile": false, "enable_torch_compile_debug_mode": false, "torch_compile_max_bs": 32, "torchao_config": "", "speculative_algorithm": null, "speculative_draft_model_path": null, "speculative_draft_model_revision": null, "speculative_draft_load_format": null, "speculative_num_steps": null, "speculative_eagle_topk": null, "speculative_num_draft_tokens": null, "speculative_dflash_block_size": null, "speculative_dspark_block_size": null, "speculative_dspark_sps_table_path": null, "speculative_dspark_confidence_sts_path": null, "speculative_dspark_align_verify_tokens_to_graph_tier": false, "speculative_accept_threshold_single": 1.0, "speculative_accept_threshold_acc": 1.0, "speculative_use_rejection_sampling": false, "speculative_token_map": null, "speculative_attention_mode": "prefill", "speculative_draft_attention_backend": null, "speculative_draft_window_size": null, "speculative_moe_runner_backend": "marlin", "speculative_moe_a2a_backend": null, "speculative_draft_model_quantization": null, "speculative_skip_dp_mlp_sync": false, "enable_multi_layer_eagle": false, "speculative_adaptive": false, "speculative_adaptive_config": null, "decoupled_spec_bind_endpoint": null, "decoupled_spec_connect_endpoints": null, "decoupled_spec_rank": null, "decoupled_spec_role": "null", "spec_trace_dir": null, "speculative_ngram_min_bfs_breadth": 1, "speculative_ngram_max_bfs_breadth": 10, "speculative_ngram_match_type": "BFS", "speculative_ngram_max_trie_depth": 18, "speculative_ngram_capacity": 10000000, "speculative_ngram_external_corpus_path": null, "speculative_ngram_external_sam_budget": 0, "speculative_ngram_external_corpus_max_tokens": 10000000, "ep_size": 32, "moe_a2a_backend": "none", "moe_runner_backend": "marlin", "flashinfer_mxfp4_moe_precision": "default", "deepep_mode": "auto", "fuseep_mode": 2, "deepep_dispatcher_output_dtype": "auto", "ep_num_redundant_experts": 0, "ep_dispatch_algorithm": null, "init_expert_location": "trivial", "enable_eplb": false, "eplb_algorithm": "auto", "eplb_rebalance_num_iterations": 1000, "eplb_rebalance_layers_per_chunk": null, "eplb_min_rebalancing_utilization_threshold": 1.0, "expert_distribution_recorder_mode": null, "expert_distribution_recorder_buffer_size": 1000, "enable_expert_distribution_metrics": false, "deepep_config": null, "moe_dense_tp_size": null, "elastic_ep_backend": null, "enable_elastic_expert_backup": false, "mooncake_ib_device": null, "enable_waterfill": false, "ep_join_mode": null, "ep_join_rank_offset": 0, "elastic_ep_initial_size": null, "max_ep_size": null, "elastic_ep_scale_timeout": 600, "elastic_ep_rejoin": false, "disable_flashinfer_cutlass_moe_fp4_allgather": false, "disable_shared_experts_fusion": false, "enforce_shared_experts_fusion": false, "max_mamba_cache_size": 181, "mamba_ssm_dtype": null, "mamba_max_states_per_path": -1, "enable_mamba_cache_stochastic_rounding": false, "mamba_cache_philox_rounds": 0, "mamba_full_memory_ratio": 0.21, "mamba_radix_cache_strategy": "extra_buffer_lazy", "uses_mamba_radix_cache": false, "mamba_track_interval": 256, "enable_int8_mamba_checkpoint": false, "int8_mamba_ckpt_size": null, "linear_attn_backend": "triton", "linear_attn_decode_backend": null, "linear_attn_prefill_backend": null, "linear_attn_verify_backend": null, "enable_linear_replayssm": false, "linear_replayssm_cache_len": 16, "enable_linear_replayssm_spec": false, "enable_hierarchical_cache": false, "hicache_ratio": 2.0, "hicache_size": 0, "hicache_write_policy": "write_through", "hicache_io_backend": "kernel", "hicache_mem_layout": "page_first", "hicache_storage_backend": null, "hicache_storage_prefetch_policy": "timeout", "hicache_storage_backend_extra_config": null, "enable_hisparse": false, "hisparse_config": null, "enable_broadcast_mm_inputs_process": false, "enable_prefix_mm_cache": false, "mm_enable_dp_encoder": false, "mm_process_config": {}, "mm_processor_worker_num": 0, "mm_io_worker_num": 0, "limit_mm_data_per_request": null, "enable_mm_global_cache": false, "disable_fast_image_processor": false, "mm_feature_transport": "cpu", "keep_mm_feature_on_device": false, "enable_lora": null, "enable_lora_overlap_loading": null, "max_lora_rank": null, "lora_target_modules": null, "lora_paths": null, "max_loaded_loras": null, "max_loras_per_batch": 8, "lora_eviction_policy": "lru", "lora_backend": "csgmv", "max_lora_chunk_size": 16, "experts_shared_outer_loras": null, "lora_use_virtual_experts": false, "lora_strict_loading": false, "lora_drain_wait_threshold": 0.0, "enable_two_batch_overlap": false, "enable_single_batch_overlap": false, "tbo_token_distribution_threshold": 0.48, "cpu_offload_gb": 0, "offload_group_size": -1, "offload_num_in_group": 1, "offload_prefetch_step": 1, "offload_mode": "cpu", "enable_lmcache": false, "lmcache_config_file": null, "enable_flexkv": false, "flexkv_config_file": null, "kt_weight_path": null, "kt_method": "AMXINT4", "kt_cpuinfer": null, "kt_threadpool_count": 2, "kt_num_gpu_experts": null, "kt_max_deferred_experts_per_token": null, "dllm_algorithm": null, "dllm_algorithm_config": null, "dllm_fdfo": true, "disaggregation_mode": "decode", "disaggregation_transfer_backend": "mooncake", "disaggregation_bootstrap_port": 28800, "disaggregation_ib_device": "mlx5_0,mlx5_1,mlx5_2,mlx5_3", "disaggregation_decode_enable_radix_cache": false, "disaggregation_decode_enable_offload_kvcache": false, "num_reserved_decode_tokens": 512, "disaggregation_decode_extra_slots": 0, "disaggregation_decode_polling_interval": 1, "optimistic_prefill_attempts": 0, "encoder_only": false, "language_only": false, "encoder_transfer_backend": "zmq_to_tokenizer", "encoder_urls": [], "encoder_bootstrap_port": 8997, "encoder_register_urls": [], "enable_adaptive_dispatch_to_encoder": false, "enable_pdmux": false, "pdmux_config_path": null, "sm_group_num": 8, "custom_weight_loader": [], "weight_loader_disable_mmap": false, "weight_loader_prefetch_checkpoints": false, "weight_loader_prefetch_num_threads": 4, "weight_loader_drop_cache_after_load": false, "remote_instance_weight_loader_seed_instance_ip": null, "remote_instance_weight_loader_seed_instance_service_port": null, "remote_instance_weight_loader_send_weights_group_ports": null, "remote_instance_weight_loader_backend": "nccl", "remote_instance_weight_loader_start_seed_via_transfer_engine": false, "engine_info_bootstrap_port": 6789, "modelexpress_config": null, "download_dir": null, "model_checksum": null, "delete_ckpt_after_loading": false, "decrypted_config_file": null, "decrypted_draft_config_file": null, "checkpoint_engine_wait_weights_before_ready": false, "enable_prefill_delayer": false, "prefill_delayer_max_delay_passes": 30, "prefill_delayer_token_usage_low_watermark": null, "prefill_delayer_forward_passes_buckets": null, "prefill_delayer_wait_seconds_buckets": null, "prefill_delayer_queue_min_ratio": null, "prefill_delayer_max_delay_ms": null, "min_free_slots_delay": null, "enable_deterministic_inference": false, "rl_on_policy_target": null, "kv_canary": "none", "kv_canary_real_data": "none", "kv_canary_sweep_interval": 0, "enable_dynamic_batch_tokenizer": false, "dynamic_batch_tokenizer_batch_size": 32, "dynamic_batch_tokenizer_batch_timeout": 0.002, "enable_tokenizer_batch_encode": false, "disable_tokenizer_batch_decode": false, "debug_tensor_dump_output_folder": null, "debug_tensor_dump_layers": null, "debug_tensor_dump_input_file": null, "enable_memory_saver": false, "enable_weights_cpu_backup": false, "enable_draft_weights_cpu_backup": false, "enable_custom_logit_processor": false, "enable_return_hidden_states": false, "enable_return_routed_experts": false, "enable_return_indexer_topk": false, "disable_outlines_disk_cache": false, "enable_mis": false, "weight_cache_mode": "off", "weight_cache_socket": null, "weight_cache_timeout": 1800, "forward_hooks": null, "msprobe_dump_config": null, "_resolved_overrides": [["_sampling_backend_default", {"sampling_backend": "flashinfer"}], ["_attention_backend_default", {"attention_backend": "triton"}], ["_data_parallelism_defaults", {"enable_dp_attention": false, "enable_dp_lm_head": false}], ["_speculative_moe_runner_default", {"speculative_moe_runner_backend": "marlin"}]], "grpc_worker_threads": 4, "_quantization_explicitly_unset": false, "_cuda_graph_config_locked": [["decode", "max_bs"]], "_declarations_materialized": true, "_runtime_mutations": [["model_runner.chunked_prefix_cache_gate", {"disable_chunked_prefix_cache": true}], ["mamba_pool.memory_budget", {"max_mamba_cache_size": 181}]], "_in_override": false, "_mx_config_cache": {}, "max_speculative_num_draft_tokens": null, "last_gen_throughput": 0.09006957280512767, "memory_usage": {"weight": 58.57, "kvcache": 11.34, "token_capacity": 440256, "graph": 0.6}, "effective_max_running_requests_per_dp": 181}], "version": "0.5.16", "kv_events": null}]}, "duration": 60.64697799109854, "completed": 8, "total_input_tokens": 131072, "total_input_text_tokens": 131072, "total_input_vision_tokens": 0, "total_output_tokens": 8, "total_output_tokens_retokenized": 8, "request_throughput": 0.1319109420616836, "input_throughput": 2161.228874738624, "output_throughput": 0.1319109420616836, "total_throughput": 2161.3607856806857, "mean_e2e_latency_ms": 7572.675019968301, "median_e2e_latency_ms": 7565.415543504059, "std_e2e_latency_ms": 24.601867162878474, "p90_e2e_latency_ms": 7597.53510360606, "p95_e2e_latency_ms": 7614.480598270893, "p99_e2e_latency_ms": 7628.0369940027595, "mean_ttft_ms": 7572.641669015866, "median_ttft_ms": 7565.384251996875, "std_ttft_ms": 24.59205444610288, "p90_ttft_ms": 7597.499966225587, "p95_ttft_ms": 7614.432096167002, "p99_ttft_ms": 7627.977800120134, "mean_tpot_ms": 0.0, "median_tpot_ms": 0.0, "std_tpot_ms": 0.0, "p90_tpot_ms": 0.0, "p95_tpot_ms": 0.0, "p99_tpot_ms": 0.0, "mean_itl_ms": 0.0, "median_itl_ms": 0.0, "std_itl_ms": 0.0, "p90_itl_ms": 0.0, "p95_itl_ms": 0.0, "p99_itl_ms": 0.0, "concurrency": 0.9989186958109973, "accept_length": null, "max_output_tokens_per_s": 1.0, "max_concurrent_requests": 2, "input_lens": [16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384], "output_lens": [1, 1, 1, 1, 1, 1, 1, 1], "ttfts": [7.631364226108417, 7.564686966827139, 7.5679764919914305, 7.561987238004804, 7.582986711990088, 7.5648577329702675, 7.541363213211298, 7.565910771023482], "itls": [[], [], [], [], [], [], [], []], "generated_texts": ["\ufffd", "\ufffd", "\ufffd", "\ufffd", "\ufffd", "\ufffd", "\ufffd", "\u7684"], "errors": ["", "", "", "", "", "", "", ""]}
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/bench/prefill_16k_1_c1.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/bench/prefill_16k_1_c1.log
new file mode 100644
index 0000000..582082f
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/bench/prefill_16k_1_c1.log
@@ -0,0 +1,78 @@
+/usr/local/lib/python3.12/dist-packages/torchao/quantization/quant_api.py:1731: SyntaxWarning: invalid escape sequence '\.'
+  """Configuration class for applying different quantization configs to modules or parameters based on their fully qualified names (FQNs).
+/sgl-workspace/sglang/python/sglang/kernels/ops/attention/fla/utils.py:223: UserWarning: Triton is not supported on current platform, roll back to CPU.
+  warnings.warn(
+/sgl-workspace/sglang/python/sglang/srt/layers/quantization/awq/awq.py:52: UserWarning: Only CUDA, HIP and XPU support AWQ currently.
+  warnings.warn(f"Only CUDA, HIP and XPU support AWQ currently.")
+/sgl-workspace/sglang/python/sglang/srt/layers/quantization/gguf.py:69: UserWarning: Only CUDA, MUSA and NPU support GGUF quantization currently.
+  warnings.warn(f"Only CUDA, MUSA and NPU support GGUF quantization currently.")
+benchmark_args=Namespace(backend='sglang', base_url=None, host='174.1.60.5', port=31000, ready_check_timeout_sec=60, dataset_name='random-ids', dataset_path='', dataset_offset=0, agentic_max_turns=None, speed_bench_category=None, speed_bench_output_len=512, model=None, served_model_name=None, tokenizer='/data/hf_models/Kimi-K3', num_prompts=8, sharegpt_output_len=None, sharegpt_context_len=None, random_input_len=16384, random_output_len=1, random_range_ratio=1.0, image_count=1, image_resolution='1080p', random_image_count=False, image_format='jpeg', image_content='random', request_rate=10000.0, use_trace_timestamps=False, max_concurrency=1, output_file='/data/hzy/sskj/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/bench/prefill_16k_1_c1.jsonl', output_details=True, print_requests=False, disable_tqdm=True, disable_stream=False, return_logprob=False, top_logprobs_num=0, token_ids_logprob=None, logprob_start_len=-1, return_routed_experts=False, cache_report=False, seed=42, disable_ignore_eos=False, temperature=0.0, top_p=1.0, extra_request_body=None, apply_chat_template=False, profile=False, plot_throughput=False, profile_activities=['CPU', 'GPU'], profile_start_step=None, profile_steps=None, profile_num_steps=None, profile_by_stage=False, profile_stages=None, profile_output_dir=None, profile_prefix=None, lora_name=None, lora_request_distribution='uniform', lora_zipf_alpha=1.5, prompt_suffix='', pd_separated=False, profile_prefill_url=None, profile_decode_url=None, flush_cache=False, warmup_requests=0, tokenize_prompt=False, gsp_num_groups=64, gsp_prompts_per_group=16, gsp_system_prompt_len=2048, gsp_question_len=128, gsp_output_len=256, gsp_range_ratio=1.0, gsp_fast_prepare=False, gsp_send_routing_key=False, gsp_num_turns=1, gsp_ordered=False, gsp_group_distribution='uniform', gsp_zipf_alpha=None, mooncake_slowdown_factor=1.0, mooncake_num_rounds=1, mooncake_workload='conversation', fake_prefill=False, tag=None, header=None)
+Waiting up to 60s for http://174.1.60.5:31000/v1/models to become ready...
+Server ready in 0.0s.
+'[Errno 101] Network is unreachable' thrown while requesting HEAD https://huggingface.co/kimi-k3/resolve/main/config.json
+Retrying in 1s [Retry 1/5].
+'[Errno 101] Network is unreachable' thrown while requesting HEAD https://huggingface.co/kimi-k3/resolve/main/config.json
+Retrying in 1s [Retry 1/5].
+Fail to load tokenizer config with error=Can't load the configuration of 'kimi-k3'. If you were trying to load it from 'https://huggingface.co/models', make sure you don't have a local directory with the same name. Otherwise, make sure 'kimi-k3' is the correct path to a directory containing a config.json file
+
+WARNING It is recommended to use the `Chat` or `Instruct` model for benchmarking.
+Because when the tokenizer counts the output tokens, if there is gibberish, it might count incorrectly.
+
+Namespace(backend='sglang', base_url=None, host='174.1.60.5', port=31000, ready_check_timeout_sec=60, dataset_name='random-ids', dataset_path='', dataset_offset=0, agentic_max_turns=None, speed_bench_category=None, speed_bench_output_len=512, model='kimi-k3', served_model_name=None, tokenizer='/data/hf_models/Kimi-K3', num_prompts=8, sharegpt_output_len=None, sharegpt_context_len=None, random_input_len=16384, random_output_len=1, random_range_ratio=1.0, image_count=1, image_resolution='1080p', random_image_count=False, image_format='jpeg', image_content='random', request_rate=10000.0, use_trace_timestamps=False, max_concurrency=1, output_file='/data/hzy/sskj/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/bench/prefill_16k_1_c1.jsonl', output_details=True, print_requests=False, disable_tqdm=True, disable_stream=False, return_logprob=False, top_logprobs_num=0, token_ids_logprob=None, logprob_start_len=-1, return_routed_experts=False, cache_report=False, seed=42, disable_ignore_eos=False, temperature=0.0, top_p=1.0, extra_request_body=None, apply_chat_template=False, profile=False, plot_throughput=False, profile_activities=['CPU', 'GPU'], profile_start_step=None, profile_steps=None, profile_num_steps=None, profile_by_stage=False, profile_stages=None, profile_output_dir=None, profile_prefix=None, lora_name=None, lora_request_distribution='uniform', lora_zipf_alpha=1.5, prompt_suffix='', pd_separated=False, profile_prefill_url=None, profile_decode_url=None, flush_cache=False, warmup_requests=0, tokenize_prompt=False, gsp_num_groups=64, gsp_prompts_per_group=16, gsp_system_prompt_len=2048, gsp_question_len=128, gsp_output_len=256, gsp_range_ratio=1.0, gsp_fast_prepare=False, gsp_send_routing_key=False, gsp_num_turns=1, gsp_ordered=False, gsp_group_distribution='uniform', gsp_zipf_alpha=None, mooncake_slowdown_factor=1.0, mooncake_num_rounds=1, mooncake_workload='conversation', fake_prefill=False, tag=None, header=None)
+
+#Input tokens: 131072
+#Output tokens: 8
+Starting warmup with 0 sequences...
+Warmup completed with 0 sequences. Starting main benchmark run...
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+
+============ Serving Benchmark Result ============
+Backend:                                 sglang    
+Traffic request rate:                    10000.0   
+Max request concurrency:                 1         
+Successful requests:                     8         
+Benchmark duration (s):                  60.65     
+Total input tokens:                      131072    
+Total input text tokens:                 131072    
+Total generated tokens:                  8         
+Total generated tokens (retokenized):    8         
+Request throughput (req/s):              0.13      
+Input token throughput (tok/s):          2161.23   
+Output token throughput (tok/s):         0.13      
+Peak output token throughput (tok/s):    1.00      
+Peak concurrent requests:                2         
+Total token throughput (tok/s):          2161.36   
+Concurrency:                             1.00      
+----------------End-to-End Latency----------------
+Mean E2E Latency (ms):                   7572.68   
+Median E2E Latency (ms):                 7565.42   
+P90 E2E Latency (ms):                    7597.54   
+P95 E2E Latency (ms):                    7614.48   
+P99 E2E Latency (ms):                    7628.04   
+---------------Time to First Token----------------
+Mean TTFT (ms):                          7572.64   
+Median TTFT (ms):                        7565.38   
+P90 TTFT (ms):                           7597.50   
+P95 TTFT (ms):                           7614.43   
+P99 TTFT (ms):                           7627.98   
+-----Time per Output Token (excl. 1st token)------
+Mean TPOT (ms):                          0.00      
+Median TPOT (ms):                        0.00      
+P90 TPOT (ms):                           0.00      
+P95 TPOT (ms):                           0.00      
+P99 TPOT (ms):                           0.00      
+---------------Inter-Token Latency----------------
+Mean ITL (ms):                           0.00      
+Median ITL (ms):                         0.00      
+P90 ITL (ms):                            0.00      
+P95 ITL (ms):                            0.00      
+P99 ITL (ms):                            0.00      
+Max ITL (ms):                            0.00      
+==================================================
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/bench/prefill_16k_1_c8.cmd.txt b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/bench/prefill_16k_1_c8.cmd.txt
new file mode 100644
index 0000000..77da5a9
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/bench/prefill_16k_1_c8.cmd.txt
@@ -0,0 +1 @@
+docker run --rm --network host -v /data/hf_models/Kimi-K3:/data/hf_models/Kimi-K3:ro -v /data/hzy/sskj/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310:/data/hzy/sskj/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310 --entrypoint python3 local/sglang:kimi-k3-sm120-flashinfer-mxfp4-pd1 -m sglang.benchmark.serving --backend sglang --host 174.1.60.5 --port 31000 --tokenizer /data/hf_models/Kimi-K3 --dataset-name random-ids --random-input-len 16384 --random-output-len 1 --random-range-ratio 1.0 --num-prompts 40 --max-concurrency 8 --request-rate 10000 --warmup-requests 0 --output-file /data/hzy/sskj/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/bench/prefill_16k_1_c8.jsonl --output-details --disable-tqdm 
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/bench/prefill_16k_1_c8.jsonl b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/bench/prefill_16k_1_c8.jsonl
new file mode 100644
index 0000000..b040ad8
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/bench/prefill_16k_1_c8.jsonl
@@ -0,0 +1 @@
+{"tag": null, "backend": "sglang", "dataset_name": "random-ids", "request_rate": 10000.0, "max_concurrency": 8, "sharegpt_output_len": null, "random_input_len": 16384, "random_output_len": 1, "random_range_ratio": 1.0, "server_info": {"internal_states": [{"model_path": "/data/hf_models/Kimi-K3", "tokenizer_path": "/data/hf_models/Kimi-K3", "tokenizer_mode": "auto", "tokenizer_backend": "huggingface", "tokenizer_worker_num": 1, "detokenizer_worker_num": 1, "skip_tokenizer_init": false, "load_format": "auto", "model_loader_extra_config": "{}", "trust_remote_code": true, "context_length": null, "is_embedding": false, "enable_multimodal": null, "revision": null, "model_impl": "auto", "model_config_parser": "auto", "json_model_override_args": "{}", "dtype": "auto", "quantization": null, "quantization_param_path": null, "kv_cache_dtype": "auto", "enable_fp32_lm_head": false, "modelopt_quant": null, "modelopt_checkpoint_restore_path": null, "modelopt_checkpoint_save_path": null, "modelopt_export_path": null, "quantize_and_serve": false, "rl_quant_profile": null, "enable_tf32_matmul": false, "mem_fraction_static": 0.88, "max_running_requests": null, "max_queued_requests": null, "max_total_tokens": null, "chunked_prefill_size": 8192, "enable_dynamic_chunking": false, "max_prefill_tokens": 16384, "prefill_max_requests": null, "schedule_policy": "fcfs", "enable_priority_scheduling": false, "disable_priority_preemption": false, "default_priority_value": null, "abort_on_priority_when_disabled": false, "schedule_low_priority_values_first": false, "priority_scheduling_preemption_threshold": 10, "retraction_policy": "length", "schedule_conservativeness": 1.0, "page_size": 64, "swa_full_tokens_ratio": 0.8, "disable_hybrid_swa_memory": false, "radix_eviction_policy": "lru", "prefill_only_disable_kv_cache": false, "disable_radix_cache": true, "enable_page_major_kv_layout": false, "enable_unified_memory": false, "disable_chunked_prefix_cache": true, "disable_overlap_schedule": false, "num_continuous_decode_steps": 1, "scheduler_recv_interval": 1, "enable_mixed_chunk": false, "nccl_port": null, "dist_timeout": 3600, "dist_init_addr": "174.1.60.5:20000", "nnodes": 4, "node_rank": 0, "tp_size": 32, "dcp_size": 1, "pp_size": 1, "pp_max_micro_batch_size": 181, "pp_async_batch_depth": 0, "dp_size": 1, "load_balance_method": "round_robin", "attn_cp_size": 1, "moe_dp_size": 1, "dcp_comm_backend": "ag_rs", "dcp_replicate_q_proj": null, "dwdp_size": 1, "enable_prefill_cp": false, "cp_strategy": null, "enable_dsa_cache_layer_split": false, "enable_dsa_prefill_context_parallel": false, "dsa_prefill_cp_mode": "round-robin-split", "enable_prefill_context_parallel": false, "prefill_cp_mode": "in-seq-split", "enable_cp_decode_attn_tp": false, "enable_dp_attention": false, "enable_dp_attention_local_control_broadcast": false, "enable_dp_lm_head": false, "enable_attn_tp_input_scattered": false, "disable_attn_tp_gather": false, "enable_p2p_check": false, "device": "cuda", "base_gpu_id": 0, "gpu_id_step": 1, "random_seed": 369235288, "watchdog_timeout": 300, "soft_watchdog_timeout": null, "sleep_on_idle": false, "use_ray": false, "numa_node": null, "gc_threshold": null, "host": "0.0.0.0", "port": 30000, "fastapi_root_path": "", "smg_grpc_mode": false, "grpc_mode": false, "grpc_port": null, "sidecar": null, "sidecar_args": null, "skip_server_warmup": false, "warmups": null, "enable_http2": false, "ssl_keyfile": null, "ssl_certfile": null, "ssl_ca_certs": null, "ssl_keyfile_password": null, "enable_ssl_refresh": false, "api_key": null, "admin_api_key": null, "served_model_name": "kimi-k3", "weight_version": "default", "chat_template": null, "hf_chat_template_name": null, "completion_template": null, "file_storage_path": "sglang_storage", "enable_cache_report": false, "reasoning_parser": null, "default_chat_template_kwargs": null, "strip_thinking_cache": false, "enable_strict_thinking": false, "tool_call_parser": null, "tool_server": null, "sampling_defaults": "model", "asr_max_buffer_seconds": 60, "asr_max_concurrent_sessions": 32, "preferred_sampling_params": null, "allow_auto_truncate": false, "stream_interval": 1, "batch_notify_size": 16, "stream_response_default_include_usage": false, "incremental_streaming_output": false, "enable_streaming_session": false, "enable_session_radix_cache": false, "log_level": "info", "log_level_http": null, "log_requests": false, "log_requests_level": 2, "log_requests_format": "text", "log_requests_target": null, "uvicorn_access_log_exclude_prefixes": [], "crash_dump_folder": null, "show_time_cost": false, "enable_metrics": false, "smg_http_sidecar_port": null, "enable_mfu_metrics": false, "enable_metrics_for_all_schedulers": false, "load_snapshot_publish_interval": 15, "tokenizer_metrics_custom_labels_header": "x-custom-labels", "tokenizer_metrics_allowed_custom_labels": null, "extra_metric_labels": null, "bucket_time_to_first_token": null, "bucket_inter_token_latency": null, "bucket_e2e_request_latency": null, "prompt_tokens_buckets": null, "generation_tokens_buckets": null, "gc_warning_threshold_secs": 0.0, "decode_log_interval": 40, "enable_request_time_stats_logging": false, "kv_events_config": null, "enable_forward_pass_metrics": false, "forward_pass_metrics_worker_id": "", "forward_pass_metrics_ipc_name": null, "enable_trace": false, "trace_modules": "request", "otlp_traces_endpoint": "localhost:4317", "export_metrics_to_file": false, "export_metrics_to_file_dir": null, "stat_loggers": null, "constrained_json_whitespace_pattern": null, "constrained_json_disable_any_whitespace": false, "attention_backend": "triton", "decode_attention_backend": null, "prefill_attention_backend": null, "sampling_backend": "flashinfer", "grammar_backend": "xgrammar", "radix_cache_backend": null, "mm_attention_backend": null, "fp8_gemm_runner_backend": "auto", "fp4_gemm_runner_backend": "auto", "bf16_gemm_backend": "auto", "dsa_prefill_backend": null, "dsa_decode_backend": null, "dsa_paged_mqa_logits_backend": "auto", "dsa_topk_backend": "sgl-kernel", "disable_flashinfer_autotune": false, "flashinfer_autotune_skip_ops": null, "mamba_backend": "triton", "cuda_graph_config": {"decode": {"backend": "full", "max_bs": 16, "bs": [1, 2, 4, 8, 12, 16], "tc_compiler": "eager", "full_prefill_max_req": null}, "prefill": {"backend": "disabled", "max_bs": 2048, "bs": [4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], "tc_compiler": "eager", "full_prefill_max_req": null}}, "cuda_graph_backend_decode": null, "cuda_graph_backend_prefill": null, "cuda_graph_max_bs_decode": 16, "cuda_graph_max_bs_prefill": null, "cuda_graph_bs_decode": null, "cuda_graph_bs_prefill": null, "cuda_graph_tc_compiler": null, "disable_prefill_cuda_graph": false, "disable_decode_cuda_graph": false, "disable_cuda_graph": false, "disable_cuda_graph_padding": false, "enable_profile_cuda_graph": false, "enable_cudagraph_gc": false, "debug_cuda_graph": false, "enable_layerwise_nvtx_marker": false, "enable_nccl_nvls": false, "enable_symm_mem": false, "triton_attention_reduce_in_fp32": false, "triton_attention_num_kv_splits": 8, "triton_attention_split_tile_size": null, "flashinfer_mla_disable_ragged": false, "enable_fused_qk_norm_rope": false, "enable_precise_embedding_interpolation": false, "enable_fused_moe_sum_all_reduce": false, "enable_deepseek_v4_fp4_indexer": false, "disable_custom_all_reduce": false, "enable_mscclpp": false, "enable_torch_symm_mem": false, "enable_scattered_sconv": false, "pre_warm_nccl": false, "enable_quant_communications": false, "enable_flashinfer_allreduce_fusion": false, "enforce_disable_flashinfer_allreduce_fusion": false, "flashinfer_allreduce_fusion_backend": null, "enable_aiter_allreduce_fusion": false, "enable_torch_compile": false, "enable_torch_compile_debug_mode": false, "torch_compile_max_bs": 32, "torchao_config": "", "speculative_algorithm": null, "speculative_draft_model_path": null, "speculative_draft_model_revision": null, "speculative_draft_load_format": null, "speculative_num_steps": null, "speculative_eagle_topk": null, "speculative_num_draft_tokens": null, "speculative_dflash_block_size": null, "speculative_dspark_block_size": null, "speculative_dspark_sps_table_path": null, "speculative_dspark_confidence_sts_path": null, "speculative_dspark_align_verify_tokens_to_graph_tier": false, "speculative_accept_threshold_single": 1.0, "speculative_accept_threshold_acc": 1.0, "speculative_use_rejection_sampling": false, "speculative_token_map": null, "speculative_attention_mode": "prefill", "speculative_draft_attention_backend": null, "speculative_draft_window_size": null, "speculative_moe_runner_backend": "marlin", "speculative_moe_a2a_backend": null, "speculative_draft_model_quantization": null, "speculative_skip_dp_mlp_sync": false, "enable_multi_layer_eagle": false, "speculative_adaptive": false, "speculative_adaptive_config": null, "decoupled_spec_bind_endpoint": null, "decoupled_spec_connect_endpoints": null, "decoupled_spec_rank": null, "decoupled_spec_role": "null", "spec_trace_dir": null, "speculative_ngram_min_bfs_breadth": 1, "speculative_ngram_max_bfs_breadth": 10, "speculative_ngram_match_type": "BFS", "speculative_ngram_max_trie_depth": 18, "speculative_ngram_capacity": 10000000, "speculative_ngram_external_corpus_path": null, "speculative_ngram_external_sam_budget": 0, "speculative_ngram_external_corpus_max_tokens": 10000000, "ep_size": 32, "moe_a2a_backend": "none", "moe_runner_backend": "marlin", "flashinfer_mxfp4_moe_precision": "default", "deepep_mode": "auto", "fuseep_mode": 2, "deepep_dispatcher_output_dtype": "auto", "ep_num_redundant_experts": 0, "ep_dispatch_algorithm": null, "init_expert_location": "trivial", "enable_eplb": false, "eplb_algorithm": "auto", "eplb_rebalance_num_iterations": 1000, "eplb_rebalance_layers_per_chunk": null, "eplb_min_rebalancing_utilization_threshold": 1.0, "expert_distribution_recorder_mode": null, "expert_distribution_recorder_buffer_size": 1000, "enable_expert_distribution_metrics": false, "deepep_config": null, "moe_dense_tp_size": null, "elastic_ep_backend": null, "enable_elastic_expert_backup": false, "mooncake_ib_device": null, "enable_waterfill": false, "ep_join_mode": null, "ep_join_rank_offset": 0, "elastic_ep_initial_size": null, "max_ep_size": null, "elastic_ep_scale_timeout": 600, "elastic_ep_rejoin": false, "disable_flashinfer_cutlass_moe_fp4_allgather": false, "disable_shared_experts_fusion": false, "enforce_shared_experts_fusion": false, "max_mamba_cache_size": 181, "mamba_ssm_dtype": null, "mamba_max_states_per_path": -1, "enable_mamba_cache_stochastic_rounding": false, "mamba_cache_philox_rounds": 0, "mamba_full_memory_ratio": 0.21, "mamba_radix_cache_strategy": "extra_buffer_lazy", "uses_mamba_radix_cache": false, "mamba_track_interval": 256, "enable_int8_mamba_checkpoint": false, "int8_mamba_ckpt_size": null, "linear_attn_backend": "triton", "linear_attn_decode_backend": null, "linear_attn_prefill_backend": null, "linear_attn_verify_backend": null, "enable_linear_replayssm": false, "linear_replayssm_cache_len": 16, "enable_linear_replayssm_spec": false, "enable_hierarchical_cache": false, "hicache_ratio": 2.0, "hicache_size": 0, "hicache_write_policy": "write_through", "hicache_io_backend": "kernel", "hicache_mem_layout": "page_first", "hicache_storage_backend": null, "hicache_storage_prefetch_policy": "timeout", "hicache_storage_backend_extra_config": null, "enable_hisparse": false, "hisparse_config": null, "enable_broadcast_mm_inputs_process": false, "enable_prefix_mm_cache": false, "mm_enable_dp_encoder": false, "mm_process_config": {}, "mm_processor_worker_num": 0, "mm_io_worker_num": 0, "limit_mm_data_per_request": null, "enable_mm_global_cache": false, "disable_fast_image_processor": false, "mm_feature_transport": "cpu", "keep_mm_feature_on_device": false, "enable_lora": null, "enable_lora_overlap_loading": null, "max_lora_rank": null, "lora_target_modules": null, "lora_paths": null, "max_loaded_loras": null, "max_loras_per_batch": 8, "lora_eviction_policy": "lru", "lora_backend": "csgmv", "max_lora_chunk_size": 16, "experts_shared_outer_loras": null, "lora_use_virtual_experts": false, "lora_strict_loading": false, "lora_drain_wait_threshold": 0.0, "enable_two_batch_overlap": false, "enable_single_batch_overlap": false, "tbo_token_distribution_threshold": 0.48, "cpu_offload_gb": 0, "offload_group_size": -1, "offload_num_in_group": 1, "offload_prefetch_step": 1, "offload_mode": "cpu", "enable_lmcache": false, "lmcache_config_file": null, "enable_flexkv": false, "flexkv_config_file": null, "kt_weight_path": null, "kt_method": "AMXINT4", "kt_cpuinfer": null, "kt_threadpool_count": 2, "kt_num_gpu_experts": null, "kt_max_deferred_experts_per_token": null, "dllm_algorithm": null, "dllm_algorithm_config": null, "dllm_fdfo": true, "disaggregation_mode": "decode", "disaggregation_transfer_backend": "mooncake", "disaggregation_bootstrap_port": 28800, "disaggregation_ib_device": "mlx5_0,mlx5_1,mlx5_2,mlx5_3", "disaggregation_decode_enable_radix_cache": false, "disaggregation_decode_enable_offload_kvcache": false, "num_reserved_decode_tokens": 512, "disaggregation_decode_extra_slots": 0, "disaggregation_decode_polling_interval": 1, "optimistic_prefill_attempts": 0, "encoder_only": false, "language_only": false, "encoder_transfer_backend": "zmq_to_tokenizer", "encoder_urls": [], "encoder_bootstrap_port": 8997, "encoder_register_urls": [], "enable_adaptive_dispatch_to_encoder": false, "enable_pdmux": false, "pdmux_config_path": null, "sm_group_num": 8, "custom_weight_loader": [], "weight_loader_disable_mmap": false, "weight_loader_prefetch_checkpoints": false, "weight_loader_prefetch_num_threads": 4, "weight_loader_drop_cache_after_load": false, "remote_instance_weight_loader_seed_instance_ip": null, "remote_instance_weight_loader_seed_instance_service_port": null, "remote_instance_weight_loader_send_weights_group_ports": null, "remote_instance_weight_loader_backend": "nccl", "remote_instance_weight_loader_start_seed_via_transfer_engine": false, "engine_info_bootstrap_port": 6789, "modelexpress_config": null, "download_dir": null, "model_checksum": null, "delete_ckpt_after_loading": false, "decrypted_config_file": null, "decrypted_draft_config_file": null, "checkpoint_engine_wait_weights_before_ready": false, "enable_prefill_delayer": false, "prefill_delayer_max_delay_passes": 30, "prefill_delayer_token_usage_low_watermark": null, "prefill_delayer_forward_passes_buckets": null, "prefill_delayer_wait_seconds_buckets": null, "prefill_delayer_queue_min_ratio": null, "prefill_delayer_max_delay_ms": null, "min_free_slots_delay": null, "enable_deterministic_inference": false, "rl_on_policy_target": null, "kv_canary": "none", "kv_canary_real_data": "none", "kv_canary_sweep_interval": 0, "enable_dynamic_batch_tokenizer": false, "dynamic_batch_tokenizer_batch_size": 32, "dynamic_batch_tokenizer_batch_timeout": 0.002, "enable_tokenizer_batch_encode": false, "disable_tokenizer_batch_decode": false, "debug_tensor_dump_output_folder": null, "debug_tensor_dump_layers": null, "debug_tensor_dump_input_file": null, "enable_memory_saver": false, "enable_weights_cpu_backup": false, "enable_draft_weights_cpu_backup": false, "enable_custom_logit_processor": false, "enable_return_hidden_states": false, "enable_return_routed_experts": false, "enable_return_indexer_topk": false, "disable_outlines_disk_cache": false, "enable_mis": false, "weight_cache_mode": "off", "weight_cache_socket": null, "weight_cache_timeout": 1800, "forward_hooks": null, "msprobe_dump_config": null, "_resolved_overrides": [["_sampling_backend_default", {"sampling_backend": "flashinfer"}], ["_attention_backend_default", {"attention_backend": "triton"}], ["_data_parallelism_defaults", {"enable_dp_attention": false, "enable_dp_lm_head": false}], ["_speculative_moe_runner_default", {"speculative_moe_runner_backend": "marlin"}]], "grpc_worker_threads": 4, "_quantization_explicitly_unset": false, "_cuda_graph_config_locked": [["decode", "max_bs"]], "_declarations_materialized": true, "_runtime_mutations": [["model_runner.chunked_prefix_cache_gate", {"disable_chunked_prefix_cache": true}], ["mamba_pool.memory_budget", {"max_mamba_cache_size": 181}]], "_in_override": false, "_mx_config_cache": {}, "max_speculative_num_draft_tokens": null, "last_gen_throughput": 0.09006957280512767, "memory_usage": {"weight": 58.57, "kvcache": 11.34, "token_capacity": 440256, "graph": 0.6}, "effective_max_running_requests_per_dp": 181}], "prefill": [{"model_path": "/data/hf_models/Kimi-K3", "tokenizer_path": "/data/hf_models/Kimi-K3", "tokenizer_mode": "auto", "tokenizer_backend": "huggingface", "tokenizer_worker_num": 1, "detokenizer_worker_num": 1, "skip_tokenizer_init": false, "load_format": "auto", "model_loader_extra_config": "{}", "trust_remote_code": true, "context_length": null, "is_embedding": false, "enable_multimodal": null, "revision": null, "model_impl": "auto", "model_config_parser": "auto", "json_model_override_args": "{}", "dtype": "auto", "quantization": null, "quantization_param_path": null, "kv_cache_dtype": "auto", "enable_fp32_lm_head": false, "modelopt_quant": null, "modelopt_checkpoint_restore_path": null, "modelopt_checkpoint_save_path": null, "modelopt_export_path": null, "quantize_and_serve": false, "rl_quant_profile": null, "enable_tf32_matmul": false, "mem_fraction_static": 0.88, "max_running_requests": null, "max_queued_requests": null, "max_total_tokens": null, "chunked_prefill_size": 8192, "enable_dynamic_chunking": false, "max_prefill_tokens": 16384, "prefill_max_requests": null, "schedule_policy": "fcfs", "enable_priority_scheduling": false, "disable_priority_preemption": false, "default_priority_value": null, "abort_on_priority_when_disabled": false, "schedule_low_priority_values_first": false, "priority_scheduling_preemption_threshold": 10, "retraction_policy": "length", "schedule_conservativeness": 1.0, "page_size": 64, "swa_full_tokens_ratio": 0.8, "disable_hybrid_swa_memory": false, "radix_eviction_policy": "lru", "prefill_only_disable_kv_cache": false, "disable_radix_cache": true, "enable_page_major_kv_layout": false, "enable_unified_memory": false, "disable_chunked_prefix_cache": false, "disable_overlap_schedule": true, "num_continuous_decode_steps": 1, "scheduler_recv_interval": 1, "enable_mixed_chunk": false, "nccl_port": null, "dist_timeout": 3600, "dist_init_addr": "174.1.60.1:20000", "nnodes": 4, "node_rank": 0, "tp_size": 4, "dcp_size": 1, "pp_size": 8, "pp_max_micro_batch_size": null, "pp_async_batch_depth": 0, "dp_size": 1, "load_balance_method": "follow_bootstrap_room", "attn_cp_size": 1, "moe_dp_size": 1, "dcp_comm_backend": "ag_rs", "dcp_replicate_q_proj": null, "dwdp_size": 1, "enable_prefill_cp": false, "cp_strategy": null, "enable_dsa_cache_layer_split": false, "enable_dsa_prefill_context_parallel": false, "dsa_prefill_cp_mode": "round-robin-split", "enable_prefill_context_parallel": false, "prefill_cp_mode": "in-seq-split", "enable_cp_decode_attn_tp": false, "enable_dp_attention": false, "enable_dp_attention_local_control_broadcast": false, "enable_dp_lm_head": false, "enable_attn_tp_input_scattered": false, "disable_attn_tp_gather": false, "enable_p2p_check": false, "device": "cuda", "base_gpu_id": 0, "gpu_id_step": 1, "random_seed": 984607766, "watchdog_timeout": 300, "soft_watchdog_timeout": null, "sleep_on_idle": false, "use_ray": false, "custom_sigquit_handler": null, "numa_node": null, "gc_threshold": null, "host": "0.0.0.0", "port": 30000, "fastapi_root_path": "", "smg_grpc_mode": false, "grpc_mode": false, "grpc_port": null, "sidecar": null, "sidecar_args": null, "skip_server_warmup": false, "warmups": null, "enable_http2": false, "ssl_keyfile": null, "ssl_certfile": null, "ssl_ca_certs": null, "ssl_keyfile_password": null, "enable_ssl_refresh": false, "api_key": null, "admin_api_key": null, "served_model_name": "kimi-k3", "weight_version": "default", "chat_template": null, "hf_chat_template_name": null, "completion_template": null, "file_storage_path": "sglang_storage", "enable_cache_report": false, "reasoning_parser": null, "default_chat_template_kwargs": null, "strip_thinking_cache": false, "enable_strict_thinking": false, "tool_call_parser": null, "tool_server": null, "sampling_defaults": "model", "asr_max_buffer_seconds": 60, "asr_max_concurrent_sessions": 32, "preferred_sampling_params": null, "allow_auto_truncate": false, "stream_interval": 1, "batch_notify_size": 16, "stream_response_default_include_usage": false, "incremental_streaming_output": false, "enable_streaming_session": false, "enable_session_radix_cache": false, "log_level": "info", "log_level_http": null, "log_requests": false, "log_requests_level": 2, "log_requests_format": "text", "log_requests_target": null, "uvicorn_access_log_exclude_prefixes": [], "crash_dump_folder": null, "show_time_cost": false, "enable_metrics": false, "smg_http_sidecar_port": null, "enable_mfu_metrics": false, "enable_metrics_for_all_schedulers": false, "load_snapshot_publish_interval": 15, "tokenizer_metrics_custom_labels_header": "x-custom-labels", "tokenizer_metrics_allowed_custom_labels": null, "extra_metric_labels": null, "bucket_time_to_first_token": null, "bucket_inter_token_latency": null, "bucket_e2e_request_latency": null, "prompt_tokens_buckets": null, "generation_tokens_buckets": null, "gc_warning_threshold_secs": 0.0, "decode_log_interval": 40, "enable_request_time_stats_logging": false, "kv_events_config": null, "enable_forward_pass_metrics": false, "forward_pass_metrics_worker_id": "", "forward_pass_metrics_ipc_name": null, "enable_trace": false, "trace_modules": "request", "otlp_traces_endpoint": "localhost:4317", "export_metrics_to_file": false, "export_metrics_to_file_dir": null, "stat_loggers": null, "constrained_json_whitespace_pattern": null, "constrained_json_disable_any_whitespace": false, "attention_backend": "triton", "decode_attention_backend": null, "prefill_attention_backend": null, "sampling_backend": "flashinfer", "grammar_backend": "xgrammar", "radix_cache_backend": null, "mm_attention_backend": null, "fp8_gemm_runner_backend": "auto", "fp4_gemm_runner_backend": "auto", "bf16_gemm_backend": "auto", "dsa_prefill_backend": null, "dsa_decode_backend": null, "dsa_paged_mqa_logits_backend": "auto", "dsa_topk_backend": "sgl-kernel", "disable_flashinfer_autotune": false, "flashinfer_autotune_skip_ops": null, "mamba_backend": "triton", "cuda_graph_config": {"decode": {"backend": "disabled", "max_bs": 16, "bs": [1, 2, 4, 8, 12, 16], "tc_compiler": "eager", "full_prefill_max_req": null}, "prefill": {"backend": "disabled", "max_bs": 2048, "bs": [4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], "tc_compiler": "eager", "full_prefill_max_req": null}}, "cuda_graph_backend_decode": null, "cuda_graph_backend_prefill": null, "cuda_graph_max_bs_decode": 16, "cuda_graph_max_bs_prefill": null, "cuda_graph_bs_decode": null, "cuda_graph_bs_prefill": null, "cuda_graph_tc_compiler": null, "disable_prefill_cuda_graph": false, "disable_decode_cuda_graph": false, "disable_cuda_graph": false, "disable_cuda_graph_padding": false, "enable_profile_cuda_graph": false, "enable_cudagraph_gc": false, "debug_cuda_graph": false, "enable_layerwise_nvtx_marker": false, "enable_nccl_nvls": false, "enable_symm_mem": false, "triton_attention_reduce_in_fp32": false, "triton_attention_num_kv_splits": 8, "triton_attention_split_tile_size": null, "flashinfer_mla_disable_ragged": false, "enable_fused_qk_norm_rope": false, "enable_precise_embedding_interpolation": false, "enable_fused_moe_sum_all_reduce": false, "enable_deepseek_v4_fp4_indexer": false, "disable_custom_all_reduce": false, "enable_mscclpp": false, "enable_torch_symm_mem": false, "enable_scattered_sconv": false, "pre_warm_nccl": false, "enable_quant_communications": false, "enable_flashinfer_allreduce_fusion": false, "enforce_disable_flashinfer_allreduce_fusion": false, "flashinfer_allreduce_fusion_backend": null, "enable_aiter_allreduce_fusion": false, "enable_torch_compile": false, "enable_torch_compile_debug_mode": false, "torch_compile_max_bs": 32, "torchao_config": "", "speculative_algorithm": null, "speculative_draft_model_path": null, "speculative_draft_model_revision": null, "speculative_draft_load_format": null, "speculative_num_steps": null, "speculative_eagle_topk": null, "speculative_num_draft_tokens": null, "speculative_dflash_block_size": null, "speculative_dspark_block_size": null, "speculative_dspark_sps_table_path": null, "speculative_dspark_confidence_sts_path": null, "speculative_dspark_align_verify_tokens_to_graph_tier": false, "speculative_accept_threshold_single": 1.0, "speculative_accept_threshold_acc": 1.0, "speculative_use_rejection_sampling": false, "speculative_token_map": null, "speculative_attention_mode": "prefill", "speculative_draft_attention_backend": null, "speculative_draft_window_size": null, "speculative_moe_runner_backend": "flashinfer_mxfp4", "speculative_moe_a2a_backend": null, "speculative_draft_model_quantization": null, "speculative_skip_dp_mlp_sync": false, "enable_multi_layer_eagle": false, "speculative_adaptive": false, "speculative_adaptive_config": null, "decoupled_spec_bind_endpoint": null, "decoupled_spec_connect_endpoints": null, "decoupled_spec_rank": null, "decoupled_spec_role": "null", "spec_trace_dir": null, "speculative_ngram_min_bfs_breadth": 1, "speculative_ngram_max_bfs_breadth": 10, "speculative_ngram_match_type": "BFS", "speculative_ngram_max_trie_depth": 18, "speculative_ngram_capacity": 10000000, "speculative_ngram_external_corpus_path": null, "speculative_ngram_external_sam_budget": 0, "speculative_ngram_external_corpus_max_tokens": 10000000, "ep_size": 4, "moe_a2a_backend": "none", "moe_runner_backend": "flashinfer_mxfp4", "flashinfer_mxfp4_moe_precision": "default", "deepep_mode": "auto", "fuseep_mode": 2, "deepep_dispatcher_output_dtype": "auto", "ep_num_redundant_experts": 0, "ep_dispatch_algorithm": null, "init_expert_location": "trivial", "enable_eplb": false, "eplb_algorithm": "auto", "eplb_rebalance_num_iterations": 1000, "eplb_rebalance_layers_per_chunk": null, "eplb_min_rebalancing_utilization_threshold": 1.0, "expert_distribution_recorder_mode": null, "expert_distribution_recorder_buffer_size": 1000, "enable_expert_distribution_metrics": false, "deepep_config": null, "moe_dense_tp_size": null, "elastic_ep_backend": null, "enable_elastic_expert_backup": false, "mooncake_ib_device": null, "enable_waterfill": false, "ep_join_mode": null, "ep_join_rank_offset": 0, "elastic_ep_initial_size": null, "max_ep_size": null, "elastic_ep_scale_timeout": 600, "elastic_ep_rejoin": false, "disable_flashinfer_cutlass_moe_fp4_allgather": false, "disable_shared_experts_fusion": false, "enforce_shared_experts_fusion": false, "max_mamba_cache_size": null, "mamba_ssm_dtype": null, "mamba_max_states_per_path": -1, "enable_mamba_cache_stochastic_rounding": false, "mamba_cache_philox_rounds": 0, "mamba_full_memory_ratio": 0.36, "mamba_radix_cache_strategy": "extra_buffer_lazy", "uses_mamba_radix_cache": false, "mamba_track_interval": 256, "enable_int8_mamba_checkpoint": false, "int8_mamba_ckpt_size": null, "linear_attn_backend": "triton", "linear_attn_decode_backend": null, "linear_attn_prefill_backend": null, "linear_attn_verify_backend": null, "enable_linear_replayssm": false, "linear_replayssm_cache_len": 16, "enable_linear_replayssm_spec": false, "enable_hierarchical_cache": false, "hicache_ratio": 2.0, "hicache_size": 0, "hicache_write_policy": "write_through", "hicache_io_backend": "kernel", "hicache_mem_layout": "page_first", "hicache_storage_backend": null, "hicache_storage_prefetch_policy": "timeout", "hicache_storage_backend_extra_config": null, "enable_hisparse": false, "hisparse_config": null, "enable_broadcast_mm_inputs_process": false, "enable_prefix_mm_cache": false, "mm_enable_dp_encoder": false, "mm_process_config": {}, "mm_processor_worker_num": 0, "mm_io_worker_num": 0, "limit_mm_data_per_request": null, "enable_mm_global_cache": false, "disable_fast_image_processor": false, "mm_feature_transport": "cpu", "keep_mm_feature_on_device": false, "enable_lora": null, "enable_lora_overlap_loading": null, "max_lora_rank": null, "lora_target_modules": null, "lora_paths": null, "max_loaded_loras": null, "max_loras_per_batch": 8, "lora_eviction_policy": "lru", "lora_backend": "csgmv", "max_lora_chunk_size": 16, "experts_shared_outer_loras": null, "lora_use_virtual_experts": false, "lora_strict_loading": false, "lora_drain_wait_threshold": 0.0, "enable_two_batch_overlap": false, "enable_single_batch_overlap": false, "tbo_token_distribution_threshold": 0.48, "cpu_offload_gb": 0, "offload_group_size": -1, "offload_num_in_group": 1, "offload_prefetch_step": 1, "offload_mode": "cpu", "enable_lmcache": false, "lmcache_config_file": null, "enable_flexkv": false, "flexkv_config_file": null, "kt_weight_path": null, "kt_method": "AMXINT4", "kt_cpuinfer": null, "kt_threadpool_count": 2, "kt_num_gpu_experts": null, "kt_max_deferred_experts_per_token": null, "dllm_algorithm": null, "dllm_algorithm_config": null, "dllm_fdfo": true, "disaggregation_mode": "prefill", "disaggregation_transfer_backend": "mooncake", "disaggregation_bootstrap_port": 28800, "disaggregation_ib_device": "mlx5_0,mlx5_1,mlx5_2,mlx5_3", "disaggregation_decode_enable_radix_cache": false, "disaggregation_decode_enable_offload_kvcache": false, "num_reserved_decode_tokens": 512, "disaggregation_decode_extra_slots": null, "disaggregation_decode_polling_interval": 1, "optimistic_prefill_attempts": 0, "encoder_only": false, "language_only": false, "encoder_transfer_backend": "zmq_to_tokenizer", "encoder_urls": [], "encoder_bootstrap_port": 8997, "encoder_register_urls": [], "enable_adaptive_dispatch_to_encoder": false, "enable_pdmux": false, "pdmux_config_path": null, "sm_group_num": 8, "custom_weight_loader": [], "weight_loader_disable_mmap": false, "weight_loader_prefetch_checkpoints": false, "weight_loader_prefetch_num_threads": 4, "weight_loader_drop_cache_after_load": false, "remote_instance_weight_loader_seed_instance_ip": null, "remote_instance_weight_loader_seed_instance_service_port": null, "remote_instance_weight_loader_send_weights_group_ports": null, "remote_instance_weight_loader_backend": "nccl", "remote_instance_weight_loader_start_seed_via_transfer_engine": false, "engine_info_bootstrap_port": 6789, "modelexpress_config": null, "download_dir": null, "model_checksum": null, "delete_ckpt_after_loading": false, "decrypted_config_file": null, "decrypted_draft_config_file": null, "checkpoint_engine_wait_weights_before_ready": false, "enable_prefill_delayer": false, "prefill_delayer_max_delay_passes": 30, "prefill_delayer_token_usage_low_watermark": null, "prefill_delayer_forward_passes_buckets": null, "prefill_delayer_wait_seconds_buckets": null, "prefill_delayer_queue_min_ratio": null, "prefill_delayer_max_delay_ms": null, "min_free_slots_delay": null, "enable_deterministic_inference": false, "rl_on_policy_target": null, "kv_canary": "none", "kv_canary_real_data": "none", "kv_canary_sweep_interval": 0, "enable_dynamic_batch_tokenizer": false, "dynamic_batch_tokenizer_batch_size": 32, "dynamic_batch_tokenizer_batch_timeout": 0.002, "enable_tokenizer_batch_encode": false, "disable_tokenizer_batch_decode": false, "debug_tensor_dump_output_folder": null, "debug_tensor_dump_layers": null, "debug_tensor_dump_input_file": null, "enable_memory_saver": false, "enable_weights_cpu_backup": false, "enable_draft_weights_cpu_backup": false, "enable_custom_logit_processor": false, "enable_return_hidden_states": false, "enable_return_routed_experts": false, "enable_return_indexer_topk": false, "disable_outlines_disk_cache": false, "enable_mis": false, "weight_cache_mode": "off", "weight_cache_socket": null, "weight_cache_timeout": 1800, "forward_hooks": null, "msprobe_dump_config": null, "status": "ready", "max_total_num_tokens": 3801408, "max_req_input_len": 1048570, "internal_states": [{"model_path": "/data/hf_models/Kimi-K3", "tokenizer_path": "/data/hf_models/Kimi-K3", "tokenizer_mode": "auto", "tokenizer_backend": "huggingface", "tokenizer_worker_num": 1, "detokenizer_worker_num": 1, "skip_tokenizer_init": false, "load_format": "auto", "model_loader_extra_config": "{}", "trust_remote_code": true, "context_length": null, "is_embedding": false, "enable_multimodal": null, "revision": null, "model_impl": "auto", "model_config_parser": "auto", "json_model_override_args": "{}", "dtype": "auto", "quantization": null, "quantization_param_path": null, "kv_cache_dtype": "auto", "enable_fp32_lm_head": false, "modelopt_quant": null, "modelopt_checkpoint_restore_path": null, "modelopt_checkpoint_save_path": null, "modelopt_export_path": null, "quantize_and_serve": false, "rl_quant_profile": null, "enable_tf32_matmul": false, "mem_fraction_static": 0.88, "max_running_requests": null, "max_queued_requests": null, "max_total_tokens": null, "chunked_prefill_size": 8192, "enable_dynamic_chunking": false, "max_prefill_tokens": 16384, "prefill_max_requests": null, "schedule_policy": "fcfs", "enable_priority_scheduling": false, "disable_priority_preemption": false, "default_priority_value": null, "abort_on_priority_when_disabled": false, "schedule_low_priority_values_first": false, "priority_scheduling_preemption_threshold": 10, "retraction_policy": "length", "schedule_conservativeness": 1.0, "page_size": 64, "swa_full_tokens_ratio": 0.8, "disable_hybrid_swa_memory": false, "radix_eviction_policy": "lru", "prefill_only_disable_kv_cache": false, "disable_radix_cache": true, "enable_page_major_kv_layout": false, "enable_unified_memory": false, "disable_chunked_prefix_cache": true, "disable_overlap_schedule": true, "num_continuous_decode_steps": 1, "scheduler_recv_interval": 1, "enable_mixed_chunk": false, "nccl_port": null, "dist_timeout": 3600, "dist_init_addr": "174.1.60.1:20000", "nnodes": 4, "node_rank": 0, "tp_size": 4, "dcp_size": 1, "pp_size": 8, "pp_max_micro_batch_size": 6, "pp_async_batch_depth": 0, "dp_size": 1, "load_balance_method": "follow_bootstrap_room", "attn_cp_size": 1, "moe_dp_size": 1, "dcp_comm_backend": "ag_rs", "dcp_replicate_q_proj": null, "dwdp_size": 1, "enable_prefill_cp": false, "cp_strategy": null, "enable_dsa_cache_layer_split": false, "enable_dsa_prefill_context_parallel": false, "dsa_prefill_cp_mode": "round-robin-split", "enable_prefill_context_parallel": false, "prefill_cp_mode": "in-seq-split", "enable_cp_decode_attn_tp": false, "enable_dp_attention": false, "enable_dp_attention_local_control_broadcast": false, "enable_dp_lm_head": false, "enable_attn_tp_input_scattered": false, "disable_attn_tp_gather": false, "enable_p2p_check": false, "device": "cuda", "base_gpu_id": 0, "gpu_id_step": 1, "random_seed": 984607766, "watchdog_timeout": 300, "soft_watchdog_timeout": null, "sleep_on_idle": false, "use_ray": false, "numa_node": null, "gc_threshold": null, "host": "0.0.0.0", "port": 30000, "fastapi_root_path": "", "smg_grpc_mode": false, "grpc_mode": false, "grpc_port": null, "sidecar": null, "sidecar_args": null, "skip_server_warmup": false, "warmups": null, "enable_http2": false, "ssl_keyfile": null, "ssl_certfile": null, "ssl_ca_certs": null, "ssl_keyfile_password": null, "enable_ssl_refresh": false, "api_key": null, "admin_api_key": null, "served_model_name": "kimi-k3", "weight_version": "default", "chat_template": null, "hf_chat_template_name": null, "completion_template": null, "file_storage_path": "sglang_storage", "enable_cache_report": false, "reasoning_parser": null, "default_chat_template_kwargs": null, "strip_thinking_cache": false, "enable_strict_thinking": false, "tool_call_parser": null, "tool_server": null, "sampling_defaults": "model", "asr_max_buffer_seconds": 60, "asr_max_concurrent_sessions": 32, "preferred_sampling_params": null, "allow_auto_truncate": false, "stream_interval": 1, "batch_notify_size": 16, "stream_response_default_include_usage": false, "incremental_streaming_output": false, "enable_streaming_session": false, "enable_session_radix_cache": false, "log_level": "info", "log_level_http": null, "log_requests": false, "log_requests_level": 2, "log_requests_format": "text", "log_requests_target": null, "uvicorn_access_log_exclude_prefixes": [], "crash_dump_folder": null, "show_time_cost": false, "enable_metrics": false, "smg_http_sidecar_port": null, "enable_mfu_metrics": false, "enable_metrics_for_all_schedulers": false, "load_snapshot_publish_interval": 15, "tokenizer_metrics_custom_labels_header": "x-custom-labels", "tokenizer_metrics_allowed_custom_labels": null, "extra_metric_labels": null, "bucket_time_to_first_token": null, "bucket_inter_token_latency": null, "bucket_e2e_request_latency": null, "prompt_tokens_buckets": null, "generation_tokens_buckets": null, "gc_warning_threshold_secs": 0.0, "decode_log_interval": 40, "enable_request_time_stats_logging": false, "kv_events_config": null, "enable_forward_pass_metrics": false, "forward_pass_metrics_worker_id": "", "forward_pass_metrics_ipc_name": null, "enable_trace": false, "trace_modules": "request", "otlp_traces_endpoint": "localhost:4317", "export_metrics_to_file": false, "export_metrics_to_file_dir": null, "stat_loggers": null, "constrained_json_whitespace_pattern": null, "constrained_json_disable_any_whitespace": false, "attention_backend": "triton", "decode_attention_backend": null, "prefill_attention_backend": null, "sampling_backend": "flashinfer", "grammar_backend": "xgrammar", "radix_cache_backend": null, "mm_attention_backend": null, "fp8_gemm_runner_backend": "auto", "fp4_gemm_runner_backend": "auto", "bf16_gemm_backend": "auto", "dsa_prefill_backend": null, "dsa_decode_backend": null, "dsa_paged_mqa_logits_backend": "auto", "dsa_topk_backend": "sgl-kernel", "disable_flashinfer_autotune": false, "flashinfer_autotune_skip_ops": null, "mamba_backend": "triton", "cuda_graph_config": {"decode": {"backend": "disabled", "max_bs": 16, "bs": [1, 2, 4, 8, 12, 16], "tc_compiler": "eager", "full_prefill_max_req": null}, "prefill": {"backend": "disabled", "max_bs": 2048, "bs": [4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], "tc_compiler": "eager", "full_prefill_max_req": null}}, "cuda_graph_backend_decode": null, "cuda_graph_backend_prefill": null, "cuda_graph_max_bs_decode": 16, "cuda_graph_max_bs_prefill": null, "cuda_graph_bs_decode": null, "cuda_graph_bs_prefill": null, "cuda_graph_tc_compiler": null, "disable_prefill_cuda_graph": false, "disable_decode_cuda_graph": false, "disable_cuda_graph": false, "disable_cuda_graph_padding": false, "enable_profile_cuda_graph": false, "enable_cudagraph_gc": false, "debug_cuda_graph": false, "enable_layerwise_nvtx_marker": false, "enable_nccl_nvls": false, "enable_symm_mem": false, "triton_attention_reduce_in_fp32": false, "triton_attention_num_kv_splits": 8, "triton_attention_split_tile_size": null, "flashinfer_mla_disable_ragged": false, "enable_fused_qk_norm_rope": false, "enable_precise_embedding_interpolation": false, "enable_fused_moe_sum_all_reduce": false, "enable_deepseek_v4_fp4_indexer": false, "disable_custom_all_reduce": false, "enable_mscclpp": false, "enable_torch_symm_mem": false, "enable_scattered_sconv": false, "pre_warm_nccl": false, "enable_quant_communications": false, "enable_flashinfer_allreduce_fusion": false, "enforce_disable_flashinfer_allreduce_fusion": false, "flashinfer_allreduce_fusion_backend": null, "enable_aiter_allreduce_fusion": false, "enable_torch_compile": false, "enable_torch_compile_debug_mode": false, "torch_compile_max_bs": 32, "torchao_config": "", "speculative_algorithm": null, "speculative_draft_model_path": null, "speculative_draft_model_revision": null, "speculative_draft_load_format": null, "speculative_num_steps": null, "speculative_eagle_topk": null, "speculative_num_draft_tokens": null, "speculative_dflash_block_size": null, "speculative_dspark_block_size": null, "speculative_dspark_sps_table_path": null, "speculative_dspark_confidence_sts_path": null, "speculative_dspark_align_verify_tokens_to_graph_tier": false, "speculative_accept_threshold_single": 1.0, "speculative_accept_threshold_acc": 1.0, "speculative_use_rejection_sampling": false, "speculative_token_map": null, "speculative_attention_mode": "prefill", "speculative_draft_attention_backend": null, "speculative_draft_window_size": null, "speculative_moe_runner_backend": "flashinfer_mxfp4", "speculative_moe_a2a_backend": null, "speculative_draft_model_quantization": null, "speculative_skip_dp_mlp_sync": false, "enable_multi_layer_eagle": false, "speculative_adaptive": false, "speculative_adaptive_config": null, "decoupled_spec_bind_endpoint": null, "decoupled_spec_connect_endpoints": null, "decoupled_spec_rank": null, "decoupled_spec_role": "null", "spec_trace_dir": null, "speculative_ngram_min_bfs_breadth": 1, "speculative_ngram_max_bfs_breadth": 10, "speculative_ngram_match_type": "BFS", "speculative_ngram_max_trie_depth": 18, "speculative_ngram_capacity": 10000000, "speculative_ngram_external_corpus_path": null, "speculative_ngram_external_sam_budget": 0, "speculative_ngram_external_corpus_max_tokens": 10000000, "ep_size": 4, "moe_a2a_backend": "none", "moe_runner_backend": "flashinfer_mxfp4", "flashinfer_mxfp4_moe_precision": "default", "deepep_mode": "auto", "fuseep_mode": 2, "deepep_dispatcher_output_dtype": "auto", "ep_num_redundant_experts": 0, "ep_dispatch_algorithm": null, "init_expert_location": "trivial", "enable_eplb": false, "eplb_algorithm": "auto", "eplb_rebalance_num_iterations": 1000, "eplb_rebalance_layers_per_chunk": null, "eplb_min_rebalancing_utilization_threshold": 1.0, "expert_distribution_recorder_mode": null, "expert_distribution_recorder_buffer_size": 1000, "enable_expert_distribution_metrics": false, "deepep_config": null, "moe_dense_tp_size": null, "elastic_ep_backend": null, "enable_elastic_expert_backup": false, "mooncake_ib_device": null, "enable_waterfill": false, "ep_join_mode": null, "ep_join_rank_offset": 0, "elastic_ep_initial_size": null, "max_ep_size": null, "elastic_ep_scale_timeout": 600, "elastic_ep_rejoin": false, "disable_flashinfer_cutlass_moe_fp4_allgather": false, "disable_shared_experts_fusion": false, "enforce_shared_experts_fusion": false, "max_mamba_cache_size": 54, "mamba_ssm_dtype": null, "mamba_max_states_per_path": -1, "enable_mamba_cache_stochastic_rounding": false, "mamba_cache_philox_rounds": 0, "mamba_full_memory_ratio": 0.36, "mamba_radix_cache_strategy": "extra_buffer_lazy", "uses_mamba_radix_cache": false, "mamba_track_interval": 256, "enable_int8_mamba_checkpoint": false, "int8_mamba_ckpt_size": null, "linear_attn_backend": "triton", "linear_attn_decode_backend": null, "linear_attn_prefill_backend": null, "linear_attn_verify_backend": null, "enable_linear_replayssm": false, "linear_replayssm_cache_len": 16, "enable_linear_replayssm_spec": false, "enable_hierarchical_cache": false, "hicache_ratio": 2.0, "hicache_size": 0, "hicache_write_policy": "write_through", "hicache_io_backend": "kernel", "hicache_mem_layout": "page_first", "hicache_storage_backend": null, "hicache_storage_prefetch_policy": "timeout", "hicache_storage_backend_extra_config": null, "enable_hisparse": false, "hisparse_config": null, "enable_broadcast_mm_inputs_process": false, "enable_prefix_mm_cache": false, "mm_enable_dp_encoder": false, "mm_process_config": {}, "mm_processor_worker_num": 0, "mm_io_worker_num": 0, "limit_mm_data_per_request": null, "enable_mm_global_cache": false, "disable_fast_image_processor": false, "mm_feature_transport": "cpu", "keep_mm_feature_on_device": false, "enable_lora": null, "enable_lora_overlap_loading": null, "max_lora_rank": null, "lora_target_modules": null, "lora_paths": null, "max_loaded_loras": null, "max_loras_per_batch": 8, "lora_eviction_policy": "lru", "lora_backend": "csgmv", "max_lora_chunk_size": 16, "experts_shared_outer_loras": null, "lora_use_virtual_experts": false, "lora_strict_loading": false, "lora_drain_wait_threshold": 0.0, "enable_two_batch_overlap": false, "enable_single_batch_overlap": false, "tbo_token_distribution_threshold": 0.48, "cpu_offload_gb": 0, "offload_group_size": -1, "offload_num_in_group": 1, "offload_prefetch_step": 1, "offload_mode": "cpu", "enable_lmcache": false, "lmcache_config_file": null, "enable_flexkv": false, "flexkv_config_file": null, "kt_weight_path": null, "kt_method": "AMXINT4", "kt_cpuinfer": null, "kt_threadpool_count": 2, "kt_num_gpu_experts": null, "kt_max_deferred_experts_per_token": null, "dllm_algorithm": null, "dllm_algorithm_config": null, "dllm_fdfo": true, "disaggregation_mode": "prefill", "disaggregation_transfer_backend": "mooncake", "disaggregation_bootstrap_port": 28800, "disaggregation_ib_device": "mlx5_0,mlx5_1,mlx5_2,mlx5_3", "disaggregation_decode_enable_radix_cache": false, "disaggregation_decode_enable_offload_kvcache": false, "num_reserved_decode_tokens": 512, "disaggregation_decode_extra_slots": null, "disaggregation_decode_polling_interval": 1, "optimistic_prefill_attempts": 0, "encoder_only": false, "language_only": false, "encoder_transfer_backend": "zmq_to_tokenizer", "encoder_urls": [], "encoder_bootstrap_port": 8997, "encoder_register_urls": [], "enable_adaptive_dispatch_to_encoder": false, "enable_pdmux": false, "pdmux_config_path": null, "sm_group_num": 8, "custom_weight_loader": [], "weight_loader_disable_mmap": false, "weight_loader_prefetch_checkpoints": false, "weight_loader_prefetch_num_threads": 4, "weight_loader_drop_cache_after_load": false, "remote_instance_weight_loader_seed_instance_ip": null, "remote_instance_weight_loader_seed_instance_service_port": null, "remote_instance_weight_loader_send_weights_group_ports": null, "remote_instance_weight_loader_backend": "nccl", "remote_instance_weight_loader_start_seed_via_transfer_engine": false, "engine_info_bootstrap_port": 6789, "modelexpress_config": null, "download_dir": null, "model_checksum": null, "delete_ckpt_after_loading": false, "decrypted_config_file": null, "decrypted_draft_config_file": null, "checkpoint_engine_wait_weights_before_ready": false, "enable_prefill_delayer": false, "prefill_delayer_max_delay_passes": 30, "prefill_delayer_token_usage_low_watermark": null, "prefill_delayer_forward_passes_buckets": null, "prefill_delayer_wait_seconds_buckets": null, "prefill_delayer_queue_min_ratio": null, "prefill_delayer_max_delay_ms": null, "min_free_slots_delay": null, "enable_deterministic_inference": false, "rl_on_policy_target": null, "kv_canary": "none", "kv_canary_real_data": "none", "kv_canary_sweep_interval": 0, "enable_dynamic_batch_tokenizer": false, "dynamic_batch_tokenizer_batch_size": 32, "dynamic_batch_tokenizer_batch_timeout": 0.002, "enable_tokenizer_batch_encode": false, "disable_tokenizer_batch_decode": false, "debug_tensor_dump_output_folder": null, "debug_tensor_dump_layers": null, "debug_tensor_dump_input_file": null, "enable_memory_saver": false, "enable_weights_cpu_backup": false, "enable_draft_weights_cpu_backup": false, "enable_custom_logit_processor": false, "enable_return_hidden_states": false, "enable_return_routed_experts": false, "enable_return_indexer_topk": false, "disable_outlines_disk_cache": false, "enable_mis": false, "weight_cache_mode": "off", "weight_cache_socket": null, "weight_cache_timeout": 1800, "forward_hooks": null, "msprobe_dump_config": null, "_resolved_overrides": [["_sampling_backend_default", {"sampling_backend": "flashinfer"}], ["_attention_backend_default", {"attention_backend": "triton"}], ["_data_parallelism_defaults", {"enable_dp_attention": false, "enable_dp_lm_head": false}], ["_pipeline_parallel_overlap_disable", {"disable_overlap_schedule": true}], ["_speculative_moe_runner_default", {"speculative_moe_runner_backend": "flashinfer_mxfp4"}]], "grpc_worker_threads": 4, "_quantization_explicitly_unset": false, "_cuda_graph_config_locked": [["decode", "max_bs"]], "_declarations_materialized": true, "_runtime_mutations": [["model_runner.chunked_prefix_cache_gate", {"disable_chunked_prefix_cache": true}], ["mamba_pool.memory_budget", {"max_mamba_cache_size": 54}]], "_in_override": false, "_mx_config_cache": {}, "max_speculative_num_draft_tokens": null, "last_gen_throughput": 0.0, "memory_usage": {"weight": 43.09, "kvcache": 8.16, "token_capacity": 3801408, "graph": 0}, "effective_max_running_requests_per_dp": 54}], "version": "0.5.16", "kv_events": null}], "decode": [{"model_path": "/data/hf_models/Kimi-K3", "tokenizer_path": "/data/hf_models/Kimi-K3", "tokenizer_mode": "auto", "tokenizer_backend": "huggingface", "tokenizer_worker_num": 1, "detokenizer_worker_num": 1, "skip_tokenizer_init": false, "load_format": "auto", "model_loader_extra_config": "{}", "trust_remote_code": true, "context_length": null, "is_embedding": false, "enable_multimodal": null, "revision": null, "model_impl": "auto", "model_config_parser": "auto", "json_model_override_args": "{}", "dtype": "auto", "quantization": null, "quantization_param_path": null, "kv_cache_dtype": "auto", "enable_fp32_lm_head": false, "modelopt_quant": null, "modelopt_checkpoint_restore_path": null, "modelopt_checkpoint_save_path": null, "modelopt_export_path": null, "quantize_and_serve": false, "rl_quant_profile": null, "enable_tf32_matmul": false, "mem_fraction_static": 0.88, "max_running_requests": null, "max_queued_requests": null, "max_total_tokens": null, "chunked_prefill_size": 8192, "enable_dynamic_chunking": false, "max_prefill_tokens": 16384, "prefill_max_requests": null, "schedule_policy": "fcfs", "enable_priority_scheduling": false, "disable_priority_preemption": false, "default_priority_value": null, "abort_on_priority_when_disabled": false, "schedule_low_priority_values_first": false, "priority_scheduling_preemption_threshold": 10, "retraction_policy": "length", "schedule_conservativeness": 1.0, "page_size": 64, "swa_full_tokens_ratio": 0.8, "disable_hybrid_swa_memory": false, "radix_eviction_policy": "lru", "prefill_only_disable_kv_cache": false, "disable_radix_cache": true, "enable_page_major_kv_layout": false, "enable_unified_memory": false, "disable_chunked_prefix_cache": false, "disable_overlap_schedule": false, "num_continuous_decode_steps": 1, "scheduler_recv_interval": 1, "enable_mixed_chunk": false, "nccl_port": null, "dist_timeout": 3600, "dist_init_addr": "174.1.60.5:20000", "nnodes": 4, "node_rank": 0, "tp_size": 32, "dcp_size": 1, "pp_size": 1, "pp_max_micro_batch_size": null, "pp_async_batch_depth": 0, "dp_size": 1, "load_balance_method": "round_robin", "attn_cp_size": 1, "moe_dp_size": 1, "dcp_comm_backend": "ag_rs", "dcp_replicate_q_proj": null, "dwdp_size": 1, "enable_prefill_cp": false, "cp_strategy": null, "enable_dsa_cache_layer_split": false, "enable_dsa_prefill_context_parallel": false, "dsa_prefill_cp_mode": "round-robin-split", "enable_prefill_context_parallel": false, "prefill_cp_mode": "in-seq-split", "enable_cp_decode_attn_tp": false, "enable_dp_attention": false, "enable_dp_attention_local_control_broadcast": false, "enable_dp_lm_head": false, "enable_attn_tp_input_scattered": false, "disable_attn_tp_gather": false, "enable_p2p_check": false, "device": "cuda", "base_gpu_id": 0, "gpu_id_step": 1, "random_seed": 369235288, "watchdog_timeout": 300, "soft_watchdog_timeout": null, "sleep_on_idle": false, "use_ray": false, "custom_sigquit_handler": null, "numa_node": null, "gc_threshold": null, "host": "0.0.0.0", "port": 30000, "fastapi_root_path": "", "smg_grpc_mode": false, "grpc_mode": false, "grpc_port": null, "sidecar": null, "sidecar_args": null, "skip_server_warmup": false, "warmups": null, "enable_http2": false, "ssl_keyfile": null, "ssl_certfile": null, "ssl_ca_certs": null, "ssl_keyfile_password": null, "enable_ssl_refresh": false, "api_key": null, "admin_api_key": null, "served_model_name": "kimi-k3", "weight_version": "default", "chat_template": null, "hf_chat_template_name": null, "completion_template": null, "file_storage_path": "sglang_storage", "enable_cache_report": false, "reasoning_parser": null, "default_chat_template_kwargs": null, "strip_thinking_cache": false, "enable_strict_thinking": false, "tool_call_parser": null, "tool_server": null, "sampling_defaults": "model", "asr_max_buffer_seconds": 60, "asr_max_concurrent_sessions": 32, "preferred_sampling_params": null, "allow_auto_truncate": false, "stream_interval": 1, "batch_notify_size": 16, "stream_response_default_include_usage": false, "incremental_streaming_output": false, "enable_streaming_session": false, "enable_session_radix_cache": false, "log_level": "info", "log_level_http": null, "log_requests": false, "log_requests_level": 2, "log_requests_format": "text", "log_requests_target": null, "uvicorn_access_log_exclude_prefixes": [], "crash_dump_folder": null, "show_time_cost": false, "enable_metrics": false, "smg_http_sidecar_port": null, "enable_mfu_metrics": false, "enable_metrics_for_all_schedulers": false, "load_snapshot_publish_interval": 15, "tokenizer_metrics_custom_labels_header": "x-custom-labels", "tokenizer_metrics_allowed_custom_labels": null, "extra_metric_labels": null, "bucket_time_to_first_token": null, "bucket_inter_token_latency": null, "bucket_e2e_request_latency": null, "prompt_tokens_buckets": null, "generation_tokens_buckets": null, "gc_warning_threshold_secs": 0.0, "decode_log_interval": 40, "enable_request_time_stats_logging": false, "kv_events_config": null, "enable_forward_pass_metrics": false, "forward_pass_metrics_worker_id": "", "forward_pass_metrics_ipc_name": null, "enable_trace": false, "trace_modules": "request", "otlp_traces_endpoint": "localhost:4317", "export_metrics_to_file": false, "export_metrics_to_file_dir": null, "stat_loggers": null, "constrained_json_whitespace_pattern": null, "constrained_json_disable_any_whitespace": false, "attention_backend": "triton", "decode_attention_backend": null, "prefill_attention_backend": null, "sampling_backend": "flashinfer", "grammar_backend": "xgrammar", "radix_cache_backend": null, "mm_attention_backend": null, "fp8_gemm_runner_backend": "auto", "fp4_gemm_runner_backend": "auto", "bf16_gemm_backend": "auto", "dsa_prefill_backend": null, "dsa_decode_backend": null, "dsa_paged_mqa_logits_backend": "auto", "dsa_topk_backend": "sgl-kernel", "disable_flashinfer_autotune": false, "flashinfer_autotune_skip_ops": null, "mamba_backend": "triton", "cuda_graph_config": {"decode": {"backend": "full", "max_bs": 16, "bs": [1, 2, 4, 8, 12, 16], "tc_compiler": "eager", "full_prefill_max_req": null}, "prefill": {"backend": "disabled", "max_bs": 2048, "bs": [4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], "tc_compiler": "eager", "full_prefill_max_req": null}}, "cuda_graph_backend_decode": null, "cuda_graph_backend_prefill": null, "cuda_graph_max_bs_decode": 16, "cuda_graph_max_bs_prefill": null, "cuda_graph_bs_decode": null, "cuda_graph_bs_prefill": null, "cuda_graph_tc_compiler": null, "disable_prefill_cuda_graph": false, "disable_decode_cuda_graph": false, "disable_cuda_graph": false, "disable_cuda_graph_padding": false, "enable_profile_cuda_graph": false, "enable_cudagraph_gc": false, "debug_cuda_graph": false, "enable_layerwise_nvtx_marker": false, "enable_nccl_nvls": false, "enable_symm_mem": false, "triton_attention_reduce_in_fp32": false, "triton_attention_num_kv_splits": 8, "triton_attention_split_tile_size": null, "flashinfer_mla_disable_ragged": false, "enable_fused_qk_norm_rope": false, "enable_precise_embedding_interpolation": false, "enable_fused_moe_sum_all_reduce": false, "enable_deepseek_v4_fp4_indexer": false, "disable_custom_all_reduce": false, "enable_mscclpp": false, "enable_torch_symm_mem": false, "enable_scattered_sconv": false, "pre_warm_nccl": false, "enable_quant_communications": false, "enable_flashinfer_allreduce_fusion": false, "enforce_disable_flashinfer_allreduce_fusion": false, "flashinfer_allreduce_fusion_backend": null, "enable_aiter_allreduce_fusion": false, "enable_torch_compile": false, "enable_torch_compile_debug_mode": false, "torch_compile_max_bs": 32, "torchao_config": "", "speculative_algorithm": null, "speculative_draft_model_path": null, "speculative_draft_model_revision": null, "speculative_draft_load_format": null, "speculative_num_steps": null, "speculative_eagle_topk": null, "speculative_num_draft_tokens": null, "speculative_dflash_block_size": null, "speculative_dspark_block_size": null, "speculative_dspark_sps_table_path": null, "speculative_dspark_confidence_sts_path": null, "speculative_dspark_align_verify_tokens_to_graph_tier": false, "speculative_accept_threshold_single": 1.0, "speculative_accept_threshold_acc": 1.0, "speculative_use_rejection_sampling": false, "speculative_token_map": null, "speculative_attention_mode": "prefill", "speculative_draft_attention_backend": null, "speculative_draft_window_size": null, "speculative_moe_runner_backend": "marlin", "speculative_moe_a2a_backend": null, "speculative_draft_model_quantization": null, "speculative_skip_dp_mlp_sync": false, "enable_multi_layer_eagle": false, "speculative_adaptive": false, "speculative_adaptive_config": null, "decoupled_spec_bind_endpoint": null, "decoupled_spec_connect_endpoints": null, "decoupled_spec_rank": null, "decoupled_spec_role": "null", "spec_trace_dir": null, "speculative_ngram_min_bfs_breadth": 1, "speculative_ngram_max_bfs_breadth": 10, "speculative_ngram_match_type": "BFS", "speculative_ngram_max_trie_depth": 18, "speculative_ngram_capacity": 10000000, "speculative_ngram_external_corpus_path": null, "speculative_ngram_external_sam_budget": 0, "speculative_ngram_external_corpus_max_tokens": 10000000, "ep_size": 32, "moe_a2a_backend": "none", "moe_runner_backend": "marlin", "flashinfer_mxfp4_moe_precision": "default", "deepep_mode": "auto", "fuseep_mode": 2, "deepep_dispatcher_output_dtype": "auto", "ep_num_redundant_experts": 0, "ep_dispatch_algorithm": null, "init_expert_location": "trivial", "enable_eplb": false, "eplb_algorithm": "auto", "eplb_rebalance_num_iterations": 1000, "eplb_rebalance_layers_per_chunk": null, "eplb_min_rebalancing_utilization_threshold": 1.0, "expert_distribution_recorder_mode": null, "expert_distribution_recorder_buffer_size": 1000, "enable_expert_distribution_metrics": false, "deepep_config": null, "moe_dense_tp_size": null, "elastic_ep_backend": null, "enable_elastic_expert_backup": false, "mooncake_ib_device": null, "enable_waterfill": false, "ep_join_mode": null, "ep_join_rank_offset": 0, "elastic_ep_initial_size": null, "max_ep_size": null, "elastic_ep_scale_timeout": 600, "elastic_ep_rejoin": false, "disable_flashinfer_cutlass_moe_fp4_allgather": false, "disable_shared_experts_fusion": false, "enforce_shared_experts_fusion": false, "max_mamba_cache_size": null, "mamba_ssm_dtype": null, "mamba_max_states_per_path": -1, "enable_mamba_cache_stochastic_rounding": false, "mamba_cache_philox_rounds": 0, "mamba_full_memory_ratio": 0.21, "mamba_radix_cache_strategy": "extra_buffer_lazy", "uses_mamba_radix_cache": false, "mamba_track_interval": 256, "enable_int8_mamba_checkpoint": false, "int8_mamba_ckpt_size": null, "linear_attn_backend": "triton", "linear_attn_decode_backend": null, "linear_attn_prefill_backend": null, "linear_attn_verify_backend": null, "enable_linear_replayssm": false, "linear_replayssm_cache_len": 16, "enable_linear_replayssm_spec": false, "enable_hierarchical_cache": false, "hicache_ratio": 2.0, "hicache_size": 0, "hicache_write_policy": "write_through", "hicache_io_backend": "kernel", "hicache_mem_layout": "page_first", "hicache_storage_backend": null, "hicache_storage_prefetch_policy": "timeout", "hicache_storage_backend_extra_config": null, "enable_hisparse": false, "hisparse_config": null, "enable_broadcast_mm_inputs_process": false, "enable_prefix_mm_cache": false, "mm_enable_dp_encoder": false, "mm_process_config": {}, "mm_processor_worker_num": 0, "mm_io_worker_num": 0, "limit_mm_data_per_request": null, "enable_mm_global_cache": false, "disable_fast_image_processor": false, "mm_feature_transport": "cpu", "keep_mm_feature_on_device": false, "enable_lora": null, "enable_lora_overlap_loading": null, "max_lora_rank": null, "lora_target_modules": null, "lora_paths": null, "max_loaded_loras": null, "max_loras_per_batch": 8, "lora_eviction_policy": "lru", "lora_backend": "csgmv", "max_lora_chunk_size": 16, "experts_shared_outer_loras": null, "lora_use_virtual_experts": false, "lora_strict_loading": false, "lora_drain_wait_threshold": 0.0, "enable_two_batch_overlap": false, "enable_single_batch_overlap": false, "tbo_token_distribution_threshold": 0.48, "cpu_offload_gb": 0, "offload_group_size": -1, "offload_num_in_group": 1, "offload_prefetch_step": 1, "offload_mode": "cpu", "enable_lmcache": false, "lmcache_config_file": null, "enable_flexkv": false, "flexkv_config_file": null, "kt_weight_path": null, "kt_method": "AMXINT4", "kt_cpuinfer": null, "kt_threadpool_count": 2, "kt_num_gpu_experts": null, "kt_max_deferred_experts_per_token": null, "dllm_algorithm": null, "dllm_algorithm_config": null, "dllm_fdfo": true, "disaggregation_mode": "decode", "disaggregation_transfer_backend": "mooncake", "disaggregation_bootstrap_port": 28800, "disaggregation_ib_device": "mlx5_0,mlx5_1,mlx5_2,mlx5_3", "disaggregation_decode_enable_radix_cache": false, "disaggregation_decode_enable_offload_kvcache": false, "num_reserved_decode_tokens": 512, "disaggregation_decode_extra_slots": 0, "disaggregation_decode_polling_interval": 1, "optimistic_prefill_attempts": 0, "encoder_only": false, "language_only": false, "encoder_transfer_backend": "zmq_to_tokenizer", "encoder_urls": [], "encoder_bootstrap_port": 8997, "encoder_register_urls": [], "enable_adaptive_dispatch_to_encoder": false, "enable_pdmux": false, "pdmux_config_path": null, "sm_group_num": 8, "custom_weight_loader": [], "weight_loader_disable_mmap": false, "weight_loader_prefetch_checkpoints": false, "weight_loader_prefetch_num_threads": 4, "weight_loader_drop_cache_after_load": false, "remote_instance_weight_loader_seed_instance_ip": null, "remote_instance_weight_loader_seed_instance_service_port": null, "remote_instance_weight_loader_send_weights_group_ports": null, "remote_instance_weight_loader_backend": "nccl", "remote_instance_weight_loader_start_seed_via_transfer_engine": false, "engine_info_bootstrap_port": 6789, "modelexpress_config": null, "download_dir": null, "model_checksum": null, "delete_ckpt_after_loading": false, "decrypted_config_file": null, "decrypted_draft_config_file": null, "checkpoint_engine_wait_weights_before_ready": false, "enable_prefill_delayer": false, "prefill_delayer_max_delay_passes": 30, "prefill_delayer_token_usage_low_watermark": null, "prefill_delayer_forward_passes_buckets": null, "prefill_delayer_wait_seconds_buckets": null, "prefill_delayer_queue_min_ratio": null, "prefill_delayer_max_delay_ms": null, "min_free_slots_delay": null, "enable_deterministic_inference": false, "rl_on_policy_target": null, "kv_canary": "none", "kv_canary_real_data": "none", "kv_canary_sweep_interval": 0, "enable_dynamic_batch_tokenizer": false, "dynamic_batch_tokenizer_batch_size": 32, "dynamic_batch_tokenizer_batch_timeout": 0.002, "enable_tokenizer_batch_encode": false, "disable_tokenizer_batch_decode": false, "debug_tensor_dump_output_folder": null, "debug_tensor_dump_layers": null, "debug_tensor_dump_input_file": null, "enable_memory_saver": false, "enable_weights_cpu_backup": false, "enable_draft_weights_cpu_backup": false, "enable_custom_logit_processor": false, "enable_return_hidden_states": false, "enable_return_routed_experts": false, "enable_return_indexer_topk": false, "disable_outlines_disk_cache": false, "enable_mis": false, "weight_cache_mode": "off", "weight_cache_socket": null, "weight_cache_timeout": 1800, "forward_hooks": null, "msprobe_dump_config": null, "status": "ready", "max_total_num_tokens": 440256, "max_req_input_len": 440250, "internal_states": [{"model_path": "/data/hf_models/Kimi-K3", "tokenizer_path": "/data/hf_models/Kimi-K3", "tokenizer_mode": "auto", "tokenizer_backend": "huggingface", "tokenizer_worker_num": 1, "detokenizer_worker_num": 1, "skip_tokenizer_init": false, "load_format": "auto", "model_loader_extra_config": "{}", "trust_remote_code": true, "context_length": null, "is_embedding": false, "enable_multimodal": null, "revision": null, "model_impl": "auto", "model_config_parser": "auto", "json_model_override_args": "{}", "dtype": "auto", "quantization": null, "quantization_param_path": null, "kv_cache_dtype": "auto", "enable_fp32_lm_head": false, "modelopt_quant": null, "modelopt_checkpoint_restore_path": null, "modelopt_checkpoint_save_path": null, "modelopt_export_path": null, "quantize_and_serve": false, "rl_quant_profile": null, "enable_tf32_matmul": false, "mem_fraction_static": 0.88, "max_running_requests": null, "max_queued_requests": null, "max_total_tokens": null, "chunked_prefill_size": 8192, "enable_dynamic_chunking": false, "max_prefill_tokens": 16384, "prefill_max_requests": null, "schedule_policy": "fcfs", "enable_priority_scheduling": false, "disable_priority_preemption": false, "default_priority_value": null, "abort_on_priority_when_disabled": false, "schedule_low_priority_values_first": false, "priority_scheduling_preemption_threshold": 10, "retraction_policy": "length", "schedule_conservativeness": 1.0, "page_size": 64, "swa_full_tokens_ratio": 0.8, "disable_hybrid_swa_memory": false, "radix_eviction_policy": "lru", "prefill_only_disable_kv_cache": false, "disable_radix_cache": true, "enable_page_major_kv_layout": false, "enable_unified_memory": false, "disable_chunked_prefix_cache": true, "disable_overlap_schedule": false, "num_continuous_decode_steps": 1, "scheduler_recv_interval": 1, "enable_mixed_chunk": false, "nccl_port": null, "dist_timeout": 3600, "dist_init_addr": "174.1.60.5:20000", "nnodes": 4, "node_rank": 0, "tp_size": 32, "dcp_size": 1, "pp_size": 1, "pp_max_micro_batch_size": 181, "pp_async_batch_depth": 0, "dp_size": 1, "load_balance_method": "round_robin", "attn_cp_size": 1, "moe_dp_size": 1, "dcp_comm_backend": "ag_rs", "dcp_replicate_q_proj": null, "dwdp_size": 1, "enable_prefill_cp": false, "cp_strategy": null, "enable_dsa_cache_layer_split": false, "enable_dsa_prefill_context_parallel": false, "dsa_prefill_cp_mode": "round-robin-split", "enable_prefill_context_parallel": false, "prefill_cp_mode": "in-seq-split", "enable_cp_decode_attn_tp": false, "enable_dp_attention": false, "enable_dp_attention_local_control_broadcast": false, "enable_dp_lm_head": false, "enable_attn_tp_input_scattered": false, "disable_attn_tp_gather": false, "enable_p2p_check": false, "device": "cuda", "base_gpu_id": 0, "gpu_id_step": 1, "random_seed": 369235288, "watchdog_timeout": 300, "soft_watchdog_timeout": null, "sleep_on_idle": false, "use_ray": false, "numa_node": null, "gc_threshold": null, "host": "0.0.0.0", "port": 30000, "fastapi_root_path": "", "smg_grpc_mode": false, "grpc_mode": false, "grpc_port": null, "sidecar": null, "sidecar_args": null, "skip_server_warmup": false, "warmups": null, "enable_http2": false, "ssl_keyfile": null, "ssl_certfile": null, "ssl_ca_certs": null, "ssl_keyfile_password": null, "enable_ssl_refresh": false, "api_key": null, "admin_api_key": null, "served_model_name": "kimi-k3", "weight_version": "default", "chat_template": null, "hf_chat_template_name": null, "completion_template": null, "file_storage_path": "sglang_storage", "enable_cache_report": false, "reasoning_parser": null, "default_chat_template_kwargs": null, "strip_thinking_cache": false, "enable_strict_thinking": false, "tool_call_parser": null, "tool_server": null, "sampling_defaults": "model", "asr_max_buffer_seconds": 60, "asr_max_concurrent_sessions": 32, "preferred_sampling_params": null, "allow_auto_truncate": false, "stream_interval": 1, "batch_notify_size": 16, "stream_response_default_include_usage": false, "incremental_streaming_output": false, "enable_streaming_session": false, "enable_session_radix_cache": false, "log_level": "info", "log_level_http": null, "log_requests": false, "log_requests_level": 2, "log_requests_format": "text", "log_requests_target": null, "uvicorn_access_log_exclude_prefixes": [], "crash_dump_folder": null, "show_time_cost": false, "enable_metrics": false, "smg_http_sidecar_port": null, "enable_mfu_metrics": false, "enable_metrics_for_all_schedulers": false, "load_snapshot_publish_interval": 15, "tokenizer_metrics_custom_labels_header": "x-custom-labels", "tokenizer_metrics_allowed_custom_labels": null, "extra_metric_labels": null, "bucket_time_to_first_token": null, "bucket_inter_token_latency": null, "bucket_e2e_request_latency": null, "prompt_tokens_buckets": null, "generation_tokens_buckets": null, "gc_warning_threshold_secs": 0.0, "decode_log_interval": 40, "enable_request_time_stats_logging": false, "kv_events_config": null, "enable_forward_pass_metrics": false, "forward_pass_metrics_worker_id": "", "forward_pass_metrics_ipc_name": null, "enable_trace": false, "trace_modules": "request", "otlp_traces_endpoint": "localhost:4317", "export_metrics_to_file": false, "export_metrics_to_file_dir": null, "stat_loggers": null, "constrained_json_whitespace_pattern": null, "constrained_json_disable_any_whitespace": false, "attention_backend": "triton", "decode_attention_backend": null, "prefill_attention_backend": null, "sampling_backend": "flashinfer", "grammar_backend": "xgrammar", "radix_cache_backend": null, "mm_attention_backend": null, "fp8_gemm_runner_backend": "auto", "fp4_gemm_runner_backend": "auto", "bf16_gemm_backend": "auto", "dsa_prefill_backend": null, "dsa_decode_backend": null, "dsa_paged_mqa_logits_backend": "auto", "dsa_topk_backend": "sgl-kernel", "disable_flashinfer_autotune": false, "flashinfer_autotune_skip_ops": null, "mamba_backend": "triton", "cuda_graph_config": {"decode": {"backend": "full", "max_bs": 16, "bs": [1, 2, 4, 8, 12, 16], "tc_compiler": "eager", "full_prefill_max_req": null}, "prefill": {"backend": "disabled", "max_bs": 2048, "bs": [4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], "tc_compiler": "eager", "full_prefill_max_req": null}}, "cuda_graph_backend_decode": null, "cuda_graph_backend_prefill": null, "cuda_graph_max_bs_decode": 16, "cuda_graph_max_bs_prefill": null, "cuda_graph_bs_decode": null, "cuda_graph_bs_prefill": null, "cuda_graph_tc_compiler": null, "disable_prefill_cuda_graph": false, "disable_decode_cuda_graph": false, "disable_cuda_graph": false, "disable_cuda_graph_padding": false, "enable_profile_cuda_graph": false, "enable_cudagraph_gc": false, "debug_cuda_graph": false, "enable_layerwise_nvtx_marker": false, "enable_nccl_nvls": false, "enable_symm_mem": false, "triton_attention_reduce_in_fp32": false, "triton_attention_num_kv_splits": 8, "triton_attention_split_tile_size": null, "flashinfer_mla_disable_ragged": false, "enable_fused_qk_norm_rope": false, "enable_precise_embedding_interpolation": false, "enable_fused_moe_sum_all_reduce": false, "enable_deepseek_v4_fp4_indexer": false, "disable_custom_all_reduce": false, "enable_mscclpp": false, "enable_torch_symm_mem": false, "enable_scattered_sconv": false, "pre_warm_nccl": false, "enable_quant_communications": false, "enable_flashinfer_allreduce_fusion": false, "enforce_disable_flashinfer_allreduce_fusion": false, "flashinfer_allreduce_fusion_backend": null, "enable_aiter_allreduce_fusion": false, "enable_torch_compile": false, "enable_torch_compile_debug_mode": false, "torch_compile_max_bs": 32, "torchao_config": "", "speculative_algorithm": null, "speculative_draft_model_path": null, "speculative_draft_model_revision": null, "speculative_draft_load_format": null, "speculative_num_steps": null, "speculative_eagle_topk": null, "speculative_num_draft_tokens": null, "speculative_dflash_block_size": null, "speculative_dspark_block_size": null, "speculative_dspark_sps_table_path": null, "speculative_dspark_confidence_sts_path": null, "speculative_dspark_align_verify_tokens_to_graph_tier": false, "speculative_accept_threshold_single": 1.0, "speculative_accept_threshold_acc": 1.0, "speculative_use_rejection_sampling": false, "speculative_token_map": null, "speculative_attention_mode": "prefill", "speculative_draft_attention_backend": null, "speculative_draft_window_size": null, "speculative_moe_runner_backend": "marlin", "speculative_moe_a2a_backend": null, "speculative_draft_model_quantization": null, "speculative_skip_dp_mlp_sync": false, "enable_multi_layer_eagle": false, "speculative_adaptive": false, "speculative_adaptive_config": null, "decoupled_spec_bind_endpoint": null, "decoupled_spec_connect_endpoints": null, "decoupled_spec_rank": null, "decoupled_spec_role": "null", "spec_trace_dir": null, "speculative_ngram_min_bfs_breadth": 1, "speculative_ngram_max_bfs_breadth": 10, "speculative_ngram_match_type": "BFS", "speculative_ngram_max_trie_depth": 18, "speculative_ngram_capacity": 10000000, "speculative_ngram_external_corpus_path": null, "speculative_ngram_external_sam_budget": 0, "speculative_ngram_external_corpus_max_tokens": 10000000, "ep_size": 32, "moe_a2a_backend": "none", "moe_runner_backend": "marlin", "flashinfer_mxfp4_moe_precision": "default", "deepep_mode": "auto", "fuseep_mode": 2, "deepep_dispatcher_output_dtype": "auto", "ep_num_redundant_experts": 0, "ep_dispatch_algorithm": null, "init_expert_location": "trivial", "enable_eplb": false, "eplb_algorithm": "auto", "eplb_rebalance_num_iterations": 1000, "eplb_rebalance_layers_per_chunk": null, "eplb_min_rebalancing_utilization_threshold": 1.0, "expert_distribution_recorder_mode": null, "expert_distribution_recorder_buffer_size": 1000, "enable_expert_distribution_metrics": false, "deepep_config": null, "moe_dense_tp_size": null, "elastic_ep_backend": null, "enable_elastic_expert_backup": false, "mooncake_ib_device": null, "enable_waterfill": false, "ep_join_mode": null, "ep_join_rank_offset": 0, "elastic_ep_initial_size": null, "max_ep_size": null, "elastic_ep_scale_timeout": 600, "elastic_ep_rejoin": false, "disable_flashinfer_cutlass_moe_fp4_allgather": false, "disable_shared_experts_fusion": false, "enforce_shared_experts_fusion": false, "max_mamba_cache_size": 181, "mamba_ssm_dtype": null, "mamba_max_states_per_path": -1, "enable_mamba_cache_stochastic_rounding": false, "mamba_cache_philox_rounds": 0, "mamba_full_memory_ratio": 0.21, "mamba_radix_cache_strategy": "extra_buffer_lazy", "uses_mamba_radix_cache": false, "mamba_track_interval": 256, "enable_int8_mamba_checkpoint": false, "int8_mamba_ckpt_size": null, "linear_attn_backend": "triton", "linear_attn_decode_backend": null, "linear_attn_prefill_backend": null, "linear_attn_verify_backend": null, "enable_linear_replayssm": false, "linear_replayssm_cache_len": 16, "enable_linear_replayssm_spec": false, "enable_hierarchical_cache": false, "hicache_ratio": 2.0, "hicache_size": 0, "hicache_write_policy": "write_through", "hicache_io_backend": "kernel", "hicache_mem_layout": "page_first", "hicache_storage_backend": null, "hicache_storage_prefetch_policy": "timeout", "hicache_storage_backend_extra_config": null, "enable_hisparse": false, "hisparse_config": null, "enable_broadcast_mm_inputs_process": false, "enable_prefix_mm_cache": false, "mm_enable_dp_encoder": false, "mm_process_config": {}, "mm_processor_worker_num": 0, "mm_io_worker_num": 0, "limit_mm_data_per_request": null, "enable_mm_global_cache": false, "disable_fast_image_processor": false, "mm_feature_transport": "cpu", "keep_mm_feature_on_device": false, "enable_lora": null, "enable_lora_overlap_loading": null, "max_lora_rank": null, "lora_target_modules": null, "lora_paths": null, "max_loaded_loras": null, "max_loras_per_batch": 8, "lora_eviction_policy": "lru", "lora_backend": "csgmv", "max_lora_chunk_size": 16, "experts_shared_outer_loras": null, "lora_use_virtual_experts": false, "lora_strict_loading": false, "lora_drain_wait_threshold": 0.0, "enable_two_batch_overlap": false, "enable_single_batch_overlap": false, "tbo_token_distribution_threshold": 0.48, "cpu_offload_gb": 0, "offload_group_size": -1, "offload_num_in_group": 1, "offload_prefetch_step": 1, "offload_mode": "cpu", "enable_lmcache": false, "lmcache_config_file": null, "enable_flexkv": false, "flexkv_config_file": null, "kt_weight_path": null, "kt_method": "AMXINT4", "kt_cpuinfer": null, "kt_threadpool_count": 2, "kt_num_gpu_experts": null, "kt_max_deferred_experts_per_token": null, "dllm_algorithm": null, "dllm_algorithm_config": null, "dllm_fdfo": true, "disaggregation_mode": "decode", "disaggregation_transfer_backend": "mooncake", "disaggregation_bootstrap_port": 28800, "disaggregation_ib_device": "mlx5_0,mlx5_1,mlx5_2,mlx5_3", "disaggregation_decode_enable_radix_cache": false, "disaggregation_decode_enable_offload_kvcache": false, "num_reserved_decode_tokens": 512, "disaggregation_decode_extra_slots": 0, "disaggregation_decode_polling_interval": 1, "optimistic_prefill_attempts": 0, "encoder_only": false, "language_only": false, "encoder_transfer_backend": "zmq_to_tokenizer", "encoder_urls": [], "encoder_bootstrap_port": 8997, "encoder_register_urls": [], "enable_adaptive_dispatch_to_encoder": false, "enable_pdmux": false, "pdmux_config_path": null, "sm_group_num": 8, "custom_weight_loader": [], "weight_loader_disable_mmap": false, "weight_loader_prefetch_checkpoints": false, "weight_loader_prefetch_num_threads": 4, "weight_loader_drop_cache_after_load": false, "remote_instance_weight_loader_seed_instance_ip": null, "remote_instance_weight_loader_seed_instance_service_port": null, "remote_instance_weight_loader_send_weights_group_ports": null, "remote_instance_weight_loader_backend": "nccl", "remote_instance_weight_loader_start_seed_via_transfer_engine": false, "engine_info_bootstrap_port": 6789, "modelexpress_config": null, "download_dir": null, "model_checksum": null, "delete_ckpt_after_loading": false, "decrypted_config_file": null, "decrypted_draft_config_file": null, "checkpoint_engine_wait_weights_before_ready": false, "enable_prefill_delayer": false, "prefill_delayer_max_delay_passes": 30, "prefill_delayer_token_usage_low_watermark": null, "prefill_delayer_forward_passes_buckets": null, "prefill_delayer_wait_seconds_buckets": null, "prefill_delayer_queue_min_ratio": null, "prefill_delayer_max_delay_ms": null, "min_free_slots_delay": null, "enable_deterministic_inference": false, "rl_on_policy_target": null, "kv_canary": "none", "kv_canary_real_data": "none", "kv_canary_sweep_interval": 0, "enable_dynamic_batch_tokenizer": false, "dynamic_batch_tokenizer_batch_size": 32, "dynamic_batch_tokenizer_batch_timeout": 0.002, "enable_tokenizer_batch_encode": false, "disable_tokenizer_batch_decode": false, "debug_tensor_dump_output_folder": null, "debug_tensor_dump_layers": null, "debug_tensor_dump_input_file": null, "enable_memory_saver": false, "enable_weights_cpu_backup": false, "enable_draft_weights_cpu_backup": false, "enable_custom_logit_processor": false, "enable_return_hidden_states": false, "enable_return_routed_experts": false, "enable_return_indexer_topk": false, "disable_outlines_disk_cache": false, "enable_mis": false, "weight_cache_mode": "off", "weight_cache_socket": null, "weight_cache_timeout": 1800, "forward_hooks": null, "msprobe_dump_config": null, "_resolved_overrides": [["_sampling_backend_default", {"sampling_backend": "flashinfer"}], ["_attention_backend_default", {"attention_backend": "triton"}], ["_data_parallelism_defaults", {"enable_dp_attention": false, "enable_dp_lm_head": false}], ["_speculative_moe_runner_default", {"speculative_moe_runner_backend": "marlin"}]], "grpc_worker_threads": 4, "_quantization_explicitly_unset": false, "_cuda_graph_config_locked": [["decode", "max_bs"]], "_declarations_materialized": true, "_runtime_mutations": [["model_runner.chunked_prefix_cache_gate", {"disable_chunked_prefix_cache": true}], ["mamba_pool.memory_budget", {"max_mamba_cache_size": 181}]], "_in_override": false, "_mx_config_cache": {}, "max_speculative_num_draft_tokens": null, "last_gen_throughput": 0.09006957280512767, "memory_usage": {"weight": 58.57, "kvcache": 11.34, "token_capacity": 440256, "graph": 0.6}, "effective_max_running_requests_per_dp": 181}], "version": "0.5.16", "kv_events": null}]}, "duration": 102.97423465992324, "completed": 40, "total_input_tokens": 655360, "total_input_text_tokens": 655360, "total_input_vision_tokens": 0, "total_output_tokens": 40, "total_output_tokens_retokenized": 40, "request_throughput": 0.38844668408657457, "input_throughput": 6364.310472074438, "output_throughput": 0.38844668408657457, "total_throughput": 6364.698918758524, "mean_e2e_latency_ms": 20164.965311507694, "median_e2e_latency_ms": 20554.723028908484, "std_e2e_latency_ms": 1424.4924608807214, "p90_e2e_latency_ms": 21172.218572022393, "p95_e2e_latency_ms": 21344.9914853787, "p99_e2e_latency_ms": 22867.75008213706, "mean_ttft_ms": 20164.917747490108, "median_ttft_ms": 20554.67213492375, "std_ttft_ms": 1424.495202307527, "p90_ttft_ms": 21172.178085357882, "p95_ttft_ms": 21344.942523411006, "p99_ttft_ms": 22867.70738121355, "mean_tpot_ms": 0.0, "median_tpot_ms": 0.0, "std_tpot_ms": 0.0, "p90_tpot_ms": 0.0, "p95_tpot_ms": 0.0, "p99_tpot_ms": 0.0, "mean_itl_ms": 0.0, "median_itl_ms": 0.0, "std_itl_ms": 0.0, "p90_itl_ms": 0.0, "p95_itl_ms": 0.0, "p99_itl_ms": 0.0, "concurrency": 7.833013909975964, "accept_length": null, "max_output_tokens_per_s": 4.0, "max_concurrent_requests": 10, "input_lens": [16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384, 16384], "output_lens": [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1], "ttfts": [14.792421613121405, 16.494908910011873, 18.22800297080539, 19.957135260803625, 20.50188078195788, 20.629512692801654, 21.929118582978845, 23.467788744019344, 19.42586032417603, 19.826406029984355, 19.90213485714048, 19.94499386101961, 20.625310443807393, 21.15849809604697, 21.314196415012702, 21.295297989156097, 20.63272958784364, 20.607796374941245, 20.581729213008657, 20.600462856004015, 20.566129944985732, 20.56192163587548, 20.55844004196115, 20.55355045478791, 20.5007693010848, 20.50776853202842, 20.53239204408601, 20.511765383183956, 20.569877775851637, 20.562934512039647, 20.555793815059587, 20.562208814080805, 20.551020709099248, 20.556789966998622, 20.549567875918, 20.557142625097185, 20.52026516897604, 19.905103630851954, 18.49541956698522, 17.001662496011704], "itls": [[], [], [], [], [], [], [], [], [], [], [], [], [], [], [], [], [], [], [], [], [], [], [], [], [], [], [], [], [], [], [], [], [], [], [], [], [], [], [], []], "generated_texts": ["\ufffd", "\ufffd", "\ufffd", "\ufffd", "\ufffd", "\ufffd", "\ufffd", "\u7684", "\ufffd", "\ufffd", "\ufffd", "\ufffd", "\ufffd", "\ufffd", "\ufffd", "\ufffd", "\ufffd", "(", "\ufffd", "\ufffd", "\u7684", "\ufffd", "ed", "\ufffd", "\ufffd", "\ufffd", "\ufffd", "\ufffd", "\ufffd", ".", "\ufffd", "\ufffd", "\n\n", "\ufffd", "\ufffd", "\ufffd", "\ufffd", "\ufffd", "\ufffd", "\ufffd"], "errors": ["", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", ""]}
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/bench/prefill_16k_1_c8.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/bench/prefill_16k_1_c8.log
new file mode 100644
index 0000000..ccf793e
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/bench/prefill_16k_1_c8.log
@@ -0,0 +1,110 @@
+/usr/local/lib/python3.12/dist-packages/torchao/quantization/quant_api.py:1731: SyntaxWarning: invalid escape sequence '\.'
+  """Configuration class for applying different quantization configs to modules or parameters based on their fully qualified names (FQNs).
+/sgl-workspace/sglang/python/sglang/kernels/ops/attention/fla/utils.py:223: UserWarning: Triton is not supported on current platform, roll back to CPU.
+  warnings.warn(
+/sgl-workspace/sglang/python/sglang/srt/layers/quantization/awq/awq.py:52: UserWarning: Only CUDA, HIP and XPU support AWQ currently.
+  warnings.warn(f"Only CUDA, HIP and XPU support AWQ currently.")
+/sgl-workspace/sglang/python/sglang/srt/layers/quantization/gguf.py:69: UserWarning: Only CUDA, MUSA and NPU support GGUF quantization currently.
+  warnings.warn(f"Only CUDA, MUSA and NPU support GGUF quantization currently.")
+benchmark_args=Namespace(backend='sglang', base_url=None, host='174.1.60.5', port=31000, ready_check_timeout_sec=60, dataset_name='random-ids', dataset_path='', dataset_offset=0, agentic_max_turns=None, speed_bench_category=None, speed_bench_output_len=512, model=None, served_model_name=None, tokenizer='/data/hf_models/Kimi-K3', num_prompts=40, sharegpt_output_len=None, sharegpt_context_len=None, random_input_len=16384, random_output_len=1, random_range_ratio=1.0, image_count=1, image_resolution='1080p', random_image_count=False, image_format='jpeg', image_content='random', request_rate=10000.0, use_trace_timestamps=False, max_concurrency=8, output_file='/data/hzy/sskj/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/bench/prefill_16k_1_c8.jsonl', output_details=True, print_requests=False, disable_tqdm=True, disable_stream=False, return_logprob=False, top_logprobs_num=0, token_ids_logprob=None, logprob_start_len=-1, return_routed_experts=False, cache_report=False, seed=42, disable_ignore_eos=False, temperature=0.0, top_p=1.0, extra_request_body=None, apply_chat_template=False, profile=False, plot_throughput=False, profile_activities=['CPU', 'GPU'], profile_start_step=None, profile_steps=None, profile_num_steps=None, profile_by_stage=False, profile_stages=None, profile_output_dir=None, profile_prefix=None, lora_name=None, lora_request_distribution='uniform', lora_zipf_alpha=1.5, prompt_suffix='', pd_separated=False, profile_prefill_url=None, profile_decode_url=None, flush_cache=False, warmup_requests=0, tokenize_prompt=False, gsp_num_groups=64, gsp_prompts_per_group=16, gsp_system_prompt_len=2048, gsp_question_len=128, gsp_output_len=256, gsp_range_ratio=1.0, gsp_fast_prepare=False, gsp_send_routing_key=False, gsp_num_turns=1, gsp_ordered=False, gsp_group_distribution='uniform', gsp_zipf_alpha=None, mooncake_slowdown_factor=1.0, mooncake_num_rounds=1, mooncake_workload='conversation', fake_prefill=False, tag=None, header=None)
+Waiting up to 60s for http://174.1.60.5:31000/v1/models to become ready...
+Server ready in 0.0s.
+'[Errno 101] Network is unreachable' thrown while requesting HEAD https://huggingface.co/kimi-k3/resolve/main/config.json
+Retrying in 1s [Retry 1/5].
+'[Errno 101] Network is unreachable' thrown while requesting HEAD https://huggingface.co/kimi-k3/resolve/main/config.json
+Retrying in 1s [Retry 1/5].
+Fail to load tokenizer config with error=Can't load the configuration of 'kimi-k3'. If you were trying to load it from 'https://huggingface.co/models', make sure you don't have a local directory with the same name. Otherwise, make sure 'kimi-k3' is the correct path to a directory containing a config.json file
+
+WARNING It is recommended to use the `Chat` or `Instruct` model for benchmarking.
+Because when the tokenizer counts the output tokens, if there is gibberish, it might count incorrectly.
+
+Namespace(backend='sglang', base_url=None, host='174.1.60.5', port=31000, ready_check_timeout_sec=60, dataset_name='random-ids', dataset_path='', dataset_offset=0, agentic_max_turns=None, speed_bench_category=None, speed_bench_output_len=512, model='kimi-k3', served_model_name=None, tokenizer='/data/hf_models/Kimi-K3', num_prompts=40, sharegpt_output_len=None, sharegpt_context_len=None, random_input_len=16384, random_output_len=1, random_range_ratio=1.0, image_count=1, image_resolution='1080p', random_image_count=False, image_format='jpeg', image_content='random', request_rate=10000.0, use_trace_timestamps=False, max_concurrency=8, output_file='/data/hzy/sskj/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/bench/prefill_16k_1_c8.jsonl', output_details=True, print_requests=False, disable_tqdm=True, disable_stream=False, return_logprob=False, top_logprobs_num=0, token_ids_logprob=None, logprob_start_len=-1, return_routed_experts=False, cache_report=False, seed=42, disable_ignore_eos=False, temperature=0.0, top_p=1.0, extra_request_body=None, apply_chat_template=False, profile=False, plot_throughput=False, profile_activities=['CPU', 'GPU'], profile_start_step=None, profile_steps=None, profile_num_steps=None, profile_by_stage=False, profile_stages=None, profile_output_dir=None, profile_prefix=None, lora_name=None, lora_request_distribution='uniform', lora_zipf_alpha=1.5, prompt_suffix='', pd_separated=False, profile_prefill_url=None, profile_decode_url=None, flush_cache=False, warmup_requests=0, tokenize_prompt=False, gsp_num_groups=64, gsp_prompts_per_group=16, gsp_system_prompt_len=2048, gsp_question_len=128, gsp_output_len=256, gsp_range_ratio=1.0, gsp_fast_prepare=False, gsp_send_routing_key=False, gsp_num_turns=1, gsp_ordered=False, gsp_group_distribution='uniform', gsp_zipf_alpha=None, mooncake_slowdown_factor=1.0, mooncake_num_rounds=1, mooncake_workload='conversation', fake_prefill=False, tag=None, header=None)
+
+#Input tokens: 655360
+#Output tokens: 40
+Starting warmup with 0 sequences...
+Warmup completed with 0 sequences. Starting main benchmark run...
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+Calling super().encode with {'add_special_tokens': False}
+
+============ Serving Benchmark Result ============
+Backend:                                 sglang    
+Traffic request rate:                    10000.0   
+Max request concurrency:                 8         
+Successful requests:                     40        
+Benchmark duration (s):                  102.97    
+Total input tokens:                      655360    
+Total input text tokens:                 655360    
+Total generated tokens:                  40        
+Total generated tokens (retokenized):    40        
+Request throughput (req/s):              0.39      
+Input token throughput (tok/s):          6364.31   
+Output token throughput (tok/s):         0.39      
+Peak output token throughput (tok/s):    4.00      
+Peak concurrent requests:                10        
+Total token throughput (tok/s):          6364.70   
+Concurrency:                             7.83      
+----------------End-to-End Latency----------------
+Mean E2E Latency (ms):                   20164.97  
+Median E2E Latency (ms):                 20554.72  
+P90 E2E Latency (ms):                    21172.22  
+P95 E2E Latency (ms):                    21344.99  
+P99 E2E Latency (ms):                    22867.75  
+---------------Time to First Token----------------
+Mean TTFT (ms):                          20164.92  
+Median TTFT (ms):                        20554.67  
+P90 TTFT (ms):                           21172.18  
+P95 TTFT (ms):                           21344.94  
+P99 TTFT (ms):                           22867.71  
+-----Time per Output Token (excl. 1st token)------
+Mean TPOT (ms):                          0.00      
+Median TPOT (ms):                        0.00      
+P90 TPOT (ms):                           0.00      
+P95 TPOT (ms):                           0.00      
+P99 TPOT (ms):                           0.00      
+---------------Inter-Token Latency----------------
+Mean ITL (ms):                           0.00      
+Median ITL (ms):                         0.00      
+P90 ITL (ms):                            0.00      
+P95 ITL (ms):                            0.00      
+P99 ITL (ms):                            0.00      
+Max ITL (ms):                            0.00      
+==================================================
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/bench/smoke_1k_16_c1.cmd.txt b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/bench/smoke_1k_16_c1.cmd.txt
new file mode 100644
index 0000000..9e35c80
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/bench/smoke_1k_16_c1.cmd.txt
@@ -0,0 +1 @@
+docker run --rm --network host -v /data/hf_models/Kimi-K3:/data/hf_models/Kimi-K3:ro -v /data/hzy/sskj/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310:/data/hzy/sskj/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310 --entrypoint python3 local/sglang:kimi-k3-sm120-flashinfer-mxfp4-pd1 -m sglang.benchmark.serving --backend sglang --host 174.1.60.5 --port 31000 --tokenizer /data/hf_models/Kimi-K3 --dataset-name random-ids --random-input-len 1024 --random-output-len 16 --random-range-ratio 1.0 --num-prompts 1 --max-concurrency 1 --request-rate 10000 --warmup-requests 0 --output-file /data/hzy/sskj/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/bench/smoke_1k_16_c1.jsonl --output-details --disable-tqdm --disable-stream 
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/bench/smoke_1k_16_c1.jsonl b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/bench/smoke_1k_16_c1.jsonl
new file mode 100644
index 0000000..967584a
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/bench/smoke_1k_16_c1.jsonl
@@ -0,0 +1 @@
+{"tag": null, "backend": "sglang", "dataset_name": "random-ids", "request_rate": 10000.0, "max_concurrency": 1, "sharegpt_output_len": null, "random_input_len": 1024, "random_output_len": 16, "random_range_ratio": 1.0, "server_info": {"internal_states": [{"model_path": "/data/hf_models/Kimi-K3", "tokenizer_path": "/data/hf_models/Kimi-K3", "tokenizer_mode": "auto", "tokenizer_backend": "huggingface", "tokenizer_worker_num": 1, "detokenizer_worker_num": 1, "skip_tokenizer_init": false, "load_format": "auto", "model_loader_extra_config": "{}", "trust_remote_code": true, "context_length": null, "is_embedding": false, "enable_multimodal": null, "revision": null, "model_impl": "auto", "model_config_parser": "auto", "json_model_override_args": "{}", "dtype": "auto", "quantization": null, "quantization_param_path": null, "kv_cache_dtype": "auto", "enable_fp32_lm_head": false, "modelopt_quant": null, "modelopt_checkpoint_restore_path": null, "modelopt_checkpoint_save_path": null, "modelopt_export_path": null, "quantize_and_serve": false, "rl_quant_profile": null, "enable_tf32_matmul": false, "mem_fraction_static": 0.88, "max_running_requests": null, "max_queued_requests": null, "max_total_tokens": null, "chunked_prefill_size": 8192, "enable_dynamic_chunking": false, "max_prefill_tokens": 16384, "prefill_max_requests": null, "schedule_policy": "fcfs", "enable_priority_scheduling": false, "disable_priority_preemption": false, "default_priority_value": null, "abort_on_priority_when_disabled": false, "schedule_low_priority_values_first": false, "priority_scheduling_preemption_threshold": 10, "retraction_policy": "length", "schedule_conservativeness": 1.0, "page_size": 64, "swa_full_tokens_ratio": 0.8, "disable_hybrid_swa_memory": false, "radix_eviction_policy": "lru", "prefill_only_disable_kv_cache": false, "disable_radix_cache": true, "enable_page_major_kv_layout": false, "enable_unified_memory": false, "disable_chunked_prefix_cache": true, "disable_overlap_schedule": false, "num_continuous_decode_steps": 1, "scheduler_recv_interval": 1, "enable_mixed_chunk": false, "nccl_port": null, "dist_timeout": 3600, "dist_init_addr": "174.1.60.5:20000", "nnodes": 4, "node_rank": 0, "tp_size": 32, "dcp_size": 1, "pp_size": 1, "pp_max_micro_batch_size": 181, "pp_async_batch_depth": 0, "dp_size": 1, "load_balance_method": "round_robin", "attn_cp_size": 1, "moe_dp_size": 1, "dcp_comm_backend": "ag_rs", "dcp_replicate_q_proj": null, "dwdp_size": 1, "enable_prefill_cp": false, "cp_strategy": null, "enable_dsa_cache_layer_split": false, "enable_dsa_prefill_context_parallel": false, "dsa_prefill_cp_mode": "round-robin-split", "enable_prefill_context_parallel": false, "prefill_cp_mode": "in-seq-split", "enable_cp_decode_attn_tp": false, "enable_dp_attention": false, "enable_dp_attention_local_control_broadcast": false, "enable_dp_lm_head": false, "enable_attn_tp_input_scattered": false, "disable_attn_tp_gather": false, "enable_p2p_check": false, "device": "cuda", "base_gpu_id": 0, "gpu_id_step": 1, "random_seed": 369235288, "watchdog_timeout": 300, "soft_watchdog_timeout": null, "sleep_on_idle": false, "use_ray": false, "numa_node": null, "gc_threshold": null, "host": "0.0.0.0", "port": 30000, "fastapi_root_path": "", "smg_grpc_mode": false, "grpc_mode": false, "grpc_port": null, "sidecar": null, "sidecar_args": null, "skip_server_warmup": false, "warmups": null, "enable_http2": false, "ssl_keyfile": null, "ssl_certfile": null, "ssl_ca_certs": null, "ssl_keyfile_password": null, "enable_ssl_refresh": false, "api_key": null, "admin_api_key": null, "served_model_name": "kimi-k3", "weight_version": "default", "chat_template": null, "hf_chat_template_name": null, "completion_template": null, "file_storage_path": "sglang_storage", "enable_cache_report": false, "reasoning_parser": null, "default_chat_template_kwargs": null, "strip_thinking_cache": false, "enable_strict_thinking": false, "tool_call_parser": null, "tool_server": null, "sampling_defaults": "model", "asr_max_buffer_seconds": 60, "asr_max_concurrent_sessions": 32, "preferred_sampling_params": null, "allow_auto_truncate": false, "stream_interval": 1, "batch_notify_size": 16, "stream_response_default_include_usage": false, "incremental_streaming_output": false, "enable_streaming_session": false, "enable_session_radix_cache": false, "log_level": "info", "log_level_http": null, "log_requests": false, "log_requests_level": 2, "log_requests_format": "text", "log_requests_target": null, "uvicorn_access_log_exclude_prefixes": [], "crash_dump_folder": null, "show_time_cost": false, "enable_metrics": false, "smg_http_sidecar_port": null, "enable_mfu_metrics": false, "enable_metrics_for_all_schedulers": false, "load_snapshot_publish_interval": 15, "tokenizer_metrics_custom_labels_header": "x-custom-labels", "tokenizer_metrics_allowed_custom_labels": null, "extra_metric_labels": null, "bucket_time_to_first_token": null, "bucket_inter_token_latency": null, "bucket_e2e_request_latency": null, "prompt_tokens_buckets": null, "generation_tokens_buckets": null, "gc_warning_threshold_secs": 0.0, "decode_log_interval": 40, "enable_request_time_stats_logging": false, "kv_events_config": null, "enable_forward_pass_metrics": false, "forward_pass_metrics_worker_id": "", "forward_pass_metrics_ipc_name": null, "enable_trace": false, "trace_modules": "request", "otlp_traces_endpoint": "localhost:4317", "export_metrics_to_file": false, "export_metrics_to_file_dir": null, "stat_loggers": null, "constrained_json_whitespace_pattern": null, "constrained_json_disable_any_whitespace": false, "attention_backend": "triton", "decode_attention_backend": null, "prefill_attention_backend": null, "sampling_backend": "flashinfer", "grammar_backend": "xgrammar", "radix_cache_backend": null, "mm_attention_backend": null, "fp8_gemm_runner_backend": "auto", "fp4_gemm_runner_backend": "auto", "bf16_gemm_backend": "auto", "dsa_prefill_backend": null, "dsa_decode_backend": null, "dsa_paged_mqa_logits_backend": "auto", "dsa_topk_backend": "sgl-kernel", "disable_flashinfer_autotune": false, "flashinfer_autotune_skip_ops": null, "mamba_backend": "triton", "cuda_graph_config": {"decode": {"backend": "full", "max_bs": 16, "bs": [1, 2, 4, 8, 12, 16], "tc_compiler": "eager", "full_prefill_max_req": null}, "prefill": {"backend": "disabled", "max_bs": 2048, "bs": [4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], "tc_compiler": "eager", "full_prefill_max_req": null}}, "cuda_graph_backend_decode": null, "cuda_graph_backend_prefill": null, "cuda_graph_max_bs_decode": 16, "cuda_graph_max_bs_prefill": null, "cuda_graph_bs_decode": null, "cuda_graph_bs_prefill": null, "cuda_graph_tc_compiler": null, "disable_prefill_cuda_graph": false, "disable_decode_cuda_graph": false, "disable_cuda_graph": false, "disable_cuda_graph_padding": false, "enable_profile_cuda_graph": false, "enable_cudagraph_gc": false, "debug_cuda_graph": false, "enable_layerwise_nvtx_marker": false, "enable_nccl_nvls": false, "enable_symm_mem": false, "triton_attention_reduce_in_fp32": false, "triton_attention_num_kv_splits": 8, "triton_attention_split_tile_size": null, "flashinfer_mla_disable_ragged": false, "enable_fused_qk_norm_rope": false, "enable_precise_embedding_interpolation": false, "enable_fused_moe_sum_all_reduce": false, "enable_deepseek_v4_fp4_indexer": false, "disable_custom_all_reduce": false, "enable_mscclpp": false, "enable_torch_symm_mem": false, "enable_scattered_sconv": false, "pre_warm_nccl": false, "enable_quant_communications": false, "enable_flashinfer_allreduce_fusion": false, "enforce_disable_flashinfer_allreduce_fusion": false, "flashinfer_allreduce_fusion_backend": null, "enable_aiter_allreduce_fusion": false, "enable_torch_compile": false, "enable_torch_compile_debug_mode": false, "torch_compile_max_bs": 32, "torchao_config": "", "speculative_algorithm": null, "speculative_draft_model_path": null, "speculative_draft_model_revision": null, "speculative_draft_load_format": null, "speculative_num_steps": null, "speculative_eagle_topk": null, "speculative_num_draft_tokens": null, "speculative_dflash_block_size": null, "speculative_dspark_block_size": null, "speculative_dspark_sps_table_path": null, "speculative_dspark_confidence_sts_path": null, "speculative_dspark_align_verify_tokens_to_graph_tier": false, "speculative_accept_threshold_single": 1.0, "speculative_accept_threshold_acc": 1.0, "speculative_use_rejection_sampling": false, "speculative_token_map": null, "speculative_attention_mode": "prefill", "speculative_draft_attention_backend": null, "speculative_draft_window_size": null, "speculative_moe_runner_backend": "marlin", "speculative_moe_a2a_backend": null, "speculative_draft_model_quantization": null, "speculative_skip_dp_mlp_sync": false, "enable_multi_layer_eagle": false, "speculative_adaptive": false, "speculative_adaptive_config": null, "decoupled_spec_bind_endpoint": null, "decoupled_spec_connect_endpoints": null, "decoupled_spec_rank": null, "decoupled_spec_role": "null", "spec_trace_dir": null, "speculative_ngram_min_bfs_breadth": 1, "speculative_ngram_max_bfs_breadth": 10, "speculative_ngram_match_type": "BFS", "speculative_ngram_max_trie_depth": 18, "speculative_ngram_capacity": 10000000, "speculative_ngram_external_corpus_path": null, "speculative_ngram_external_sam_budget": 0, "speculative_ngram_external_corpus_max_tokens": 10000000, "ep_size": 32, "moe_a2a_backend": "none", "moe_runner_backend": "marlin", "flashinfer_mxfp4_moe_precision": "default", "deepep_mode": "auto", "fuseep_mode": 2, "deepep_dispatcher_output_dtype": "auto", "ep_num_redundant_experts": 0, "ep_dispatch_algorithm": null, "init_expert_location": "trivial", "enable_eplb": false, "eplb_algorithm": "auto", "eplb_rebalance_num_iterations": 1000, "eplb_rebalance_layers_per_chunk": null, "eplb_min_rebalancing_utilization_threshold": 1.0, "expert_distribution_recorder_mode": null, "expert_distribution_recorder_buffer_size": 1000, "enable_expert_distribution_metrics": false, "deepep_config": null, "moe_dense_tp_size": null, "elastic_ep_backend": null, "enable_elastic_expert_backup": false, "mooncake_ib_device": null, "enable_waterfill": false, "ep_join_mode": null, "ep_join_rank_offset": 0, "elastic_ep_initial_size": null, "max_ep_size": null, "elastic_ep_scale_timeout": 600, "elastic_ep_rejoin": false, "disable_flashinfer_cutlass_moe_fp4_allgather": false, "disable_shared_experts_fusion": false, "enforce_shared_experts_fusion": false, "max_mamba_cache_size": 181, "mamba_ssm_dtype": null, "mamba_max_states_per_path": -1, "enable_mamba_cache_stochastic_rounding": false, "mamba_cache_philox_rounds": 0, "mamba_full_memory_ratio": 0.21, "mamba_radix_cache_strategy": "extra_buffer_lazy", "uses_mamba_radix_cache": false, "mamba_track_interval": 256, "enable_int8_mamba_checkpoint": false, "int8_mamba_ckpt_size": null, "linear_attn_backend": "triton", "linear_attn_decode_backend": null, "linear_attn_prefill_backend": null, "linear_attn_verify_backend": null, "enable_linear_replayssm": false, "linear_replayssm_cache_len": 16, "enable_linear_replayssm_spec": false, "enable_hierarchical_cache": false, "hicache_ratio": 2.0, "hicache_size": 0, "hicache_write_policy": "write_through", "hicache_io_backend": "kernel", "hicache_mem_layout": "page_first", "hicache_storage_backend": null, "hicache_storage_prefetch_policy": "timeout", "hicache_storage_backend_extra_config": null, "enable_hisparse": false, "hisparse_config": null, "enable_broadcast_mm_inputs_process": false, "enable_prefix_mm_cache": false, "mm_enable_dp_encoder": false, "mm_process_config": {}, "mm_processor_worker_num": 0, "mm_io_worker_num": 0, "limit_mm_data_per_request": null, "enable_mm_global_cache": false, "disable_fast_image_processor": false, "mm_feature_transport": "cpu", "keep_mm_feature_on_device": false, "enable_lora": null, "enable_lora_overlap_loading": null, "max_lora_rank": null, "lora_target_modules": null, "lora_paths": null, "max_loaded_loras": null, "max_loras_per_batch": 8, "lora_eviction_policy": "lru", "lora_backend": "csgmv", "max_lora_chunk_size": 16, "experts_shared_outer_loras": null, "lora_use_virtual_experts": false, "lora_strict_loading": false, "lora_drain_wait_threshold": 0.0, "enable_two_batch_overlap": false, "enable_single_batch_overlap": false, "tbo_token_distribution_threshold": 0.48, "cpu_offload_gb": 0, "offload_group_size": -1, "offload_num_in_group": 1, "offload_prefetch_step": 1, "offload_mode": "cpu", "enable_lmcache": false, "lmcache_config_file": null, "enable_flexkv": false, "flexkv_config_file": null, "kt_weight_path": null, "kt_method": "AMXINT4", "kt_cpuinfer": null, "kt_threadpool_count": 2, "kt_num_gpu_experts": null, "kt_max_deferred_experts_per_token": null, "dllm_algorithm": null, "dllm_algorithm_config": null, "dllm_fdfo": true, "disaggregation_mode": "decode", "disaggregation_transfer_backend": "mooncake", "disaggregation_bootstrap_port": 28800, "disaggregation_ib_device": "mlx5_0,mlx5_1,mlx5_2,mlx5_3", "disaggregation_decode_enable_radix_cache": false, "disaggregation_decode_enable_offload_kvcache": false, "num_reserved_decode_tokens": 512, "disaggregation_decode_extra_slots": 0, "disaggregation_decode_polling_interval": 1, "optimistic_prefill_attempts": 0, "encoder_only": false, "language_only": false, "encoder_transfer_backend": "zmq_to_tokenizer", "encoder_urls": [], "encoder_bootstrap_port": 8997, "encoder_register_urls": [], "enable_adaptive_dispatch_to_encoder": false, "enable_pdmux": false, "pdmux_config_path": null, "sm_group_num": 8, "custom_weight_loader": [], "weight_loader_disable_mmap": false, "weight_loader_prefetch_checkpoints": false, "weight_loader_prefetch_num_threads": 4, "weight_loader_drop_cache_after_load": false, "remote_instance_weight_loader_seed_instance_ip": null, "remote_instance_weight_loader_seed_instance_service_port": null, "remote_instance_weight_loader_send_weights_group_ports": null, "remote_instance_weight_loader_backend": "nccl", "remote_instance_weight_loader_start_seed_via_transfer_engine": false, "engine_info_bootstrap_port": 6789, "modelexpress_config": null, "download_dir": null, "model_checksum": null, "delete_ckpt_after_loading": false, "decrypted_config_file": null, "decrypted_draft_config_file": null, "checkpoint_engine_wait_weights_before_ready": false, "enable_prefill_delayer": false, "prefill_delayer_max_delay_passes": 30, "prefill_delayer_token_usage_low_watermark": null, "prefill_delayer_forward_passes_buckets": null, "prefill_delayer_wait_seconds_buckets": null, "prefill_delayer_queue_min_ratio": null, "prefill_delayer_max_delay_ms": null, "min_free_slots_delay": null, "enable_deterministic_inference": false, "rl_on_policy_target": null, "kv_canary": "none", "kv_canary_real_data": "none", "kv_canary_sweep_interval": 0, "enable_dynamic_batch_tokenizer": false, "dynamic_batch_tokenizer_batch_size": 32, "dynamic_batch_tokenizer_batch_timeout": 0.002, "enable_tokenizer_batch_encode": false, "disable_tokenizer_batch_decode": false, "debug_tensor_dump_output_folder": null, "debug_tensor_dump_layers": null, "debug_tensor_dump_input_file": null, "enable_memory_saver": false, "enable_weights_cpu_backup": false, "enable_draft_weights_cpu_backup": false, "enable_custom_logit_processor": false, "enable_return_hidden_states": false, "enable_return_routed_experts": false, "enable_return_indexer_topk": false, "disable_outlines_disk_cache": false, "enable_mis": false, "weight_cache_mode": "off", "weight_cache_socket": null, "weight_cache_timeout": 1800, "forward_hooks": null, "msprobe_dump_config": null, "_resolved_overrides": [["_sampling_backend_default", {"sampling_backend": "flashinfer"}], ["_attention_backend_default", {"attention_backend": "triton"}], ["_data_parallelism_defaults", {"enable_dp_attention": false, "enable_dp_lm_head": false}], ["_speculative_moe_runner_default", {"speculative_moe_runner_backend": "marlin"}]], "grpc_worker_threads": 4, "_quantization_explicitly_unset": false, "_cuda_graph_config_locked": [["decode", "max_bs"]], "_declarations_materialized": true, "_runtime_mutations": [["model_runner.chunked_prefix_cache_gate", {"disable_chunked_prefix_cache": true}], ["mamba_pool.memory_budget", {"max_mamba_cache_size": 181}]], "_in_override": false, "_mx_config_cache": {}, "max_speculative_num_draft_tokens": null, "last_gen_throughput": 0.09006957280512767, "memory_usage": {"weight": 58.57, "kvcache": 11.34, "token_capacity": 440256, "graph": 0.6}, "effective_max_running_requests_per_dp": 181}], "prefill": [{"model_path": "/data/hf_models/Kimi-K3", "tokenizer_path": "/data/hf_models/Kimi-K3", "tokenizer_mode": "auto", "tokenizer_backend": "huggingface", "tokenizer_worker_num": 1, "detokenizer_worker_num": 1, "skip_tokenizer_init": false, "load_format": "auto", "model_loader_extra_config": "{}", "trust_remote_code": true, "context_length": null, "is_embedding": false, "enable_multimodal": null, "revision": null, "model_impl": "auto", "model_config_parser": "auto", "json_model_override_args": "{}", "dtype": "auto", "quantization": null, "quantization_param_path": null, "kv_cache_dtype": "auto", "enable_fp32_lm_head": false, "modelopt_quant": null, "modelopt_checkpoint_restore_path": null, "modelopt_checkpoint_save_path": null, "modelopt_export_path": null, "quantize_and_serve": false, "rl_quant_profile": null, "enable_tf32_matmul": false, "mem_fraction_static": 0.88, "max_running_requests": null, "max_queued_requests": null, "max_total_tokens": null, "chunked_prefill_size": 8192, "enable_dynamic_chunking": false, "max_prefill_tokens": 16384, "prefill_max_requests": null, "schedule_policy": "fcfs", "enable_priority_scheduling": false, "disable_priority_preemption": false, "default_priority_value": null, "abort_on_priority_when_disabled": false, "schedule_low_priority_values_first": false, "priority_scheduling_preemption_threshold": 10, "retraction_policy": "length", "schedule_conservativeness": 1.0, "page_size": 64, "swa_full_tokens_ratio": 0.8, "disable_hybrid_swa_memory": false, "radix_eviction_policy": "lru", "prefill_only_disable_kv_cache": false, "disable_radix_cache": true, "enable_page_major_kv_layout": false, "enable_unified_memory": false, "disable_chunked_prefix_cache": false, "disable_overlap_schedule": true, "num_continuous_decode_steps": 1, "scheduler_recv_interval": 1, "enable_mixed_chunk": false, "nccl_port": null, "dist_timeout": 3600, "dist_init_addr": "174.1.60.1:20000", "nnodes": 4, "node_rank": 0, "tp_size": 4, "dcp_size": 1, "pp_size": 8, "pp_max_micro_batch_size": null, "pp_async_batch_depth": 0, "dp_size": 1, "load_balance_method": "follow_bootstrap_room", "attn_cp_size": 1, "moe_dp_size": 1, "dcp_comm_backend": "ag_rs", "dcp_replicate_q_proj": null, "dwdp_size": 1, "enable_prefill_cp": false, "cp_strategy": null, "enable_dsa_cache_layer_split": false, "enable_dsa_prefill_context_parallel": false, "dsa_prefill_cp_mode": "round-robin-split", "enable_prefill_context_parallel": false, "prefill_cp_mode": "in-seq-split", "enable_cp_decode_attn_tp": false, "enable_dp_attention": false, "enable_dp_attention_local_control_broadcast": false, "enable_dp_lm_head": false, "enable_attn_tp_input_scattered": false, "disable_attn_tp_gather": false, "enable_p2p_check": false, "device": "cuda", "base_gpu_id": 0, "gpu_id_step": 1, "random_seed": 984607766, "watchdog_timeout": 300, "soft_watchdog_timeout": null, "sleep_on_idle": false, "use_ray": false, "custom_sigquit_handler": null, "numa_node": null, "gc_threshold": null, "host": "0.0.0.0", "port": 30000, "fastapi_root_path": "", "smg_grpc_mode": false, "grpc_mode": false, "grpc_port": null, "sidecar": null, "sidecar_args": null, "skip_server_warmup": false, "warmups": null, "enable_http2": false, "ssl_keyfile": null, "ssl_certfile": null, "ssl_ca_certs": null, "ssl_keyfile_password": null, "enable_ssl_refresh": false, "api_key": null, "admin_api_key": null, "served_model_name": "kimi-k3", "weight_version": "default", "chat_template": null, "hf_chat_template_name": null, "completion_template": null, "file_storage_path": "sglang_storage", "enable_cache_report": false, "reasoning_parser": null, "default_chat_template_kwargs": null, "strip_thinking_cache": false, "enable_strict_thinking": false, "tool_call_parser": null, "tool_server": null, "sampling_defaults": "model", "asr_max_buffer_seconds": 60, "asr_max_concurrent_sessions": 32, "preferred_sampling_params": null, "allow_auto_truncate": false, "stream_interval": 1, "batch_notify_size": 16, "stream_response_default_include_usage": false, "incremental_streaming_output": false, "enable_streaming_session": false, "enable_session_radix_cache": false, "log_level": "info", "log_level_http": null, "log_requests": false, "log_requests_level": 2, "log_requests_format": "text", "log_requests_target": null, "uvicorn_access_log_exclude_prefixes": [], "crash_dump_folder": null, "show_time_cost": false, "enable_metrics": false, "smg_http_sidecar_port": null, "enable_mfu_metrics": false, "enable_metrics_for_all_schedulers": false, "load_snapshot_publish_interval": 15, "tokenizer_metrics_custom_labels_header": "x-custom-labels", "tokenizer_metrics_allowed_custom_labels": null, "extra_metric_labels": null, "bucket_time_to_first_token": null, "bucket_inter_token_latency": null, "bucket_e2e_request_latency": null, "prompt_tokens_buckets": null, "generation_tokens_buckets": null, "gc_warning_threshold_secs": 0.0, "decode_log_interval": 40, "enable_request_time_stats_logging": false, "kv_events_config": null, "enable_forward_pass_metrics": false, "forward_pass_metrics_worker_id": "", "forward_pass_metrics_ipc_name": null, "enable_trace": false, "trace_modules": "request", "otlp_traces_endpoint": "localhost:4317", "export_metrics_to_file": false, "export_metrics_to_file_dir": null, "stat_loggers": null, "constrained_json_whitespace_pattern": null, "constrained_json_disable_any_whitespace": false, "attention_backend": "triton", "decode_attention_backend": null, "prefill_attention_backend": null, "sampling_backend": "flashinfer", "grammar_backend": "xgrammar", "radix_cache_backend": null, "mm_attention_backend": null, "fp8_gemm_runner_backend": "auto", "fp4_gemm_runner_backend": "auto", "bf16_gemm_backend": "auto", "dsa_prefill_backend": null, "dsa_decode_backend": null, "dsa_paged_mqa_logits_backend": "auto", "dsa_topk_backend": "sgl-kernel", "disable_flashinfer_autotune": false, "flashinfer_autotune_skip_ops": null, "mamba_backend": "triton", "cuda_graph_config": {"decode": {"backend": "disabled", "max_bs": 16, "bs": [1, 2, 4, 8, 12, 16], "tc_compiler": "eager", "full_prefill_max_req": null}, "prefill": {"backend": "disabled", "max_bs": 2048, "bs": [4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], "tc_compiler": "eager", "full_prefill_max_req": null}}, "cuda_graph_backend_decode": null, "cuda_graph_backend_prefill": null, "cuda_graph_max_bs_decode": 16, "cuda_graph_max_bs_prefill": null, "cuda_graph_bs_decode": null, "cuda_graph_bs_prefill": null, "cuda_graph_tc_compiler": null, "disable_prefill_cuda_graph": false, "disable_decode_cuda_graph": false, "disable_cuda_graph": false, "disable_cuda_graph_padding": false, "enable_profile_cuda_graph": false, "enable_cudagraph_gc": false, "debug_cuda_graph": false, "enable_layerwise_nvtx_marker": false, "enable_nccl_nvls": false, "enable_symm_mem": false, "triton_attention_reduce_in_fp32": false, "triton_attention_num_kv_splits": 8, "triton_attention_split_tile_size": null, "flashinfer_mla_disable_ragged": false, "enable_fused_qk_norm_rope": false, "enable_precise_embedding_interpolation": false, "enable_fused_moe_sum_all_reduce": false, "enable_deepseek_v4_fp4_indexer": false, "disable_custom_all_reduce": false, "enable_mscclpp": false, "enable_torch_symm_mem": false, "enable_scattered_sconv": false, "pre_warm_nccl": false, "enable_quant_communications": false, "enable_flashinfer_allreduce_fusion": false, "enforce_disable_flashinfer_allreduce_fusion": false, "flashinfer_allreduce_fusion_backend": null, "enable_aiter_allreduce_fusion": false, "enable_torch_compile": false, "enable_torch_compile_debug_mode": false, "torch_compile_max_bs": 32, "torchao_config": "", "speculative_algorithm": null, "speculative_draft_model_path": null, "speculative_draft_model_revision": null, "speculative_draft_load_format": null, "speculative_num_steps": null, "speculative_eagle_topk": null, "speculative_num_draft_tokens": null, "speculative_dflash_block_size": null, "speculative_dspark_block_size": null, "speculative_dspark_sps_table_path": null, "speculative_dspark_confidence_sts_path": null, "speculative_dspark_align_verify_tokens_to_graph_tier": false, "speculative_accept_threshold_single": 1.0, "speculative_accept_threshold_acc": 1.0, "speculative_use_rejection_sampling": false, "speculative_token_map": null, "speculative_attention_mode": "prefill", "speculative_draft_attention_backend": null, "speculative_draft_window_size": null, "speculative_moe_runner_backend": "flashinfer_mxfp4", "speculative_moe_a2a_backend": null, "speculative_draft_model_quantization": null, "speculative_skip_dp_mlp_sync": false, "enable_multi_layer_eagle": false, "speculative_adaptive": false, "speculative_adaptive_config": null, "decoupled_spec_bind_endpoint": null, "decoupled_spec_connect_endpoints": null, "decoupled_spec_rank": null, "decoupled_spec_role": "null", "spec_trace_dir": null, "speculative_ngram_min_bfs_breadth": 1, "speculative_ngram_max_bfs_breadth": 10, "speculative_ngram_match_type": "BFS", "speculative_ngram_max_trie_depth": 18, "speculative_ngram_capacity": 10000000, "speculative_ngram_external_corpus_path": null, "speculative_ngram_external_sam_budget": 0, "speculative_ngram_external_corpus_max_tokens": 10000000, "ep_size": 4, "moe_a2a_backend": "none", "moe_runner_backend": "flashinfer_mxfp4", "flashinfer_mxfp4_moe_precision": "default", "deepep_mode": "auto", "fuseep_mode": 2, "deepep_dispatcher_output_dtype": "auto", "ep_num_redundant_experts": 0, "ep_dispatch_algorithm": null, "init_expert_location": "trivial", "enable_eplb": false, "eplb_algorithm": "auto", "eplb_rebalance_num_iterations": 1000, "eplb_rebalance_layers_per_chunk": null, "eplb_min_rebalancing_utilization_threshold": 1.0, "expert_distribution_recorder_mode": null, "expert_distribution_recorder_buffer_size": 1000, "enable_expert_distribution_metrics": false, "deepep_config": null, "moe_dense_tp_size": null, "elastic_ep_backend": null, "enable_elastic_expert_backup": false, "mooncake_ib_device": null, "enable_waterfill": false, "ep_join_mode": null, "ep_join_rank_offset": 0, "elastic_ep_initial_size": null, "max_ep_size": null, "elastic_ep_scale_timeout": 600, "elastic_ep_rejoin": false, "disable_flashinfer_cutlass_moe_fp4_allgather": false, "disable_shared_experts_fusion": false, "enforce_shared_experts_fusion": false, "max_mamba_cache_size": null, "mamba_ssm_dtype": null, "mamba_max_states_per_path": -1, "enable_mamba_cache_stochastic_rounding": false, "mamba_cache_philox_rounds": 0, "mamba_full_memory_ratio": 0.36, "mamba_radix_cache_strategy": "extra_buffer_lazy", "uses_mamba_radix_cache": false, "mamba_track_interval": 256, "enable_int8_mamba_checkpoint": false, "int8_mamba_ckpt_size": null, "linear_attn_backend": "triton", "linear_attn_decode_backend": null, "linear_attn_prefill_backend": null, "linear_attn_verify_backend": null, "enable_linear_replayssm": false, "linear_replayssm_cache_len": 16, "enable_linear_replayssm_spec": false, "enable_hierarchical_cache": false, "hicache_ratio": 2.0, "hicache_size": 0, "hicache_write_policy": "write_through", "hicache_io_backend": "kernel", "hicache_mem_layout": "page_first", "hicache_storage_backend": null, "hicache_storage_prefetch_policy": "timeout", "hicache_storage_backend_extra_config": null, "enable_hisparse": false, "hisparse_config": null, "enable_broadcast_mm_inputs_process": false, "enable_prefix_mm_cache": false, "mm_enable_dp_encoder": false, "mm_process_config": {}, "mm_processor_worker_num": 0, "mm_io_worker_num": 0, "limit_mm_data_per_request": null, "enable_mm_global_cache": false, "disable_fast_image_processor": false, "mm_feature_transport": "cpu", "keep_mm_feature_on_device": false, "enable_lora": null, "enable_lora_overlap_loading": null, "max_lora_rank": null, "lora_target_modules": null, "lora_paths": null, "max_loaded_loras": null, "max_loras_per_batch": 8, "lora_eviction_policy": "lru", "lora_backend": "csgmv", "max_lora_chunk_size": 16, "experts_shared_outer_loras": null, "lora_use_virtual_experts": false, "lora_strict_loading": false, "lora_drain_wait_threshold": 0.0, "enable_two_batch_overlap": false, "enable_single_batch_overlap": false, "tbo_token_distribution_threshold": 0.48, "cpu_offload_gb": 0, "offload_group_size": -1, "offload_num_in_group": 1, "offload_prefetch_step": 1, "offload_mode": "cpu", "enable_lmcache": false, "lmcache_config_file": null, "enable_flexkv": false, "flexkv_config_file": null, "kt_weight_path": null, "kt_method": "AMXINT4", "kt_cpuinfer": null, "kt_threadpool_count": 2, "kt_num_gpu_experts": null, "kt_max_deferred_experts_per_token": null, "dllm_algorithm": null, "dllm_algorithm_config": null, "dllm_fdfo": true, "disaggregation_mode": "prefill", "disaggregation_transfer_backend": "mooncake", "disaggregation_bootstrap_port": 28800, "disaggregation_ib_device": "mlx5_0,mlx5_1,mlx5_2,mlx5_3", "disaggregation_decode_enable_radix_cache": false, "disaggregation_decode_enable_offload_kvcache": false, "num_reserved_decode_tokens": 512, "disaggregation_decode_extra_slots": null, "disaggregation_decode_polling_interval": 1, "optimistic_prefill_attempts": 0, "encoder_only": false, "language_only": false, "encoder_transfer_backend": "zmq_to_tokenizer", "encoder_urls": [], "encoder_bootstrap_port": 8997, "encoder_register_urls": [], "enable_adaptive_dispatch_to_encoder": false, "enable_pdmux": false, "pdmux_config_path": null, "sm_group_num": 8, "custom_weight_loader": [], "weight_loader_disable_mmap": false, "weight_loader_prefetch_checkpoints": false, "weight_loader_prefetch_num_threads": 4, "weight_loader_drop_cache_after_load": false, "remote_instance_weight_loader_seed_instance_ip": null, "remote_instance_weight_loader_seed_instance_service_port": null, "remote_instance_weight_loader_send_weights_group_ports": null, "remote_instance_weight_loader_backend": "nccl", "remote_instance_weight_loader_start_seed_via_transfer_engine": false, "engine_info_bootstrap_port": 6789, "modelexpress_config": null, "download_dir": null, "model_checksum": null, "delete_ckpt_after_loading": false, "decrypted_config_file": null, "decrypted_draft_config_file": null, "checkpoint_engine_wait_weights_before_ready": false, "enable_prefill_delayer": false, "prefill_delayer_max_delay_passes": 30, "prefill_delayer_token_usage_low_watermark": null, "prefill_delayer_forward_passes_buckets": null, "prefill_delayer_wait_seconds_buckets": null, "prefill_delayer_queue_min_ratio": null, "prefill_delayer_max_delay_ms": null, "min_free_slots_delay": null, "enable_deterministic_inference": false, "rl_on_policy_target": null, "kv_canary": "none", "kv_canary_real_data": "none", "kv_canary_sweep_interval": 0, "enable_dynamic_batch_tokenizer": false, "dynamic_batch_tokenizer_batch_size": 32, "dynamic_batch_tokenizer_batch_timeout": 0.002, "enable_tokenizer_batch_encode": false, "disable_tokenizer_batch_decode": false, "debug_tensor_dump_output_folder": null, "debug_tensor_dump_layers": null, "debug_tensor_dump_input_file": null, "enable_memory_saver": false, "enable_weights_cpu_backup": false, "enable_draft_weights_cpu_backup": false, "enable_custom_logit_processor": false, "enable_return_hidden_states": false, "enable_return_routed_experts": false, "enable_return_indexer_topk": false, "disable_outlines_disk_cache": false, "enable_mis": false, "weight_cache_mode": "off", "weight_cache_socket": null, "weight_cache_timeout": 1800, "forward_hooks": null, "msprobe_dump_config": null, "status": "ready", "max_total_num_tokens": 3801408, "max_req_input_len": 1048570, "internal_states": [{"model_path": "/data/hf_models/Kimi-K3", "tokenizer_path": "/data/hf_models/Kimi-K3", "tokenizer_mode": "auto", "tokenizer_backend": "huggingface", "tokenizer_worker_num": 1, "detokenizer_worker_num": 1, "skip_tokenizer_init": false, "load_format": "auto", "model_loader_extra_config": "{}", "trust_remote_code": true, "context_length": null, "is_embedding": false, "enable_multimodal": null, "revision": null, "model_impl": "auto", "model_config_parser": "auto", "json_model_override_args": "{}", "dtype": "auto", "quantization": null, "quantization_param_path": null, "kv_cache_dtype": "auto", "enable_fp32_lm_head": false, "modelopt_quant": null, "modelopt_checkpoint_restore_path": null, "modelopt_checkpoint_save_path": null, "modelopt_export_path": null, "quantize_and_serve": false, "rl_quant_profile": null, "enable_tf32_matmul": false, "mem_fraction_static": 0.88, "max_running_requests": null, "max_queued_requests": null, "max_total_tokens": null, "chunked_prefill_size": 8192, "enable_dynamic_chunking": false, "max_prefill_tokens": 16384, "prefill_max_requests": null, "schedule_policy": "fcfs", "enable_priority_scheduling": false, "disable_priority_preemption": false, "default_priority_value": null, "abort_on_priority_when_disabled": false, "schedule_low_priority_values_first": false, "priority_scheduling_preemption_threshold": 10, "retraction_policy": "length", "schedule_conservativeness": 1.0, "page_size": 64, "swa_full_tokens_ratio": 0.8, "disable_hybrid_swa_memory": false, "radix_eviction_policy": "lru", "prefill_only_disable_kv_cache": false, "disable_radix_cache": true, "enable_page_major_kv_layout": false, "enable_unified_memory": false, "disable_chunked_prefix_cache": true, "disable_overlap_schedule": true, "num_continuous_decode_steps": 1, "scheduler_recv_interval": 1, "enable_mixed_chunk": false, "nccl_port": null, "dist_timeout": 3600, "dist_init_addr": "174.1.60.1:20000", "nnodes": 4, "node_rank": 0, "tp_size": 4, "dcp_size": 1, "pp_size": 8, "pp_max_micro_batch_size": 6, "pp_async_batch_depth": 0, "dp_size": 1, "load_balance_method": "follow_bootstrap_room", "attn_cp_size": 1, "moe_dp_size": 1, "dcp_comm_backend": "ag_rs", "dcp_replicate_q_proj": null, "dwdp_size": 1, "enable_prefill_cp": false, "cp_strategy": null, "enable_dsa_cache_layer_split": false, "enable_dsa_prefill_context_parallel": false, "dsa_prefill_cp_mode": "round-robin-split", "enable_prefill_context_parallel": false, "prefill_cp_mode": "in-seq-split", "enable_cp_decode_attn_tp": false, "enable_dp_attention": false, "enable_dp_attention_local_control_broadcast": false, "enable_dp_lm_head": false, "enable_attn_tp_input_scattered": false, "disable_attn_tp_gather": false, "enable_p2p_check": false, "device": "cuda", "base_gpu_id": 0, "gpu_id_step": 1, "random_seed": 984607766, "watchdog_timeout": 300, "soft_watchdog_timeout": null, "sleep_on_idle": false, "use_ray": false, "numa_node": null, "gc_threshold": null, "host": "0.0.0.0", "port": 30000, "fastapi_root_path": "", "smg_grpc_mode": false, "grpc_mode": false, "grpc_port": null, "sidecar": null, "sidecar_args": null, "skip_server_warmup": false, "warmups": null, "enable_http2": false, "ssl_keyfile": null, "ssl_certfile": null, "ssl_ca_certs": null, "ssl_keyfile_password": null, "enable_ssl_refresh": false, "api_key": null, "admin_api_key": null, "served_model_name": "kimi-k3", "weight_version": "default", "chat_template": null, "hf_chat_template_name": null, "completion_template": null, "file_storage_path": "sglang_storage", "enable_cache_report": false, "reasoning_parser": null, "default_chat_template_kwargs": null, "strip_thinking_cache": false, "enable_strict_thinking": false, "tool_call_parser": null, "tool_server": null, "sampling_defaults": "model", "asr_max_buffer_seconds": 60, "asr_max_concurrent_sessions": 32, "preferred_sampling_params": null, "allow_auto_truncate": false, "stream_interval": 1, "batch_notify_size": 16, "stream_response_default_include_usage": false, "incremental_streaming_output": false, "enable_streaming_session": false, "enable_session_radix_cache": false, "log_level": "info", "log_level_http": null, "log_requests": false, "log_requests_level": 2, "log_requests_format": "text", "log_requests_target": null, "uvicorn_access_log_exclude_prefixes": [], "crash_dump_folder": null, "show_time_cost": false, "enable_metrics": false, "smg_http_sidecar_port": null, "enable_mfu_metrics": false, "enable_metrics_for_all_schedulers": false, "load_snapshot_publish_interval": 15, "tokenizer_metrics_custom_labels_header": "x-custom-labels", "tokenizer_metrics_allowed_custom_labels": null, "extra_metric_labels": null, "bucket_time_to_first_token": null, "bucket_inter_token_latency": null, "bucket_e2e_request_latency": null, "prompt_tokens_buckets": null, "generation_tokens_buckets": null, "gc_warning_threshold_secs": 0.0, "decode_log_interval": 40, "enable_request_time_stats_logging": false, "kv_events_config": null, "enable_forward_pass_metrics": false, "forward_pass_metrics_worker_id": "", "forward_pass_metrics_ipc_name": null, "enable_trace": false, "trace_modules": "request", "otlp_traces_endpoint": "localhost:4317", "export_metrics_to_file": false, "export_metrics_to_file_dir": null, "stat_loggers": null, "constrained_json_whitespace_pattern": null, "constrained_json_disable_any_whitespace": false, "attention_backend": "triton", "decode_attention_backend": null, "prefill_attention_backend": null, "sampling_backend": "flashinfer", "grammar_backend": "xgrammar", "radix_cache_backend": null, "mm_attention_backend": null, "fp8_gemm_runner_backend": "auto", "fp4_gemm_runner_backend": "auto", "bf16_gemm_backend": "auto", "dsa_prefill_backend": null, "dsa_decode_backend": null, "dsa_paged_mqa_logits_backend": "auto", "dsa_topk_backend": "sgl-kernel", "disable_flashinfer_autotune": false, "flashinfer_autotune_skip_ops": null, "mamba_backend": "triton", "cuda_graph_config": {"decode": {"backend": "disabled", "max_bs": 16, "bs": [1, 2, 4, 8, 12, 16], "tc_compiler": "eager", "full_prefill_max_req": null}, "prefill": {"backend": "disabled", "max_bs": 2048, "bs": [4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], "tc_compiler": "eager", "full_prefill_max_req": null}}, "cuda_graph_backend_decode": null, "cuda_graph_backend_prefill": null, "cuda_graph_max_bs_decode": 16, "cuda_graph_max_bs_prefill": null, "cuda_graph_bs_decode": null, "cuda_graph_bs_prefill": null, "cuda_graph_tc_compiler": null, "disable_prefill_cuda_graph": false, "disable_decode_cuda_graph": false, "disable_cuda_graph": false, "disable_cuda_graph_padding": false, "enable_profile_cuda_graph": false, "enable_cudagraph_gc": false, "debug_cuda_graph": false, "enable_layerwise_nvtx_marker": false, "enable_nccl_nvls": false, "enable_symm_mem": false, "triton_attention_reduce_in_fp32": false, "triton_attention_num_kv_splits": 8, "triton_attention_split_tile_size": null, "flashinfer_mla_disable_ragged": false, "enable_fused_qk_norm_rope": false, "enable_precise_embedding_interpolation": false, "enable_fused_moe_sum_all_reduce": false, "enable_deepseek_v4_fp4_indexer": false, "disable_custom_all_reduce": false, "enable_mscclpp": false, "enable_torch_symm_mem": false, "enable_scattered_sconv": false, "pre_warm_nccl": false, "enable_quant_communications": false, "enable_flashinfer_allreduce_fusion": false, "enforce_disable_flashinfer_allreduce_fusion": false, "flashinfer_allreduce_fusion_backend": null, "enable_aiter_allreduce_fusion": false, "enable_torch_compile": false, "enable_torch_compile_debug_mode": false, "torch_compile_max_bs": 32, "torchao_config": "", "speculative_algorithm": null, "speculative_draft_model_path": null, "speculative_draft_model_revision": null, "speculative_draft_load_format": null, "speculative_num_steps": null, "speculative_eagle_topk": null, "speculative_num_draft_tokens": null, "speculative_dflash_block_size": null, "speculative_dspark_block_size": null, "speculative_dspark_sps_table_path": null, "speculative_dspark_confidence_sts_path": null, "speculative_dspark_align_verify_tokens_to_graph_tier": false, "speculative_accept_threshold_single": 1.0, "speculative_accept_threshold_acc": 1.0, "speculative_use_rejection_sampling": false, "speculative_token_map": null, "speculative_attention_mode": "prefill", "speculative_draft_attention_backend": null, "speculative_draft_window_size": null, "speculative_moe_runner_backend": "flashinfer_mxfp4", "speculative_moe_a2a_backend": null, "speculative_draft_model_quantization": null, "speculative_skip_dp_mlp_sync": false, "enable_multi_layer_eagle": false, "speculative_adaptive": false, "speculative_adaptive_config": null, "decoupled_spec_bind_endpoint": null, "decoupled_spec_connect_endpoints": null, "decoupled_spec_rank": null, "decoupled_spec_role": "null", "spec_trace_dir": null, "speculative_ngram_min_bfs_breadth": 1, "speculative_ngram_max_bfs_breadth": 10, "speculative_ngram_match_type": "BFS", "speculative_ngram_max_trie_depth": 18, "speculative_ngram_capacity": 10000000, "speculative_ngram_external_corpus_path": null, "speculative_ngram_external_sam_budget": 0, "speculative_ngram_external_corpus_max_tokens": 10000000, "ep_size": 4, "moe_a2a_backend": "none", "moe_runner_backend": "flashinfer_mxfp4", "flashinfer_mxfp4_moe_precision": "default", "deepep_mode": "auto", "fuseep_mode": 2, "deepep_dispatcher_output_dtype": "auto", "ep_num_redundant_experts": 0, "ep_dispatch_algorithm": null, "init_expert_location": "trivial", "enable_eplb": false, "eplb_algorithm": "auto", "eplb_rebalance_num_iterations": 1000, "eplb_rebalance_layers_per_chunk": null, "eplb_min_rebalancing_utilization_threshold": 1.0, "expert_distribution_recorder_mode": null, "expert_distribution_recorder_buffer_size": 1000, "enable_expert_distribution_metrics": false, "deepep_config": null, "moe_dense_tp_size": null, "elastic_ep_backend": null, "enable_elastic_expert_backup": false, "mooncake_ib_device": null, "enable_waterfill": false, "ep_join_mode": null, "ep_join_rank_offset": 0, "elastic_ep_initial_size": null, "max_ep_size": null, "elastic_ep_scale_timeout": 600, "elastic_ep_rejoin": false, "disable_flashinfer_cutlass_moe_fp4_allgather": false, "disable_shared_experts_fusion": false, "enforce_shared_experts_fusion": false, "max_mamba_cache_size": 54, "mamba_ssm_dtype": null, "mamba_max_states_per_path": -1, "enable_mamba_cache_stochastic_rounding": false, "mamba_cache_philox_rounds": 0, "mamba_full_memory_ratio": 0.36, "mamba_radix_cache_strategy": "extra_buffer_lazy", "uses_mamba_radix_cache": false, "mamba_track_interval": 256, "enable_int8_mamba_checkpoint": false, "int8_mamba_ckpt_size": null, "linear_attn_backend": "triton", "linear_attn_decode_backend": null, "linear_attn_prefill_backend": null, "linear_attn_verify_backend": null, "enable_linear_replayssm": false, "linear_replayssm_cache_len": 16, "enable_linear_replayssm_spec": false, "enable_hierarchical_cache": false, "hicache_ratio": 2.0, "hicache_size": 0, "hicache_write_policy": "write_through", "hicache_io_backend": "kernel", "hicache_mem_layout": "page_first", "hicache_storage_backend": null, "hicache_storage_prefetch_policy": "timeout", "hicache_storage_backend_extra_config": null, "enable_hisparse": false, "hisparse_config": null, "enable_broadcast_mm_inputs_process": false, "enable_prefix_mm_cache": false, "mm_enable_dp_encoder": false, "mm_process_config": {}, "mm_processor_worker_num": 0, "mm_io_worker_num": 0, "limit_mm_data_per_request": null, "enable_mm_global_cache": false, "disable_fast_image_processor": false, "mm_feature_transport": "cpu", "keep_mm_feature_on_device": false, "enable_lora": null, "enable_lora_overlap_loading": null, "max_lora_rank": null, "lora_target_modules": null, "lora_paths": null, "max_loaded_loras": null, "max_loras_per_batch": 8, "lora_eviction_policy": "lru", "lora_backend": "csgmv", "max_lora_chunk_size": 16, "experts_shared_outer_loras": null, "lora_use_virtual_experts": false, "lora_strict_loading": false, "lora_drain_wait_threshold": 0.0, "enable_two_batch_overlap": false, "enable_single_batch_overlap": false, "tbo_token_distribution_threshold": 0.48, "cpu_offload_gb": 0, "offload_group_size": -1, "offload_num_in_group": 1, "offload_prefetch_step": 1, "offload_mode": "cpu", "enable_lmcache": false, "lmcache_config_file": null, "enable_flexkv": false, "flexkv_config_file": null, "kt_weight_path": null, "kt_method": "AMXINT4", "kt_cpuinfer": null, "kt_threadpool_count": 2, "kt_num_gpu_experts": null, "kt_max_deferred_experts_per_token": null, "dllm_algorithm": null, "dllm_algorithm_config": null, "dllm_fdfo": true, "disaggregation_mode": "prefill", "disaggregation_transfer_backend": "mooncake", "disaggregation_bootstrap_port": 28800, "disaggregation_ib_device": "mlx5_0,mlx5_1,mlx5_2,mlx5_3", "disaggregation_decode_enable_radix_cache": false, "disaggregation_decode_enable_offload_kvcache": false, "num_reserved_decode_tokens": 512, "disaggregation_decode_extra_slots": null, "disaggregation_decode_polling_interval": 1, "optimistic_prefill_attempts": 0, "encoder_only": false, "language_only": false, "encoder_transfer_backend": "zmq_to_tokenizer", "encoder_urls": [], "encoder_bootstrap_port": 8997, "encoder_register_urls": [], "enable_adaptive_dispatch_to_encoder": false, "enable_pdmux": false, "pdmux_config_path": null, "sm_group_num": 8, "custom_weight_loader": [], "weight_loader_disable_mmap": false, "weight_loader_prefetch_checkpoints": false, "weight_loader_prefetch_num_threads": 4, "weight_loader_drop_cache_after_load": false, "remote_instance_weight_loader_seed_instance_ip": null, "remote_instance_weight_loader_seed_instance_service_port": null, "remote_instance_weight_loader_send_weights_group_ports": null, "remote_instance_weight_loader_backend": "nccl", "remote_instance_weight_loader_start_seed_via_transfer_engine": false, "engine_info_bootstrap_port": 6789, "modelexpress_config": null, "download_dir": null, "model_checksum": null, "delete_ckpt_after_loading": false, "decrypted_config_file": null, "decrypted_draft_config_file": null, "checkpoint_engine_wait_weights_before_ready": false, "enable_prefill_delayer": false, "prefill_delayer_max_delay_passes": 30, "prefill_delayer_token_usage_low_watermark": null, "prefill_delayer_forward_passes_buckets": null, "prefill_delayer_wait_seconds_buckets": null, "prefill_delayer_queue_min_ratio": null, "prefill_delayer_max_delay_ms": null, "min_free_slots_delay": null, "enable_deterministic_inference": false, "rl_on_policy_target": null, "kv_canary": "none", "kv_canary_real_data": "none", "kv_canary_sweep_interval": 0, "enable_dynamic_batch_tokenizer": false, "dynamic_batch_tokenizer_batch_size": 32, "dynamic_batch_tokenizer_batch_timeout": 0.002, "enable_tokenizer_batch_encode": false, "disable_tokenizer_batch_decode": false, "debug_tensor_dump_output_folder": null, "debug_tensor_dump_layers": null, "debug_tensor_dump_input_file": null, "enable_memory_saver": false, "enable_weights_cpu_backup": false, "enable_draft_weights_cpu_backup": false, "enable_custom_logit_processor": false, "enable_return_hidden_states": false, "enable_return_routed_experts": false, "enable_return_indexer_topk": false, "disable_outlines_disk_cache": false, "enable_mis": false, "weight_cache_mode": "off", "weight_cache_socket": null, "weight_cache_timeout": 1800, "forward_hooks": null, "msprobe_dump_config": null, "_resolved_overrides": [["_sampling_backend_default", {"sampling_backend": "flashinfer"}], ["_attention_backend_default", {"attention_backend": "triton"}], ["_data_parallelism_defaults", {"enable_dp_attention": false, "enable_dp_lm_head": false}], ["_pipeline_parallel_overlap_disable", {"disable_overlap_schedule": true}], ["_speculative_moe_runner_default", {"speculative_moe_runner_backend": "flashinfer_mxfp4"}]], "grpc_worker_threads": 4, "_quantization_explicitly_unset": false, "_cuda_graph_config_locked": [["decode", "max_bs"]], "_declarations_materialized": true, "_runtime_mutations": [["model_runner.chunked_prefix_cache_gate", {"disable_chunked_prefix_cache": true}], ["mamba_pool.memory_budget", {"max_mamba_cache_size": 54}]], "_in_override": false, "_mx_config_cache": {}, "max_speculative_num_draft_tokens": null, "last_gen_throughput": 0.0, "memory_usage": {"weight": 43.09, "kvcache": 8.16, "token_capacity": 3801408, "graph": 0}, "effective_max_running_requests_per_dp": 54}], "version": "0.5.16", "kv_events": null}], "decode": [{"model_path": "/data/hf_models/Kimi-K3", "tokenizer_path": "/data/hf_models/Kimi-K3", "tokenizer_mode": "auto", "tokenizer_backend": "huggingface", "tokenizer_worker_num": 1, "detokenizer_worker_num": 1, "skip_tokenizer_init": false, "load_format": "auto", "model_loader_extra_config": "{}", "trust_remote_code": true, "context_length": null, "is_embedding": false, "enable_multimodal": null, "revision": null, "model_impl": "auto", "model_config_parser": "auto", "json_model_override_args": "{}", "dtype": "auto", "quantization": null, "quantization_param_path": null, "kv_cache_dtype": "auto", "enable_fp32_lm_head": false, "modelopt_quant": null, "modelopt_checkpoint_restore_path": null, "modelopt_checkpoint_save_path": null, "modelopt_export_path": null, "quantize_and_serve": false, "rl_quant_profile": null, "enable_tf32_matmul": false, "mem_fraction_static": 0.88, "max_running_requests": null, "max_queued_requests": null, "max_total_tokens": null, "chunked_prefill_size": 8192, "enable_dynamic_chunking": false, "max_prefill_tokens": 16384, "prefill_max_requests": null, "schedule_policy": "fcfs", "enable_priority_scheduling": false, "disable_priority_preemption": false, "default_priority_value": null, "abort_on_priority_when_disabled": false, "schedule_low_priority_values_first": false, "priority_scheduling_preemption_threshold": 10, "retraction_policy": "length", "schedule_conservativeness": 1.0, "page_size": 64, "swa_full_tokens_ratio": 0.8, "disable_hybrid_swa_memory": false, "radix_eviction_policy": "lru", "prefill_only_disable_kv_cache": false, "disable_radix_cache": true, "enable_page_major_kv_layout": false, "enable_unified_memory": false, "disable_chunked_prefix_cache": false, "disable_overlap_schedule": false, "num_continuous_decode_steps": 1, "scheduler_recv_interval": 1, "enable_mixed_chunk": false, "nccl_port": null, "dist_timeout": 3600, "dist_init_addr": "174.1.60.5:20000", "nnodes": 4, "node_rank": 0, "tp_size": 32, "dcp_size": 1, "pp_size": 1, "pp_max_micro_batch_size": null, "pp_async_batch_depth": 0, "dp_size": 1, "load_balance_method": "round_robin", "attn_cp_size": 1, "moe_dp_size": 1, "dcp_comm_backend": "ag_rs", "dcp_replicate_q_proj": null, "dwdp_size": 1, "enable_prefill_cp": false, "cp_strategy": null, "enable_dsa_cache_layer_split": false, "enable_dsa_prefill_context_parallel": false, "dsa_prefill_cp_mode": "round-robin-split", "enable_prefill_context_parallel": false, "prefill_cp_mode": "in-seq-split", "enable_cp_decode_attn_tp": false, "enable_dp_attention": false, "enable_dp_attention_local_control_broadcast": false, "enable_dp_lm_head": false, "enable_attn_tp_input_scattered": false, "disable_attn_tp_gather": false, "enable_p2p_check": false, "device": "cuda", "base_gpu_id": 0, "gpu_id_step": 1, "random_seed": 369235288, "watchdog_timeout": 300, "soft_watchdog_timeout": null, "sleep_on_idle": false, "use_ray": false, "custom_sigquit_handler": null, "numa_node": null, "gc_threshold": null, "host": "0.0.0.0", "port": 30000, "fastapi_root_path": "", "smg_grpc_mode": false, "grpc_mode": false, "grpc_port": null, "sidecar": null, "sidecar_args": null, "skip_server_warmup": false, "warmups": null, "enable_http2": false, "ssl_keyfile": null, "ssl_certfile": null, "ssl_ca_certs": null, "ssl_keyfile_password": null, "enable_ssl_refresh": false, "api_key": null, "admin_api_key": null, "served_model_name": "kimi-k3", "weight_version": "default", "chat_template": null, "hf_chat_template_name": null, "completion_template": null, "file_storage_path": "sglang_storage", "enable_cache_report": false, "reasoning_parser": null, "default_chat_template_kwargs": null, "strip_thinking_cache": false, "enable_strict_thinking": false, "tool_call_parser": null, "tool_server": null, "sampling_defaults": "model", "asr_max_buffer_seconds": 60, "asr_max_concurrent_sessions": 32, "preferred_sampling_params": null, "allow_auto_truncate": false, "stream_interval": 1, "batch_notify_size": 16, "stream_response_default_include_usage": false, "incremental_streaming_output": false, "enable_streaming_session": false, "enable_session_radix_cache": false, "log_level": "info", "log_level_http": null, "log_requests": false, "log_requests_level": 2, "log_requests_format": "text", "log_requests_target": null, "uvicorn_access_log_exclude_prefixes": [], "crash_dump_folder": null, "show_time_cost": false, "enable_metrics": false, "smg_http_sidecar_port": null, "enable_mfu_metrics": false, "enable_metrics_for_all_schedulers": false, "load_snapshot_publish_interval": 15, "tokenizer_metrics_custom_labels_header": "x-custom-labels", "tokenizer_metrics_allowed_custom_labels": null, "extra_metric_labels": null, "bucket_time_to_first_token": null, "bucket_inter_token_latency": null, "bucket_e2e_request_latency": null, "prompt_tokens_buckets": null, "generation_tokens_buckets": null, "gc_warning_threshold_secs": 0.0, "decode_log_interval": 40, "enable_request_time_stats_logging": false, "kv_events_config": null, "enable_forward_pass_metrics": false, "forward_pass_metrics_worker_id": "", "forward_pass_metrics_ipc_name": null, "enable_trace": false, "trace_modules": "request", "otlp_traces_endpoint": "localhost:4317", "export_metrics_to_file": false, "export_metrics_to_file_dir": null, "stat_loggers": null, "constrained_json_whitespace_pattern": null, "constrained_json_disable_any_whitespace": false, "attention_backend": "triton", "decode_attention_backend": null, "prefill_attention_backend": null, "sampling_backend": "flashinfer", "grammar_backend": "xgrammar", "radix_cache_backend": null, "mm_attention_backend": null, "fp8_gemm_runner_backend": "auto", "fp4_gemm_runner_backend": "auto", "bf16_gemm_backend": "auto", "dsa_prefill_backend": null, "dsa_decode_backend": null, "dsa_paged_mqa_logits_backend": "auto", "dsa_topk_backend": "sgl-kernel", "disable_flashinfer_autotune": false, "flashinfer_autotune_skip_ops": null, "mamba_backend": "triton", "cuda_graph_config": {"decode": {"backend": "full", "max_bs": 16, "bs": [1, 2, 4, 8, 12, 16], "tc_compiler": "eager", "full_prefill_max_req": null}, "prefill": {"backend": "disabled", "max_bs": 2048, "bs": [4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], "tc_compiler": "eager", "full_prefill_max_req": null}}, "cuda_graph_backend_decode": null, "cuda_graph_backend_prefill": null, "cuda_graph_max_bs_decode": 16, "cuda_graph_max_bs_prefill": null, "cuda_graph_bs_decode": null, "cuda_graph_bs_prefill": null, "cuda_graph_tc_compiler": null, "disable_prefill_cuda_graph": false, "disable_decode_cuda_graph": false, "disable_cuda_graph": false, "disable_cuda_graph_padding": false, "enable_profile_cuda_graph": false, "enable_cudagraph_gc": false, "debug_cuda_graph": false, "enable_layerwise_nvtx_marker": false, "enable_nccl_nvls": false, "enable_symm_mem": false, "triton_attention_reduce_in_fp32": false, "triton_attention_num_kv_splits": 8, "triton_attention_split_tile_size": null, "flashinfer_mla_disable_ragged": false, "enable_fused_qk_norm_rope": false, "enable_precise_embedding_interpolation": false, "enable_fused_moe_sum_all_reduce": false, "enable_deepseek_v4_fp4_indexer": false, "disable_custom_all_reduce": false, "enable_mscclpp": false, "enable_torch_symm_mem": false, "enable_scattered_sconv": false, "pre_warm_nccl": false, "enable_quant_communications": false, "enable_flashinfer_allreduce_fusion": false, "enforce_disable_flashinfer_allreduce_fusion": false, "flashinfer_allreduce_fusion_backend": null, "enable_aiter_allreduce_fusion": false, "enable_torch_compile": false, "enable_torch_compile_debug_mode": false, "torch_compile_max_bs": 32, "torchao_config": "", "speculative_algorithm": null, "speculative_draft_model_path": null, "speculative_draft_model_revision": null, "speculative_draft_load_format": null, "speculative_num_steps": null, "speculative_eagle_topk": null, "speculative_num_draft_tokens": null, "speculative_dflash_block_size": null, "speculative_dspark_block_size": null, "speculative_dspark_sps_table_path": null, "speculative_dspark_confidence_sts_path": null, "speculative_dspark_align_verify_tokens_to_graph_tier": false, "speculative_accept_threshold_single": 1.0, "speculative_accept_threshold_acc": 1.0, "speculative_use_rejection_sampling": false, "speculative_token_map": null, "speculative_attention_mode": "prefill", "speculative_draft_attention_backend": null, "speculative_draft_window_size": null, "speculative_moe_runner_backend": "marlin", "speculative_moe_a2a_backend": null, "speculative_draft_model_quantization": null, "speculative_skip_dp_mlp_sync": false, "enable_multi_layer_eagle": false, "speculative_adaptive": false, "speculative_adaptive_config": null, "decoupled_spec_bind_endpoint": null, "decoupled_spec_connect_endpoints": null, "decoupled_spec_rank": null, "decoupled_spec_role": "null", "spec_trace_dir": null, "speculative_ngram_min_bfs_breadth": 1, "speculative_ngram_max_bfs_breadth": 10, "speculative_ngram_match_type": "BFS", "speculative_ngram_max_trie_depth": 18, "speculative_ngram_capacity": 10000000, "speculative_ngram_external_corpus_path": null, "speculative_ngram_external_sam_budget": 0, "speculative_ngram_external_corpus_max_tokens": 10000000, "ep_size": 32, "moe_a2a_backend": "none", "moe_runner_backend": "marlin", "flashinfer_mxfp4_moe_precision": "default", "deepep_mode": "auto", "fuseep_mode": 2, "deepep_dispatcher_output_dtype": "auto", "ep_num_redundant_experts": 0, "ep_dispatch_algorithm": null, "init_expert_location": "trivial", "enable_eplb": false, "eplb_algorithm": "auto", "eplb_rebalance_num_iterations": 1000, "eplb_rebalance_layers_per_chunk": null, "eplb_min_rebalancing_utilization_threshold": 1.0, "expert_distribution_recorder_mode": null, "expert_distribution_recorder_buffer_size": 1000, "enable_expert_distribution_metrics": false, "deepep_config": null, "moe_dense_tp_size": null, "elastic_ep_backend": null, "enable_elastic_expert_backup": false, "mooncake_ib_device": null, "enable_waterfill": false, "ep_join_mode": null, "ep_join_rank_offset": 0, "elastic_ep_initial_size": null, "max_ep_size": null, "elastic_ep_scale_timeout": 600, "elastic_ep_rejoin": false, "disable_flashinfer_cutlass_moe_fp4_allgather": false, "disable_shared_experts_fusion": false, "enforce_shared_experts_fusion": false, "max_mamba_cache_size": null, "mamba_ssm_dtype": null, "mamba_max_states_per_path": -1, "enable_mamba_cache_stochastic_rounding": false, "mamba_cache_philox_rounds": 0, "mamba_full_memory_ratio": 0.21, "mamba_radix_cache_strategy": "extra_buffer_lazy", "uses_mamba_radix_cache": false, "mamba_track_interval": 256, "enable_int8_mamba_checkpoint": false, "int8_mamba_ckpt_size": null, "linear_attn_backend": "triton", "linear_attn_decode_backend": null, "linear_attn_prefill_backend": null, "linear_attn_verify_backend": null, "enable_linear_replayssm": false, "linear_replayssm_cache_len": 16, "enable_linear_replayssm_spec": false, "enable_hierarchical_cache": false, "hicache_ratio": 2.0, "hicache_size": 0, "hicache_write_policy": "write_through", "hicache_io_backend": "kernel", "hicache_mem_layout": "page_first", "hicache_storage_backend": null, "hicache_storage_prefetch_policy": "timeout", "hicache_storage_backend_extra_config": null, "enable_hisparse": false, "hisparse_config": null, "enable_broadcast_mm_inputs_process": false, "enable_prefix_mm_cache": false, "mm_enable_dp_encoder": false, "mm_process_config": {}, "mm_processor_worker_num": 0, "mm_io_worker_num": 0, "limit_mm_data_per_request": null, "enable_mm_global_cache": false, "disable_fast_image_processor": false, "mm_feature_transport": "cpu", "keep_mm_feature_on_device": false, "enable_lora": null, "enable_lora_overlap_loading": null, "max_lora_rank": null, "lora_target_modules": null, "lora_paths": null, "max_loaded_loras": null, "max_loras_per_batch": 8, "lora_eviction_policy": "lru", "lora_backend": "csgmv", "max_lora_chunk_size": 16, "experts_shared_outer_loras": null, "lora_use_virtual_experts": false, "lora_strict_loading": false, "lora_drain_wait_threshold": 0.0, "enable_two_batch_overlap": false, "enable_single_batch_overlap": false, "tbo_token_distribution_threshold": 0.48, "cpu_offload_gb": 0, "offload_group_size": -1, "offload_num_in_group": 1, "offload_prefetch_step": 1, "offload_mode": "cpu", "enable_lmcache": false, "lmcache_config_file": null, "enable_flexkv": false, "flexkv_config_file": null, "kt_weight_path": null, "kt_method": "AMXINT4", "kt_cpuinfer": null, "kt_threadpool_count": 2, "kt_num_gpu_experts": null, "kt_max_deferred_experts_per_token": null, "dllm_algorithm": null, "dllm_algorithm_config": null, "dllm_fdfo": true, "disaggregation_mode": "decode", "disaggregation_transfer_backend": "mooncake", "disaggregation_bootstrap_port": 28800, "disaggregation_ib_device": "mlx5_0,mlx5_1,mlx5_2,mlx5_3", "disaggregation_decode_enable_radix_cache": false, "disaggregation_decode_enable_offload_kvcache": false, "num_reserved_decode_tokens": 512, "disaggregation_decode_extra_slots": 0, "disaggregation_decode_polling_interval": 1, "optimistic_prefill_attempts": 0, "encoder_only": false, "language_only": false, "encoder_transfer_backend": "zmq_to_tokenizer", "encoder_urls": [], "encoder_bootstrap_port": 8997, "encoder_register_urls": [], "enable_adaptive_dispatch_to_encoder": false, "enable_pdmux": false, "pdmux_config_path": null, "sm_group_num": 8, "custom_weight_loader": [], "weight_loader_disable_mmap": false, "weight_loader_prefetch_checkpoints": false, "weight_loader_prefetch_num_threads": 4, "weight_loader_drop_cache_after_load": false, "remote_instance_weight_loader_seed_instance_ip": null, "remote_instance_weight_loader_seed_instance_service_port": null, "remote_instance_weight_loader_send_weights_group_ports": null, "remote_instance_weight_loader_backend": "nccl", "remote_instance_weight_loader_start_seed_via_transfer_engine": false, "engine_info_bootstrap_port": 6789, "modelexpress_config": null, "download_dir": null, "model_checksum": null, "delete_ckpt_after_loading": false, "decrypted_config_file": null, "decrypted_draft_config_file": null, "checkpoint_engine_wait_weights_before_ready": false, "enable_prefill_delayer": false, "prefill_delayer_max_delay_passes": 30, "prefill_delayer_token_usage_low_watermark": null, "prefill_delayer_forward_passes_buckets": null, "prefill_delayer_wait_seconds_buckets": null, "prefill_delayer_queue_min_ratio": null, "prefill_delayer_max_delay_ms": null, "min_free_slots_delay": null, "enable_deterministic_inference": false, "rl_on_policy_target": null, "kv_canary": "none", "kv_canary_real_data": "none", "kv_canary_sweep_interval": 0, "enable_dynamic_batch_tokenizer": false, "dynamic_batch_tokenizer_batch_size": 32, "dynamic_batch_tokenizer_batch_timeout": 0.002, "enable_tokenizer_batch_encode": false, "disable_tokenizer_batch_decode": false, "debug_tensor_dump_output_folder": null, "debug_tensor_dump_layers": null, "debug_tensor_dump_input_file": null, "enable_memory_saver": false, "enable_weights_cpu_backup": false, "enable_draft_weights_cpu_backup": false, "enable_custom_logit_processor": false, "enable_return_hidden_states": false, "enable_return_routed_experts": false, "enable_return_indexer_topk": false, "disable_outlines_disk_cache": false, "enable_mis": false, "weight_cache_mode": "off", "weight_cache_socket": null, "weight_cache_timeout": 1800, "forward_hooks": null, "msprobe_dump_config": null, "status": "ready", "max_total_num_tokens": 440256, "max_req_input_len": 440250, "internal_states": [{"model_path": "/data/hf_models/Kimi-K3", "tokenizer_path": "/data/hf_models/Kimi-K3", "tokenizer_mode": "auto", "tokenizer_backend": "huggingface", "tokenizer_worker_num": 1, "detokenizer_worker_num": 1, "skip_tokenizer_init": false, "load_format": "auto", "model_loader_extra_config": "{}", "trust_remote_code": true, "context_length": null, "is_embedding": false, "enable_multimodal": null, "revision": null, "model_impl": "auto", "model_config_parser": "auto", "json_model_override_args": "{}", "dtype": "auto", "quantization": null, "quantization_param_path": null, "kv_cache_dtype": "auto", "enable_fp32_lm_head": false, "modelopt_quant": null, "modelopt_checkpoint_restore_path": null, "modelopt_checkpoint_save_path": null, "modelopt_export_path": null, "quantize_and_serve": false, "rl_quant_profile": null, "enable_tf32_matmul": false, "mem_fraction_static": 0.88, "max_running_requests": null, "max_queued_requests": null, "max_total_tokens": null, "chunked_prefill_size": 8192, "enable_dynamic_chunking": false, "max_prefill_tokens": 16384, "prefill_max_requests": null, "schedule_policy": "fcfs", "enable_priority_scheduling": false, "disable_priority_preemption": false, "default_priority_value": null, "abort_on_priority_when_disabled": false, "schedule_low_priority_values_first": false, "priority_scheduling_preemption_threshold": 10, "retraction_policy": "length", "schedule_conservativeness": 1.0, "page_size": 64, "swa_full_tokens_ratio": 0.8, "disable_hybrid_swa_memory": false, "radix_eviction_policy": "lru", "prefill_only_disable_kv_cache": false, "disable_radix_cache": true, "enable_page_major_kv_layout": false, "enable_unified_memory": false, "disable_chunked_prefix_cache": true, "disable_overlap_schedule": false, "num_continuous_decode_steps": 1, "scheduler_recv_interval": 1, "enable_mixed_chunk": false, "nccl_port": null, "dist_timeout": 3600, "dist_init_addr": "174.1.60.5:20000", "nnodes": 4, "node_rank": 0, "tp_size": 32, "dcp_size": 1, "pp_size": 1, "pp_max_micro_batch_size": 181, "pp_async_batch_depth": 0, "dp_size": 1, "load_balance_method": "round_robin", "attn_cp_size": 1, "moe_dp_size": 1, "dcp_comm_backend": "ag_rs", "dcp_replicate_q_proj": null, "dwdp_size": 1, "enable_prefill_cp": false, "cp_strategy": null, "enable_dsa_cache_layer_split": false, "enable_dsa_prefill_context_parallel": false, "dsa_prefill_cp_mode": "round-robin-split", "enable_prefill_context_parallel": false, "prefill_cp_mode": "in-seq-split", "enable_cp_decode_attn_tp": false, "enable_dp_attention": false, "enable_dp_attention_local_control_broadcast": false, "enable_dp_lm_head": false, "enable_attn_tp_input_scattered": false, "disable_attn_tp_gather": false, "enable_p2p_check": false, "device": "cuda", "base_gpu_id": 0, "gpu_id_step": 1, "random_seed": 369235288, "watchdog_timeout": 300, "soft_watchdog_timeout": null, "sleep_on_idle": false, "use_ray": false, "numa_node": null, "gc_threshold": null, "host": "0.0.0.0", "port": 30000, "fastapi_root_path": "", "smg_grpc_mode": false, "grpc_mode": false, "grpc_port": null, "sidecar": null, "sidecar_args": null, "skip_server_warmup": false, "warmups": null, "enable_http2": false, "ssl_keyfile": null, "ssl_certfile": null, "ssl_ca_certs": null, "ssl_keyfile_password": null, "enable_ssl_refresh": false, "api_key": null, "admin_api_key": null, "served_model_name": "kimi-k3", "weight_version": "default", "chat_template": null, "hf_chat_template_name": null, "completion_template": null, "file_storage_path": "sglang_storage", "enable_cache_report": false, "reasoning_parser": null, "default_chat_template_kwargs": null, "strip_thinking_cache": false, "enable_strict_thinking": false, "tool_call_parser": null, "tool_server": null, "sampling_defaults": "model", "asr_max_buffer_seconds": 60, "asr_max_concurrent_sessions": 32, "preferred_sampling_params": null, "allow_auto_truncate": false, "stream_interval": 1, "batch_notify_size": 16, "stream_response_default_include_usage": false, "incremental_streaming_output": false, "enable_streaming_session": false, "enable_session_radix_cache": false, "log_level": "info", "log_level_http": null, "log_requests": false, "log_requests_level": 2, "log_requests_format": "text", "log_requests_target": null, "uvicorn_access_log_exclude_prefixes": [], "crash_dump_folder": null, "show_time_cost": false, "enable_metrics": false, "smg_http_sidecar_port": null, "enable_mfu_metrics": false, "enable_metrics_for_all_schedulers": false, "load_snapshot_publish_interval": 15, "tokenizer_metrics_custom_labels_header": "x-custom-labels", "tokenizer_metrics_allowed_custom_labels": null, "extra_metric_labels": null, "bucket_time_to_first_token": null, "bucket_inter_token_latency": null, "bucket_e2e_request_latency": null, "prompt_tokens_buckets": null, "generation_tokens_buckets": null, "gc_warning_threshold_secs": 0.0, "decode_log_interval": 40, "enable_request_time_stats_logging": false, "kv_events_config": null, "enable_forward_pass_metrics": false, "forward_pass_metrics_worker_id": "", "forward_pass_metrics_ipc_name": null, "enable_trace": false, "trace_modules": "request", "otlp_traces_endpoint": "localhost:4317", "export_metrics_to_file": false, "export_metrics_to_file_dir": null, "stat_loggers": null, "constrained_json_whitespace_pattern": null, "constrained_json_disable_any_whitespace": false, "attention_backend": "triton", "decode_attention_backend": null, "prefill_attention_backend": null, "sampling_backend": "flashinfer", "grammar_backend": "xgrammar", "radix_cache_backend": null, "mm_attention_backend": null, "fp8_gemm_runner_backend": "auto", "fp4_gemm_runner_backend": "auto", "bf16_gemm_backend": "auto", "dsa_prefill_backend": null, "dsa_decode_backend": null, "dsa_paged_mqa_logits_backend": "auto", "dsa_topk_backend": "sgl-kernel", "disable_flashinfer_autotune": false, "flashinfer_autotune_skip_ops": null, "mamba_backend": "triton", "cuda_graph_config": {"decode": {"backend": "full", "max_bs": 16, "bs": [1, 2, 4, 8, 12, 16], "tc_compiler": "eager", "full_prefill_max_req": null}, "prefill": {"backend": "disabled", "max_bs": 2048, "bs": [4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], "tc_compiler": "eager", "full_prefill_max_req": null}}, "cuda_graph_backend_decode": null, "cuda_graph_backend_prefill": null, "cuda_graph_max_bs_decode": 16, "cuda_graph_max_bs_prefill": null, "cuda_graph_bs_decode": null, "cuda_graph_bs_prefill": null, "cuda_graph_tc_compiler": null, "disable_prefill_cuda_graph": false, "disable_decode_cuda_graph": false, "disable_cuda_graph": false, "disable_cuda_graph_padding": false, "enable_profile_cuda_graph": false, "enable_cudagraph_gc": false, "debug_cuda_graph": false, "enable_layerwise_nvtx_marker": false, "enable_nccl_nvls": false, "enable_symm_mem": false, "triton_attention_reduce_in_fp32": false, "triton_attention_num_kv_splits": 8, "triton_attention_split_tile_size": null, "flashinfer_mla_disable_ragged": false, "enable_fused_qk_norm_rope": false, "enable_precise_embedding_interpolation": false, "enable_fused_moe_sum_all_reduce": false, "enable_deepseek_v4_fp4_indexer": false, "disable_custom_all_reduce": false, "enable_mscclpp": false, "enable_torch_symm_mem": false, "enable_scattered_sconv": false, "pre_warm_nccl": false, "enable_quant_communications": false, "enable_flashinfer_allreduce_fusion": false, "enforce_disable_flashinfer_allreduce_fusion": false, "flashinfer_allreduce_fusion_backend": null, "enable_aiter_allreduce_fusion": false, "enable_torch_compile": false, "enable_torch_compile_debug_mode": false, "torch_compile_max_bs": 32, "torchao_config": "", "speculative_algorithm": null, "speculative_draft_model_path": null, "speculative_draft_model_revision": null, "speculative_draft_load_format": null, "speculative_num_steps": null, "speculative_eagle_topk": null, "speculative_num_draft_tokens": null, "speculative_dflash_block_size": null, "speculative_dspark_block_size": null, "speculative_dspark_sps_table_path": null, "speculative_dspark_confidence_sts_path": null, "speculative_dspark_align_verify_tokens_to_graph_tier": false, "speculative_accept_threshold_single": 1.0, "speculative_accept_threshold_acc": 1.0, "speculative_use_rejection_sampling": false, "speculative_token_map": null, "speculative_attention_mode": "prefill", "speculative_draft_attention_backend": null, "speculative_draft_window_size": null, "speculative_moe_runner_backend": "marlin", "speculative_moe_a2a_backend": null, "speculative_draft_model_quantization": null, "speculative_skip_dp_mlp_sync": false, "enable_multi_layer_eagle": false, "speculative_adaptive": false, "speculative_adaptive_config": null, "decoupled_spec_bind_endpoint": null, "decoupled_spec_connect_endpoints": null, "decoupled_spec_rank": null, "decoupled_spec_role": "null", "spec_trace_dir": null, "speculative_ngram_min_bfs_breadth": 1, "speculative_ngram_max_bfs_breadth": 10, "speculative_ngram_match_type": "BFS", "speculative_ngram_max_trie_depth": 18, "speculative_ngram_capacity": 10000000, "speculative_ngram_external_corpus_path": null, "speculative_ngram_external_sam_budget": 0, "speculative_ngram_external_corpus_max_tokens": 10000000, "ep_size": 32, "moe_a2a_backend": "none", "moe_runner_backend": "marlin", "flashinfer_mxfp4_moe_precision": "default", "deepep_mode": "auto", "fuseep_mode": 2, "deepep_dispatcher_output_dtype": "auto", "ep_num_redundant_experts": 0, "ep_dispatch_algorithm": null, "init_expert_location": "trivial", "enable_eplb": false, "eplb_algorithm": "auto", "eplb_rebalance_num_iterations": 1000, "eplb_rebalance_layers_per_chunk": null, "eplb_min_rebalancing_utilization_threshold": 1.0, "expert_distribution_recorder_mode": null, "expert_distribution_recorder_buffer_size": 1000, "enable_expert_distribution_metrics": false, "deepep_config": null, "moe_dense_tp_size": null, "elastic_ep_backend": null, "enable_elastic_expert_backup": false, "mooncake_ib_device": null, "enable_waterfill": false, "ep_join_mode": null, "ep_join_rank_offset": 0, "elastic_ep_initial_size": null, "max_ep_size": null, "elastic_ep_scale_timeout": 600, "elastic_ep_rejoin": false, "disable_flashinfer_cutlass_moe_fp4_allgather": false, "disable_shared_experts_fusion": false, "enforce_shared_experts_fusion": false, "max_mamba_cache_size": 181, "mamba_ssm_dtype": null, "mamba_max_states_per_path": -1, "enable_mamba_cache_stochastic_rounding": false, "mamba_cache_philox_rounds": 0, "mamba_full_memory_ratio": 0.21, "mamba_radix_cache_strategy": "extra_buffer_lazy", "uses_mamba_radix_cache": false, "mamba_track_interval": 256, "enable_int8_mamba_checkpoint": false, "int8_mamba_ckpt_size": null, "linear_attn_backend": "triton", "linear_attn_decode_backend": null, "linear_attn_prefill_backend": null, "linear_attn_verify_backend": null, "enable_linear_replayssm": false, "linear_replayssm_cache_len": 16, "enable_linear_replayssm_spec": false, "enable_hierarchical_cache": false, "hicache_ratio": 2.0, "hicache_size": 0, "hicache_write_policy": "write_through", "hicache_io_backend": "kernel", "hicache_mem_layout": "page_first", "hicache_storage_backend": null, "hicache_storage_prefetch_policy": "timeout", "hicache_storage_backend_extra_config": null, "enable_hisparse": false, "hisparse_config": null, "enable_broadcast_mm_inputs_process": false, "enable_prefix_mm_cache": false, "mm_enable_dp_encoder": false, "mm_process_config": {}, "mm_processor_worker_num": 0, "mm_io_worker_num": 0, "limit_mm_data_per_request": null, "enable_mm_global_cache": false, "disable_fast_image_processor": false, "mm_feature_transport": "cpu", "keep_mm_feature_on_device": false, "enable_lora": null, "enable_lora_overlap_loading": null, "max_lora_rank": null, "lora_target_modules": null, "lora_paths": null, "max_loaded_loras": null, "max_loras_per_batch": 8, "lora_eviction_policy": "lru", "lora_backend": "csgmv", "max_lora_chunk_size": 16, "experts_shared_outer_loras": null, "lora_use_virtual_experts": false, "lora_strict_loading": false, "lora_drain_wait_threshold": 0.0, "enable_two_batch_overlap": false, "enable_single_batch_overlap": false, "tbo_token_distribution_threshold": 0.48, "cpu_offload_gb": 0, "offload_group_size": -1, "offload_num_in_group": 1, "offload_prefetch_step": 1, "offload_mode": "cpu", "enable_lmcache": false, "lmcache_config_file": null, "enable_flexkv": false, "flexkv_config_file": null, "kt_weight_path": null, "kt_method": "AMXINT4", "kt_cpuinfer": null, "kt_threadpool_count": 2, "kt_num_gpu_experts": null, "kt_max_deferred_experts_per_token": null, "dllm_algorithm": null, "dllm_algorithm_config": null, "dllm_fdfo": true, "disaggregation_mode": "decode", "disaggregation_transfer_backend": "mooncake", "disaggregation_bootstrap_port": 28800, "disaggregation_ib_device": "mlx5_0,mlx5_1,mlx5_2,mlx5_3", "disaggregation_decode_enable_radix_cache": false, "disaggregation_decode_enable_offload_kvcache": false, "num_reserved_decode_tokens": 512, "disaggregation_decode_extra_slots": 0, "disaggregation_decode_polling_interval": 1, "optimistic_prefill_attempts": 0, "encoder_only": false, "language_only": false, "encoder_transfer_backend": "zmq_to_tokenizer", "encoder_urls": [], "encoder_bootstrap_port": 8997, "encoder_register_urls": [], "enable_adaptive_dispatch_to_encoder": false, "enable_pdmux": false, "pdmux_config_path": null, "sm_group_num": 8, "custom_weight_loader": [], "weight_loader_disable_mmap": false, "weight_loader_prefetch_checkpoints": false, "weight_loader_prefetch_num_threads": 4, "weight_loader_drop_cache_after_load": false, "remote_instance_weight_loader_seed_instance_ip": null, "remote_instance_weight_loader_seed_instance_service_port": null, "remote_instance_weight_loader_send_weights_group_ports": null, "remote_instance_weight_loader_backend": "nccl", "remote_instance_weight_loader_start_seed_via_transfer_engine": false, "engine_info_bootstrap_port": 6789, "modelexpress_config": null, "download_dir": null, "model_checksum": null, "delete_ckpt_after_loading": false, "decrypted_config_file": null, "decrypted_draft_config_file": null, "checkpoint_engine_wait_weights_before_ready": false, "enable_prefill_delayer": false, "prefill_delayer_max_delay_passes": 30, "prefill_delayer_token_usage_low_watermark": null, "prefill_delayer_forward_passes_buckets": null, "prefill_delayer_wait_seconds_buckets": null, "prefill_delayer_queue_min_ratio": null, "prefill_delayer_max_delay_ms": null, "min_free_slots_delay": null, "enable_deterministic_inference": false, "rl_on_policy_target": null, "kv_canary": "none", "kv_canary_real_data": "none", "kv_canary_sweep_interval": 0, "enable_dynamic_batch_tokenizer": false, "dynamic_batch_tokenizer_batch_size": 32, "dynamic_batch_tokenizer_batch_timeout": 0.002, "enable_tokenizer_batch_encode": false, "disable_tokenizer_batch_decode": false, "debug_tensor_dump_output_folder": null, "debug_tensor_dump_layers": null, "debug_tensor_dump_input_file": null, "enable_memory_saver": false, "enable_weights_cpu_backup": false, "enable_draft_weights_cpu_backup": false, "enable_custom_logit_processor": false, "enable_return_hidden_states": false, "enable_return_routed_experts": false, "enable_return_indexer_topk": false, "disable_outlines_disk_cache": false, "enable_mis": false, "weight_cache_mode": "off", "weight_cache_socket": null, "weight_cache_timeout": 1800, "forward_hooks": null, "msprobe_dump_config": null, "_resolved_overrides": [["_sampling_backend_default", {"sampling_backend": "flashinfer"}], ["_attention_backend_default", {"attention_backend": "triton"}], ["_data_parallelism_defaults", {"enable_dp_attention": false, "enable_dp_lm_head": false}], ["_speculative_moe_runner_default", {"speculative_moe_runner_backend": "marlin"}]], "grpc_worker_threads": 4, "_quantization_explicitly_unset": false, "_cuda_graph_config_locked": [["decode", "max_bs"]], "_declarations_materialized": true, "_runtime_mutations": [["model_runner.chunked_prefix_cache_gate", {"disable_chunked_prefix_cache": true}], ["mamba_pool.memory_budget", {"max_mamba_cache_size": 181}]], "_in_override": false, "_mx_config_cache": {}, "max_speculative_num_draft_tokens": null, "last_gen_throughput": 0.09006957280512767, "memory_usage": {"weight": 58.57, "kvcache": 11.34, "token_capacity": 440256, "graph": 0.6}, "effective_max_running_requests_per_dp": 181}], "version": "0.5.16", "kv_events": null}]}, "duration": 2.0154006520751864, "completed": 1, "total_input_tokens": 1024, "total_input_text_tokens": 1024, "total_input_vision_tokens": 0, "total_output_tokens": 16, "total_output_tokens_retokenized": 16, "request_throughput": 0.49617925794076506, "input_throughput": 508.0875601313434, "output_throughput": 7.938868127052241, "total_throughput": 516.0264282583956, "mean_e2e_latency_ms": 1955.6031040847301, "median_e2e_latency_ms": 1955.6031040847301, "std_e2e_latency_ms": 0.0, "p90_e2e_latency_ms": 1955.6031040847301, "p95_e2e_latency_ms": 1955.6031040847301, "p99_e2e_latency_ms": 1955.6031040847301, "mean_ttft_ms": 1955.715852091089, "median_ttft_ms": 1955.715852091089, "std_ttft_ms": 0.0, "p90_ttft_ms": 1955.715852091089, "p95_ttft_ms": 1955.715852091089, "p99_ttft_ms": 1955.715852091089, "mean_tpot_ms": -0.0075165337572495146, "median_tpot_ms": -0.0075165337572495146, "std_tpot_ms": 0.0, "p90_tpot_ms": -0.0075165337572495146, "p95_tpot_ms": -0.0075165337572495146, "p99_tpot_ms": -0.0075165337572495146, "mean_itl_ms": 0.0, "median_itl_ms": 0.0, "std_itl_ms": 0.0, "p90_itl_ms": 0.0, "p95_itl_ms": 0.0, "p99_itl_ms": 0.0, "concurrency": 0.9703296970114181, "accept_length": null, "max_output_tokens_per_s": 1.0, "max_concurrent_requests": 1, "input_lens": [1024], "output_lens": [16], "ttfts": [1.955715852091089], "itls": [[]], "generated_texts": ["\ufffd-inspired\ufffd-inspired\ufffd-inspired\ufffd-inspired"], "errors": [""]}
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/bench/smoke_1k_16_c1.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/bench/smoke_1k_16_c1.log
new file mode 100644
index 0000000..59c1736
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/bench/smoke_1k_16_c1.log
@@ -0,0 +1,71 @@
+/usr/local/lib/python3.12/dist-packages/torchao/quantization/quant_api.py:1731: SyntaxWarning: invalid escape sequence '\.'
+  """Configuration class for applying different quantization configs to modules or parameters based on their fully qualified names (FQNs).
+/sgl-workspace/sglang/python/sglang/kernels/ops/attention/fla/utils.py:223: UserWarning: Triton is not supported on current platform, roll back to CPU.
+  warnings.warn(
+/sgl-workspace/sglang/python/sglang/srt/layers/quantization/awq/awq.py:52: UserWarning: Only CUDA, HIP and XPU support AWQ currently.
+  warnings.warn(f"Only CUDA, HIP and XPU support AWQ currently.")
+/sgl-workspace/sglang/python/sglang/srt/layers/quantization/gguf.py:69: UserWarning: Only CUDA, MUSA and NPU support GGUF quantization currently.
+  warnings.warn(f"Only CUDA, MUSA and NPU support GGUF quantization currently.")
+benchmark_args=Namespace(backend='sglang', base_url=None, host='174.1.60.5', port=31000, ready_check_timeout_sec=60, dataset_name='random-ids', dataset_path='', dataset_offset=0, agentic_max_turns=None, speed_bench_category=None, speed_bench_output_len=512, model=None, served_model_name=None, tokenizer='/data/hf_models/Kimi-K3', num_prompts=1, sharegpt_output_len=None, sharegpt_context_len=None, random_input_len=1024, random_output_len=16, random_range_ratio=1.0, image_count=1, image_resolution='1080p', random_image_count=False, image_format='jpeg', image_content='random', request_rate=10000.0, use_trace_timestamps=False, max_concurrency=1, output_file='/data/hzy/sskj/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/bench/smoke_1k_16_c1.jsonl', output_details=True, print_requests=False, disable_tqdm=True, disable_stream=True, return_logprob=False, top_logprobs_num=0, token_ids_logprob=None, logprob_start_len=-1, return_routed_experts=False, cache_report=False, seed=42, disable_ignore_eos=False, temperature=0.0, top_p=1.0, extra_request_body=None, apply_chat_template=False, profile=False, plot_throughput=False, profile_activities=['CPU', 'GPU'], profile_start_step=None, profile_steps=None, profile_num_steps=None, profile_by_stage=False, profile_stages=None, profile_output_dir=None, profile_prefix=None, lora_name=None, lora_request_distribution='uniform', lora_zipf_alpha=1.5, prompt_suffix='', pd_separated=False, profile_prefill_url=None, profile_decode_url=None, flush_cache=False, warmup_requests=0, tokenize_prompt=False, gsp_num_groups=64, gsp_prompts_per_group=16, gsp_system_prompt_len=2048, gsp_question_len=128, gsp_output_len=256, gsp_range_ratio=1.0, gsp_fast_prepare=False, gsp_send_routing_key=False, gsp_num_turns=1, gsp_ordered=False, gsp_group_distribution='uniform', gsp_zipf_alpha=None, mooncake_slowdown_factor=1.0, mooncake_num_rounds=1, mooncake_workload='conversation', fake_prefill=False, tag=None, header=None)
+Waiting up to 60s for http://174.1.60.5:31000/v1/models to become ready...
+Server ready in 0.0s.
+'[Errno 101] Network is unreachable' thrown while requesting HEAD https://huggingface.co/kimi-k3/resolve/main/config.json
+Retrying in 1s [Retry 1/5].
+'[Errno 101] Network is unreachable' thrown while requesting HEAD https://huggingface.co/kimi-k3/resolve/main/config.json
+Retrying in 1s [Retry 1/5].
+Fail to load tokenizer config with error=Can't load the configuration of 'kimi-k3'. If you were trying to load it from 'https://huggingface.co/models', make sure you don't have a local directory with the same name. Otherwise, make sure 'kimi-k3' is the correct path to a directory containing a config.json file
+
+WARNING It is recommended to use the `Chat` or `Instruct` model for benchmarking.
+Because when the tokenizer counts the output tokens, if there is gibberish, it might count incorrectly.
+
+Namespace(backend='sglang', base_url=None, host='174.1.60.5', port=31000, ready_check_timeout_sec=60, dataset_name='random-ids', dataset_path='', dataset_offset=0, agentic_max_turns=None, speed_bench_category=None, speed_bench_output_len=512, model='kimi-k3', served_model_name=None, tokenizer='/data/hf_models/Kimi-K3', num_prompts=1, sharegpt_output_len=None, sharegpt_context_len=None, random_input_len=1024, random_output_len=16, random_range_ratio=1.0, image_count=1, image_resolution='1080p', random_image_count=False, image_format='jpeg', image_content='random', request_rate=10000.0, use_trace_timestamps=False, max_concurrency=1, output_file='/data/hzy/sskj/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/bench/smoke_1k_16_c1.jsonl', output_details=True, print_requests=False, disable_tqdm=True, disable_stream=True, return_logprob=False, top_logprobs_num=0, token_ids_logprob=None, logprob_start_len=-1, return_routed_experts=False, cache_report=False, seed=42, disable_ignore_eos=False, temperature=0.0, top_p=1.0, extra_request_body=None, apply_chat_template=False, profile=False, plot_throughput=False, profile_activities=['CPU', 'GPU'], profile_start_step=None, profile_steps=None, profile_num_steps=None, profile_by_stage=False, profile_stages=None, profile_output_dir=None, profile_prefix=None, lora_name=None, lora_request_distribution='uniform', lora_zipf_alpha=1.5, prompt_suffix='', pd_separated=False, profile_prefill_url=None, profile_decode_url=None, flush_cache=False, warmup_requests=0, tokenize_prompt=False, gsp_num_groups=64, gsp_prompts_per_group=16, gsp_system_prompt_len=2048, gsp_question_len=128, gsp_output_len=256, gsp_range_ratio=1.0, gsp_fast_prepare=False, gsp_send_routing_key=False, gsp_num_turns=1, gsp_ordered=False, gsp_group_distribution='uniform', gsp_zipf_alpha=None, mooncake_slowdown_factor=1.0, mooncake_num_rounds=1, mooncake_workload='conversation', fake_prefill=False, tag=None, header=None)
+
+#Input tokens: 1024
+#Output tokens: 16
+Starting warmup with 0 sequences...
+Warmup completed with 0 sequences. Starting main benchmark run...
+Calling super().encode with {'add_special_tokens': False}
+
+============ Serving Benchmark Result ============
+Backend:                                 sglang    
+Traffic request rate:                    10000.0   
+Max request concurrency:                 1         
+Successful requests:                     1         
+Benchmark duration (s):                  2.02      
+Total input tokens:                      1024      
+Total input text tokens:                 1024      
+Total generated tokens:                  16        
+Total generated tokens (retokenized):    16        
+Request throughput (req/s):              0.50      
+Input token throughput (tok/s):          508.09    
+Output token throughput (tok/s):         7.94      
+Peak output token throughput (tok/s):    1.00      
+Peak concurrent requests:                1         
+Total token throughput (tok/s):          516.03    
+Concurrency:                             0.97      
+----------------End-to-End Latency----------------
+Mean E2E Latency (ms):                   1955.60   
+Median E2E Latency (ms):                 1955.60   
+P90 E2E Latency (ms):                    1955.60   
+P95 E2E Latency (ms):                    1955.60   
+P99 E2E Latency (ms):                    1955.60   
+---------------Time to First Token----------------
+Mean TTFT (ms):                          1955.72   
+Median TTFT (ms):                        1955.72   
+P90 TTFT (ms):                           1955.72   
+P95 TTFT (ms):                           1955.72   
+P99 TTFT (ms):                           1955.72   
+-----Time per Output Token (excl. 1st token)------
+Mean TPOT (ms):                          -0.01     
+Median TPOT (ms):                        -0.01     
+P90 TPOT (ms):                           -0.01     
+P95 TPOT (ms):                           -0.01     
+P99 TPOT (ms):                           -0.01     
+---------------Inter-Token Latency----------------
+Mean ITL (ms):                           0.00      
+Median ITL (ms):                         0.00      
+P90 ITL (ms):                            0.00      
+P95 ITL (ms):                            0.00      
+P99 ITL (ms):                            0.00      
+Max ITL (ms):                            0.00      
+==================================================
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/bench/smoke_64_8_c1.cmd.txt b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/bench/smoke_64_8_c1.cmd.txt
new file mode 100644
index 0000000..ced444a
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/bench/smoke_64_8_c1.cmd.txt
@@ -0,0 +1 @@
+docker run --rm --network host -v /data/hf_models/Kimi-K3:/data/hf_models/Kimi-K3:ro -v /data/hzy/sskj/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310:/data/hzy/sskj/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310 --entrypoint python3 local/sglang:kimi-k3-sm120-flashinfer-mxfp4-pd1 -m sglang.benchmark.serving --backend sglang --host 174.1.60.5 --port 31000 --tokenizer /data/hf_models/Kimi-K3 --dataset-name random-ids --random-input-len 64 --random-output-len 8 --random-range-ratio 1.0 --num-prompts 1 --max-concurrency 1 --request-rate 10000 --warmup-requests 0 --output-file /data/hzy/sskj/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/bench/smoke_64_8_c1.jsonl --output-details --disable-tqdm 
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/bench/smoke_64_8_c1.jsonl b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/bench/smoke_64_8_c1.jsonl
new file mode 100644
index 0000000..3f9ef01
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/bench/smoke_64_8_c1.jsonl
@@ -0,0 +1 @@
+{"tag": null, "backend": "sglang", "dataset_name": "random-ids", "request_rate": 10000.0, "max_concurrency": 1, "sharegpt_output_len": null, "random_input_len": 64, "random_output_len": 8, "random_range_ratio": 1.0, "server_info": {"internal_states": [{"model_path": "/data/hf_models/Kimi-K3", "tokenizer_path": "/data/hf_models/Kimi-K3", "tokenizer_mode": "auto", "tokenizer_backend": "huggingface", "tokenizer_worker_num": 1, "detokenizer_worker_num": 1, "skip_tokenizer_init": false, "load_format": "auto", "model_loader_extra_config": "{}", "trust_remote_code": true, "context_length": null, "is_embedding": false, "enable_multimodal": null, "revision": null, "model_impl": "auto", "model_config_parser": "auto", "json_model_override_args": "{}", "dtype": "auto", "quantization": null, "quantization_param_path": null, "kv_cache_dtype": "auto", "enable_fp32_lm_head": false, "modelopt_quant": null, "modelopt_checkpoint_restore_path": null, "modelopt_checkpoint_save_path": null, "modelopt_export_path": null, "quantize_and_serve": false, "rl_quant_profile": null, "enable_tf32_matmul": false, "mem_fraction_static": 0.88, "max_running_requests": null, "max_queued_requests": null, "max_total_tokens": null, "chunked_prefill_size": 8192, "enable_dynamic_chunking": false, "max_prefill_tokens": 16384, "prefill_max_requests": null, "schedule_policy": "fcfs", "enable_priority_scheduling": false, "disable_priority_preemption": false, "default_priority_value": null, "abort_on_priority_when_disabled": false, "schedule_low_priority_values_first": false, "priority_scheduling_preemption_threshold": 10, "retraction_policy": "length", "schedule_conservativeness": 1.0, "page_size": 64, "swa_full_tokens_ratio": 0.8, "disable_hybrid_swa_memory": false, "radix_eviction_policy": "lru", "prefill_only_disable_kv_cache": false, "disable_radix_cache": true, "enable_page_major_kv_layout": false, "enable_unified_memory": false, "disable_chunked_prefix_cache": true, "disable_overlap_schedule": false, "num_continuous_decode_steps": 1, "scheduler_recv_interval": 1, "enable_mixed_chunk": false, "nccl_port": null, "dist_timeout": 3600, "dist_init_addr": "174.1.60.5:20000", "nnodes": 4, "node_rank": 0, "tp_size": 32, "dcp_size": 1, "pp_size": 1, "pp_max_micro_batch_size": 181, "pp_async_batch_depth": 0, "dp_size": 1, "load_balance_method": "round_robin", "attn_cp_size": 1, "moe_dp_size": 1, "dcp_comm_backend": "ag_rs", "dcp_replicate_q_proj": null, "dwdp_size": 1, "enable_prefill_cp": false, "cp_strategy": null, "enable_dsa_cache_layer_split": false, "enable_dsa_prefill_context_parallel": false, "dsa_prefill_cp_mode": "round-robin-split", "enable_prefill_context_parallel": false, "prefill_cp_mode": "in-seq-split", "enable_cp_decode_attn_tp": false, "enable_dp_attention": false, "enable_dp_attention_local_control_broadcast": false, "enable_dp_lm_head": false, "enable_attn_tp_input_scattered": false, "disable_attn_tp_gather": false, "enable_p2p_check": false, "device": "cuda", "base_gpu_id": 0, "gpu_id_step": 1, "random_seed": 369235288, "watchdog_timeout": 300, "soft_watchdog_timeout": null, "sleep_on_idle": false, "use_ray": false, "numa_node": null, "gc_threshold": null, "host": "0.0.0.0", "port": 30000, "fastapi_root_path": "", "smg_grpc_mode": false, "grpc_mode": false, "grpc_port": null, "sidecar": null, "sidecar_args": null, "skip_server_warmup": false, "warmups": null, "enable_http2": false, "ssl_keyfile": null, "ssl_certfile": null, "ssl_ca_certs": null, "ssl_keyfile_password": null, "enable_ssl_refresh": false, "api_key": null, "admin_api_key": null, "served_model_name": "kimi-k3", "weight_version": "default", "chat_template": null, "hf_chat_template_name": null, "completion_template": null, "file_storage_path": "sglang_storage", "enable_cache_report": false, "reasoning_parser": null, "default_chat_template_kwargs": null, "strip_thinking_cache": false, "enable_strict_thinking": false, "tool_call_parser": null, "tool_server": null, "sampling_defaults": "model", "asr_max_buffer_seconds": 60, "asr_max_concurrent_sessions": 32, "preferred_sampling_params": null, "allow_auto_truncate": false, "stream_interval": 1, "batch_notify_size": 16, "stream_response_default_include_usage": false, "incremental_streaming_output": false, "enable_streaming_session": false, "enable_session_radix_cache": false, "log_level": "info", "log_level_http": null, "log_requests": false, "log_requests_level": 2, "log_requests_format": "text", "log_requests_target": null, "uvicorn_access_log_exclude_prefixes": [], "crash_dump_folder": null, "show_time_cost": false, "enable_metrics": false, "smg_http_sidecar_port": null, "enable_mfu_metrics": false, "enable_metrics_for_all_schedulers": false, "load_snapshot_publish_interval": 15, "tokenizer_metrics_custom_labels_header": "x-custom-labels", "tokenizer_metrics_allowed_custom_labels": null, "extra_metric_labels": null, "bucket_time_to_first_token": null, "bucket_inter_token_latency": null, "bucket_e2e_request_latency": null, "prompt_tokens_buckets": null, "generation_tokens_buckets": null, "gc_warning_threshold_secs": 0.0, "decode_log_interval": 40, "enable_request_time_stats_logging": false, "kv_events_config": null, "enable_forward_pass_metrics": false, "forward_pass_metrics_worker_id": "", "forward_pass_metrics_ipc_name": null, "enable_trace": false, "trace_modules": "request", "otlp_traces_endpoint": "localhost:4317", "export_metrics_to_file": false, "export_metrics_to_file_dir": null, "stat_loggers": null, "constrained_json_whitespace_pattern": null, "constrained_json_disable_any_whitespace": false, "attention_backend": "triton", "decode_attention_backend": null, "prefill_attention_backend": null, "sampling_backend": "flashinfer", "grammar_backend": "xgrammar", "radix_cache_backend": null, "mm_attention_backend": null, "fp8_gemm_runner_backend": "auto", "fp4_gemm_runner_backend": "auto", "bf16_gemm_backend": "auto", "dsa_prefill_backend": null, "dsa_decode_backend": null, "dsa_paged_mqa_logits_backend": "auto", "dsa_topk_backend": "sgl-kernel", "disable_flashinfer_autotune": false, "flashinfer_autotune_skip_ops": null, "mamba_backend": "triton", "cuda_graph_config": {"decode": {"backend": "full", "max_bs": 16, "bs": [1, 2, 4, 8, 12, 16], "tc_compiler": "eager", "full_prefill_max_req": null}, "prefill": {"backend": "disabled", "max_bs": 2048, "bs": [4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], "tc_compiler": "eager", "full_prefill_max_req": null}}, "cuda_graph_backend_decode": null, "cuda_graph_backend_prefill": null, "cuda_graph_max_bs_decode": 16, "cuda_graph_max_bs_prefill": null, "cuda_graph_bs_decode": null, "cuda_graph_bs_prefill": null, "cuda_graph_tc_compiler": null, "disable_prefill_cuda_graph": false, "disable_decode_cuda_graph": false, "disable_cuda_graph": false, "disable_cuda_graph_padding": false, "enable_profile_cuda_graph": false, "enable_cudagraph_gc": false, "debug_cuda_graph": false, "enable_layerwise_nvtx_marker": false, "enable_nccl_nvls": false, "enable_symm_mem": false, "triton_attention_reduce_in_fp32": false, "triton_attention_num_kv_splits": 8, "triton_attention_split_tile_size": null, "flashinfer_mla_disable_ragged": false, "enable_fused_qk_norm_rope": false, "enable_precise_embedding_interpolation": false, "enable_fused_moe_sum_all_reduce": false, "enable_deepseek_v4_fp4_indexer": false, "disable_custom_all_reduce": false, "enable_mscclpp": false, "enable_torch_symm_mem": false, "enable_scattered_sconv": false, "pre_warm_nccl": false, "enable_quant_communications": false, "enable_flashinfer_allreduce_fusion": false, "enforce_disable_flashinfer_allreduce_fusion": false, "flashinfer_allreduce_fusion_backend": null, "enable_aiter_allreduce_fusion": false, "enable_torch_compile": false, "enable_torch_compile_debug_mode": false, "torch_compile_max_bs": 32, "torchao_config": "", "speculative_algorithm": null, "speculative_draft_model_path": null, "speculative_draft_model_revision": null, "speculative_draft_load_format": null, "speculative_num_steps": null, "speculative_eagle_topk": null, "speculative_num_draft_tokens": null, "speculative_dflash_block_size": null, "speculative_dspark_block_size": null, "speculative_dspark_sps_table_path": null, "speculative_dspark_confidence_sts_path": null, "speculative_dspark_align_verify_tokens_to_graph_tier": false, "speculative_accept_threshold_single": 1.0, "speculative_accept_threshold_acc": 1.0, "speculative_use_rejection_sampling": false, "speculative_token_map": null, "speculative_attention_mode": "prefill", "speculative_draft_attention_backend": null, "speculative_draft_window_size": null, "speculative_moe_runner_backend": "marlin", "speculative_moe_a2a_backend": null, "speculative_draft_model_quantization": null, "speculative_skip_dp_mlp_sync": false, "enable_multi_layer_eagle": false, "speculative_adaptive": false, "speculative_adaptive_config": null, "decoupled_spec_bind_endpoint": null, "decoupled_spec_connect_endpoints": null, "decoupled_spec_rank": null, "decoupled_spec_role": "null", "spec_trace_dir": null, "speculative_ngram_min_bfs_breadth": 1, "speculative_ngram_max_bfs_breadth": 10, "speculative_ngram_match_type": "BFS", "speculative_ngram_max_trie_depth": 18, "speculative_ngram_capacity": 10000000, "speculative_ngram_external_corpus_path": null, "speculative_ngram_external_sam_budget": 0, "speculative_ngram_external_corpus_max_tokens": 10000000, "ep_size": 32, "moe_a2a_backend": "none", "moe_runner_backend": "marlin", "flashinfer_mxfp4_moe_precision": "default", "deepep_mode": "auto", "fuseep_mode": 2, "deepep_dispatcher_output_dtype": "auto", "ep_num_redundant_experts": 0, "ep_dispatch_algorithm": null, "init_expert_location": "trivial", "enable_eplb": false, "eplb_algorithm": "auto", "eplb_rebalance_num_iterations": 1000, "eplb_rebalance_layers_per_chunk": null, "eplb_min_rebalancing_utilization_threshold": 1.0, "expert_distribution_recorder_mode": null, "expert_distribution_recorder_buffer_size": 1000, "enable_expert_distribution_metrics": false, "deepep_config": null, "moe_dense_tp_size": null, "elastic_ep_backend": null, "enable_elastic_expert_backup": false, "mooncake_ib_device": null, "enable_waterfill": false, "ep_join_mode": null, "ep_join_rank_offset": 0, "elastic_ep_initial_size": null, "max_ep_size": null, "elastic_ep_scale_timeout": 600, "elastic_ep_rejoin": false, "disable_flashinfer_cutlass_moe_fp4_allgather": false, "disable_shared_experts_fusion": false, "enforce_shared_experts_fusion": false, "max_mamba_cache_size": 181, "mamba_ssm_dtype": null, "mamba_max_states_per_path": -1, "enable_mamba_cache_stochastic_rounding": false, "mamba_cache_philox_rounds": 0, "mamba_full_memory_ratio": 0.21, "mamba_radix_cache_strategy": "extra_buffer_lazy", "uses_mamba_radix_cache": false, "mamba_track_interval": 256, "enable_int8_mamba_checkpoint": false, "int8_mamba_ckpt_size": null, "linear_attn_backend": "triton", "linear_attn_decode_backend": null, "linear_attn_prefill_backend": null, "linear_attn_verify_backend": null, "enable_linear_replayssm": false, "linear_replayssm_cache_len": 16, "enable_linear_replayssm_spec": false, "enable_hierarchical_cache": false, "hicache_ratio": 2.0, "hicache_size": 0, "hicache_write_policy": "write_through", "hicache_io_backend": "kernel", "hicache_mem_layout": "page_first", "hicache_storage_backend": null, "hicache_storage_prefetch_policy": "timeout", "hicache_storage_backend_extra_config": null, "enable_hisparse": false, "hisparse_config": null, "enable_broadcast_mm_inputs_process": false, "enable_prefix_mm_cache": false, "mm_enable_dp_encoder": false, "mm_process_config": {}, "mm_processor_worker_num": 0, "mm_io_worker_num": 0, "limit_mm_data_per_request": null, "enable_mm_global_cache": false, "disable_fast_image_processor": false, "mm_feature_transport": "cpu", "keep_mm_feature_on_device": false, "enable_lora": null, "enable_lora_overlap_loading": null, "max_lora_rank": null, "lora_target_modules": null, "lora_paths": null, "max_loaded_loras": null, "max_loras_per_batch": 8, "lora_eviction_policy": "lru", "lora_backend": "csgmv", "max_lora_chunk_size": 16, "experts_shared_outer_loras": null, "lora_use_virtual_experts": false, "lora_strict_loading": false, "lora_drain_wait_threshold": 0.0, "enable_two_batch_overlap": false, "enable_single_batch_overlap": false, "tbo_token_distribution_threshold": 0.48, "cpu_offload_gb": 0, "offload_group_size": -1, "offload_num_in_group": 1, "offload_prefetch_step": 1, "offload_mode": "cpu", "enable_lmcache": false, "lmcache_config_file": null, "enable_flexkv": false, "flexkv_config_file": null, "kt_weight_path": null, "kt_method": "AMXINT4", "kt_cpuinfer": null, "kt_threadpool_count": 2, "kt_num_gpu_experts": null, "kt_max_deferred_experts_per_token": null, "dllm_algorithm": null, "dllm_algorithm_config": null, "dllm_fdfo": true, "disaggregation_mode": "decode", "disaggregation_transfer_backend": "mooncake", "disaggregation_bootstrap_port": 28800, "disaggregation_ib_device": "mlx5_0,mlx5_1,mlx5_2,mlx5_3", "disaggregation_decode_enable_radix_cache": false, "disaggregation_decode_enable_offload_kvcache": false, "num_reserved_decode_tokens": 512, "disaggregation_decode_extra_slots": 0, "disaggregation_decode_polling_interval": 1, "optimistic_prefill_attempts": 0, "encoder_only": false, "language_only": false, "encoder_transfer_backend": "zmq_to_tokenizer", "encoder_urls": [], "encoder_bootstrap_port": 8997, "encoder_register_urls": [], "enable_adaptive_dispatch_to_encoder": false, "enable_pdmux": false, "pdmux_config_path": null, "sm_group_num": 8, "custom_weight_loader": [], "weight_loader_disable_mmap": false, "weight_loader_prefetch_checkpoints": false, "weight_loader_prefetch_num_threads": 4, "weight_loader_drop_cache_after_load": false, "remote_instance_weight_loader_seed_instance_ip": null, "remote_instance_weight_loader_seed_instance_service_port": null, "remote_instance_weight_loader_send_weights_group_ports": null, "remote_instance_weight_loader_backend": "nccl", "remote_instance_weight_loader_start_seed_via_transfer_engine": false, "engine_info_bootstrap_port": 6789, "modelexpress_config": null, "download_dir": null, "model_checksum": null, "delete_ckpt_after_loading": false, "decrypted_config_file": null, "decrypted_draft_config_file": null, "checkpoint_engine_wait_weights_before_ready": false, "enable_prefill_delayer": false, "prefill_delayer_max_delay_passes": 30, "prefill_delayer_token_usage_low_watermark": null, "prefill_delayer_forward_passes_buckets": null, "prefill_delayer_wait_seconds_buckets": null, "prefill_delayer_queue_min_ratio": null, "prefill_delayer_max_delay_ms": null, "min_free_slots_delay": null, "enable_deterministic_inference": false, "rl_on_policy_target": null, "kv_canary": "none", "kv_canary_real_data": "none", "kv_canary_sweep_interval": 0, "enable_dynamic_batch_tokenizer": false, "dynamic_batch_tokenizer_batch_size": 32, "dynamic_batch_tokenizer_batch_timeout": 0.002, "enable_tokenizer_batch_encode": false, "disable_tokenizer_batch_decode": false, "debug_tensor_dump_output_folder": null, "debug_tensor_dump_layers": null, "debug_tensor_dump_input_file": null, "enable_memory_saver": false, "enable_weights_cpu_backup": false, "enable_draft_weights_cpu_backup": false, "enable_custom_logit_processor": false, "enable_return_hidden_states": false, "enable_return_routed_experts": false, "enable_return_indexer_topk": false, "disable_outlines_disk_cache": false, "enable_mis": false, "weight_cache_mode": "off", "weight_cache_socket": null, "weight_cache_timeout": 1800, "forward_hooks": null, "msprobe_dump_config": null, "_resolved_overrides": [["_sampling_backend_default", {"sampling_backend": "flashinfer"}], ["_attention_backend_default", {"attention_backend": "triton"}], ["_data_parallelism_defaults", {"enable_dp_attention": false, "enable_dp_lm_head": false}], ["_speculative_moe_runner_default", {"speculative_moe_runner_backend": "marlin"}]], "grpc_worker_threads": 4, "_quantization_explicitly_unset": false, "_cuda_graph_config_locked": [["decode", "max_bs"]], "_declarations_materialized": true, "_runtime_mutations": [["model_runner.chunked_prefix_cache_gate", {"disable_chunked_prefix_cache": true}], ["mamba_pool.memory_budget", {"max_mamba_cache_size": 181}]], "_in_override": false, "_mx_config_cache": {}, "max_speculative_num_draft_tokens": null, "last_gen_throughput": 0.09006957280512767, "memory_usage": {"weight": 58.57, "kvcache": 11.34, "token_capacity": 440256, "graph": 0.6}, "effective_max_running_requests_per_dp": 181}], "prefill": [{"model_path": "/data/hf_models/Kimi-K3", "tokenizer_path": "/data/hf_models/Kimi-K3", "tokenizer_mode": "auto", "tokenizer_backend": "huggingface", "tokenizer_worker_num": 1, "detokenizer_worker_num": 1, "skip_tokenizer_init": false, "load_format": "auto", "model_loader_extra_config": "{}", "trust_remote_code": true, "context_length": null, "is_embedding": false, "enable_multimodal": null, "revision": null, "model_impl": "auto", "model_config_parser": "auto", "json_model_override_args": "{}", "dtype": "auto", "quantization": null, "quantization_param_path": null, "kv_cache_dtype": "auto", "enable_fp32_lm_head": false, "modelopt_quant": null, "modelopt_checkpoint_restore_path": null, "modelopt_checkpoint_save_path": null, "modelopt_export_path": null, "quantize_and_serve": false, "rl_quant_profile": null, "enable_tf32_matmul": false, "mem_fraction_static": 0.88, "max_running_requests": null, "max_queued_requests": null, "max_total_tokens": null, "chunked_prefill_size": 8192, "enable_dynamic_chunking": false, "max_prefill_tokens": 16384, "prefill_max_requests": null, "schedule_policy": "fcfs", "enable_priority_scheduling": false, "disable_priority_preemption": false, "default_priority_value": null, "abort_on_priority_when_disabled": false, "schedule_low_priority_values_first": false, "priority_scheduling_preemption_threshold": 10, "retraction_policy": "length", "schedule_conservativeness": 1.0, "page_size": 64, "swa_full_tokens_ratio": 0.8, "disable_hybrid_swa_memory": false, "radix_eviction_policy": "lru", "prefill_only_disable_kv_cache": false, "disable_radix_cache": true, "enable_page_major_kv_layout": false, "enable_unified_memory": false, "disable_chunked_prefix_cache": false, "disable_overlap_schedule": true, "num_continuous_decode_steps": 1, "scheduler_recv_interval": 1, "enable_mixed_chunk": false, "nccl_port": null, "dist_timeout": 3600, "dist_init_addr": "174.1.60.1:20000", "nnodes": 4, "node_rank": 0, "tp_size": 4, "dcp_size": 1, "pp_size": 8, "pp_max_micro_batch_size": null, "pp_async_batch_depth": 0, "dp_size": 1, "load_balance_method": "follow_bootstrap_room", "attn_cp_size": 1, "moe_dp_size": 1, "dcp_comm_backend": "ag_rs", "dcp_replicate_q_proj": null, "dwdp_size": 1, "enable_prefill_cp": false, "cp_strategy": null, "enable_dsa_cache_layer_split": false, "enable_dsa_prefill_context_parallel": false, "dsa_prefill_cp_mode": "round-robin-split", "enable_prefill_context_parallel": false, "prefill_cp_mode": "in-seq-split", "enable_cp_decode_attn_tp": false, "enable_dp_attention": false, "enable_dp_attention_local_control_broadcast": false, "enable_dp_lm_head": false, "enable_attn_tp_input_scattered": false, "disable_attn_tp_gather": false, "enable_p2p_check": false, "device": "cuda", "base_gpu_id": 0, "gpu_id_step": 1, "random_seed": 984607766, "watchdog_timeout": 300, "soft_watchdog_timeout": null, "sleep_on_idle": false, "use_ray": false, "custom_sigquit_handler": null, "numa_node": null, "gc_threshold": null, "host": "0.0.0.0", "port": 30000, "fastapi_root_path": "", "smg_grpc_mode": false, "grpc_mode": false, "grpc_port": null, "sidecar": null, "sidecar_args": null, "skip_server_warmup": false, "warmups": null, "enable_http2": false, "ssl_keyfile": null, "ssl_certfile": null, "ssl_ca_certs": null, "ssl_keyfile_password": null, "enable_ssl_refresh": false, "api_key": null, "admin_api_key": null, "served_model_name": "kimi-k3", "weight_version": "default", "chat_template": null, "hf_chat_template_name": null, "completion_template": null, "file_storage_path": "sglang_storage", "enable_cache_report": false, "reasoning_parser": null, "default_chat_template_kwargs": null, "strip_thinking_cache": false, "enable_strict_thinking": false, "tool_call_parser": null, "tool_server": null, "sampling_defaults": "model", "asr_max_buffer_seconds": 60, "asr_max_concurrent_sessions": 32, "preferred_sampling_params": null, "allow_auto_truncate": false, "stream_interval": 1, "batch_notify_size": 16, "stream_response_default_include_usage": false, "incremental_streaming_output": false, "enable_streaming_session": false, "enable_session_radix_cache": false, "log_level": "info", "log_level_http": null, "log_requests": false, "log_requests_level": 2, "log_requests_format": "text", "log_requests_target": null, "uvicorn_access_log_exclude_prefixes": [], "crash_dump_folder": null, "show_time_cost": false, "enable_metrics": false, "smg_http_sidecar_port": null, "enable_mfu_metrics": false, "enable_metrics_for_all_schedulers": false, "load_snapshot_publish_interval": 15, "tokenizer_metrics_custom_labels_header": "x-custom-labels", "tokenizer_metrics_allowed_custom_labels": null, "extra_metric_labels": null, "bucket_time_to_first_token": null, "bucket_inter_token_latency": null, "bucket_e2e_request_latency": null, "prompt_tokens_buckets": null, "generation_tokens_buckets": null, "gc_warning_threshold_secs": 0.0, "decode_log_interval": 40, "enable_request_time_stats_logging": false, "kv_events_config": null, "enable_forward_pass_metrics": false, "forward_pass_metrics_worker_id": "", "forward_pass_metrics_ipc_name": null, "enable_trace": false, "trace_modules": "request", "otlp_traces_endpoint": "localhost:4317", "export_metrics_to_file": false, "export_metrics_to_file_dir": null, "stat_loggers": null, "constrained_json_whitespace_pattern": null, "constrained_json_disable_any_whitespace": false, "attention_backend": "triton", "decode_attention_backend": null, "prefill_attention_backend": null, "sampling_backend": "flashinfer", "grammar_backend": "xgrammar", "radix_cache_backend": null, "mm_attention_backend": null, "fp8_gemm_runner_backend": "auto", "fp4_gemm_runner_backend": "auto", "bf16_gemm_backend": "auto", "dsa_prefill_backend": null, "dsa_decode_backend": null, "dsa_paged_mqa_logits_backend": "auto", "dsa_topk_backend": "sgl-kernel", "disable_flashinfer_autotune": false, "flashinfer_autotune_skip_ops": null, "mamba_backend": "triton", "cuda_graph_config": {"decode": {"backend": "disabled", "max_bs": 16, "bs": [1, 2, 4, 8, 12, 16], "tc_compiler": "eager", "full_prefill_max_req": null}, "prefill": {"backend": "disabled", "max_bs": 2048, "bs": [4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], "tc_compiler": "eager", "full_prefill_max_req": null}}, "cuda_graph_backend_decode": null, "cuda_graph_backend_prefill": null, "cuda_graph_max_bs_decode": 16, "cuda_graph_max_bs_prefill": null, "cuda_graph_bs_decode": null, "cuda_graph_bs_prefill": null, "cuda_graph_tc_compiler": null, "disable_prefill_cuda_graph": false, "disable_decode_cuda_graph": false, "disable_cuda_graph": false, "disable_cuda_graph_padding": false, "enable_profile_cuda_graph": false, "enable_cudagraph_gc": false, "debug_cuda_graph": false, "enable_layerwise_nvtx_marker": false, "enable_nccl_nvls": false, "enable_symm_mem": false, "triton_attention_reduce_in_fp32": false, "triton_attention_num_kv_splits": 8, "triton_attention_split_tile_size": null, "flashinfer_mla_disable_ragged": false, "enable_fused_qk_norm_rope": false, "enable_precise_embedding_interpolation": false, "enable_fused_moe_sum_all_reduce": false, "enable_deepseek_v4_fp4_indexer": false, "disable_custom_all_reduce": false, "enable_mscclpp": false, "enable_torch_symm_mem": false, "enable_scattered_sconv": false, "pre_warm_nccl": false, "enable_quant_communications": false, "enable_flashinfer_allreduce_fusion": false, "enforce_disable_flashinfer_allreduce_fusion": false, "flashinfer_allreduce_fusion_backend": null, "enable_aiter_allreduce_fusion": false, "enable_torch_compile": false, "enable_torch_compile_debug_mode": false, "torch_compile_max_bs": 32, "torchao_config": "", "speculative_algorithm": null, "speculative_draft_model_path": null, "speculative_draft_model_revision": null, "speculative_draft_load_format": null, "speculative_num_steps": null, "speculative_eagle_topk": null, "speculative_num_draft_tokens": null, "speculative_dflash_block_size": null, "speculative_dspark_block_size": null, "speculative_dspark_sps_table_path": null, "speculative_dspark_confidence_sts_path": null, "speculative_dspark_align_verify_tokens_to_graph_tier": false, "speculative_accept_threshold_single": 1.0, "speculative_accept_threshold_acc": 1.0, "speculative_use_rejection_sampling": false, "speculative_token_map": null, "speculative_attention_mode": "prefill", "speculative_draft_attention_backend": null, "speculative_draft_window_size": null, "speculative_moe_runner_backend": "flashinfer_mxfp4", "speculative_moe_a2a_backend": null, "speculative_draft_model_quantization": null, "speculative_skip_dp_mlp_sync": false, "enable_multi_layer_eagle": false, "speculative_adaptive": false, "speculative_adaptive_config": null, "decoupled_spec_bind_endpoint": null, "decoupled_spec_connect_endpoints": null, "decoupled_spec_rank": null, "decoupled_spec_role": "null", "spec_trace_dir": null, "speculative_ngram_min_bfs_breadth": 1, "speculative_ngram_max_bfs_breadth": 10, "speculative_ngram_match_type": "BFS", "speculative_ngram_max_trie_depth": 18, "speculative_ngram_capacity": 10000000, "speculative_ngram_external_corpus_path": null, "speculative_ngram_external_sam_budget": 0, "speculative_ngram_external_corpus_max_tokens": 10000000, "ep_size": 4, "moe_a2a_backend": "none", "moe_runner_backend": "flashinfer_mxfp4", "flashinfer_mxfp4_moe_precision": "default", "deepep_mode": "auto", "fuseep_mode": 2, "deepep_dispatcher_output_dtype": "auto", "ep_num_redundant_experts": 0, "ep_dispatch_algorithm": null, "init_expert_location": "trivial", "enable_eplb": false, "eplb_algorithm": "auto", "eplb_rebalance_num_iterations": 1000, "eplb_rebalance_layers_per_chunk": null, "eplb_min_rebalancing_utilization_threshold": 1.0, "expert_distribution_recorder_mode": null, "expert_distribution_recorder_buffer_size": 1000, "enable_expert_distribution_metrics": false, "deepep_config": null, "moe_dense_tp_size": null, "elastic_ep_backend": null, "enable_elastic_expert_backup": false, "mooncake_ib_device": null, "enable_waterfill": false, "ep_join_mode": null, "ep_join_rank_offset": 0, "elastic_ep_initial_size": null, "max_ep_size": null, "elastic_ep_scale_timeout": 600, "elastic_ep_rejoin": false, "disable_flashinfer_cutlass_moe_fp4_allgather": false, "disable_shared_experts_fusion": false, "enforce_shared_experts_fusion": false, "max_mamba_cache_size": null, "mamba_ssm_dtype": null, "mamba_max_states_per_path": -1, "enable_mamba_cache_stochastic_rounding": false, "mamba_cache_philox_rounds": 0, "mamba_full_memory_ratio": 0.36, "mamba_radix_cache_strategy": "extra_buffer_lazy", "uses_mamba_radix_cache": false, "mamba_track_interval": 256, "enable_int8_mamba_checkpoint": false, "int8_mamba_ckpt_size": null, "linear_attn_backend": "triton", "linear_attn_decode_backend": null, "linear_attn_prefill_backend": null, "linear_attn_verify_backend": null, "enable_linear_replayssm": false, "linear_replayssm_cache_len": 16, "enable_linear_replayssm_spec": false, "enable_hierarchical_cache": false, "hicache_ratio": 2.0, "hicache_size": 0, "hicache_write_policy": "write_through", "hicache_io_backend": "kernel", "hicache_mem_layout": "page_first", "hicache_storage_backend": null, "hicache_storage_prefetch_policy": "timeout", "hicache_storage_backend_extra_config": null, "enable_hisparse": false, "hisparse_config": null, "enable_broadcast_mm_inputs_process": false, "enable_prefix_mm_cache": false, "mm_enable_dp_encoder": false, "mm_process_config": {}, "mm_processor_worker_num": 0, "mm_io_worker_num": 0, "limit_mm_data_per_request": null, "enable_mm_global_cache": false, "disable_fast_image_processor": false, "mm_feature_transport": "cpu", "keep_mm_feature_on_device": false, "enable_lora": null, "enable_lora_overlap_loading": null, "max_lora_rank": null, "lora_target_modules": null, "lora_paths": null, "max_loaded_loras": null, "max_loras_per_batch": 8, "lora_eviction_policy": "lru", "lora_backend": "csgmv", "max_lora_chunk_size": 16, "experts_shared_outer_loras": null, "lora_use_virtual_experts": false, "lora_strict_loading": false, "lora_drain_wait_threshold": 0.0, "enable_two_batch_overlap": false, "enable_single_batch_overlap": false, "tbo_token_distribution_threshold": 0.48, "cpu_offload_gb": 0, "offload_group_size": -1, "offload_num_in_group": 1, "offload_prefetch_step": 1, "offload_mode": "cpu", "enable_lmcache": false, "lmcache_config_file": null, "enable_flexkv": false, "flexkv_config_file": null, "kt_weight_path": null, "kt_method": "AMXINT4", "kt_cpuinfer": null, "kt_threadpool_count": 2, "kt_num_gpu_experts": null, "kt_max_deferred_experts_per_token": null, "dllm_algorithm": null, "dllm_algorithm_config": null, "dllm_fdfo": true, "disaggregation_mode": "prefill", "disaggregation_transfer_backend": "mooncake", "disaggregation_bootstrap_port": 28800, "disaggregation_ib_device": "mlx5_0,mlx5_1,mlx5_2,mlx5_3", "disaggregation_decode_enable_radix_cache": false, "disaggregation_decode_enable_offload_kvcache": false, "num_reserved_decode_tokens": 512, "disaggregation_decode_extra_slots": null, "disaggregation_decode_polling_interval": 1, "optimistic_prefill_attempts": 0, "encoder_only": false, "language_only": false, "encoder_transfer_backend": "zmq_to_tokenizer", "encoder_urls": [], "encoder_bootstrap_port": 8997, "encoder_register_urls": [], "enable_adaptive_dispatch_to_encoder": false, "enable_pdmux": false, "pdmux_config_path": null, "sm_group_num": 8, "custom_weight_loader": [], "weight_loader_disable_mmap": false, "weight_loader_prefetch_checkpoints": false, "weight_loader_prefetch_num_threads": 4, "weight_loader_drop_cache_after_load": false, "remote_instance_weight_loader_seed_instance_ip": null, "remote_instance_weight_loader_seed_instance_service_port": null, "remote_instance_weight_loader_send_weights_group_ports": null, "remote_instance_weight_loader_backend": "nccl", "remote_instance_weight_loader_start_seed_via_transfer_engine": false, "engine_info_bootstrap_port": 6789, "modelexpress_config": null, "download_dir": null, "model_checksum": null, "delete_ckpt_after_loading": false, "decrypted_config_file": null, "decrypted_draft_config_file": null, "checkpoint_engine_wait_weights_before_ready": false, "enable_prefill_delayer": false, "prefill_delayer_max_delay_passes": 30, "prefill_delayer_token_usage_low_watermark": null, "prefill_delayer_forward_passes_buckets": null, "prefill_delayer_wait_seconds_buckets": null, "prefill_delayer_queue_min_ratio": null, "prefill_delayer_max_delay_ms": null, "min_free_slots_delay": null, "enable_deterministic_inference": false, "rl_on_policy_target": null, "kv_canary": "none", "kv_canary_real_data": "none", "kv_canary_sweep_interval": 0, "enable_dynamic_batch_tokenizer": false, "dynamic_batch_tokenizer_batch_size": 32, "dynamic_batch_tokenizer_batch_timeout": 0.002, "enable_tokenizer_batch_encode": false, "disable_tokenizer_batch_decode": false, "debug_tensor_dump_output_folder": null, "debug_tensor_dump_layers": null, "debug_tensor_dump_input_file": null, "enable_memory_saver": false, "enable_weights_cpu_backup": false, "enable_draft_weights_cpu_backup": false, "enable_custom_logit_processor": false, "enable_return_hidden_states": false, "enable_return_routed_experts": false, "enable_return_indexer_topk": false, "disable_outlines_disk_cache": false, "enable_mis": false, "weight_cache_mode": "off", "weight_cache_socket": null, "weight_cache_timeout": 1800, "forward_hooks": null, "msprobe_dump_config": null, "status": "ready", "max_total_num_tokens": 3801408, "max_req_input_len": 1048570, "internal_states": [{"model_path": "/data/hf_models/Kimi-K3", "tokenizer_path": "/data/hf_models/Kimi-K3", "tokenizer_mode": "auto", "tokenizer_backend": "huggingface", "tokenizer_worker_num": 1, "detokenizer_worker_num": 1, "skip_tokenizer_init": false, "load_format": "auto", "model_loader_extra_config": "{}", "trust_remote_code": true, "context_length": null, "is_embedding": false, "enable_multimodal": null, "revision": null, "model_impl": "auto", "model_config_parser": "auto", "json_model_override_args": "{}", "dtype": "auto", "quantization": null, "quantization_param_path": null, "kv_cache_dtype": "auto", "enable_fp32_lm_head": false, "modelopt_quant": null, "modelopt_checkpoint_restore_path": null, "modelopt_checkpoint_save_path": null, "modelopt_export_path": null, "quantize_and_serve": false, "rl_quant_profile": null, "enable_tf32_matmul": false, "mem_fraction_static": 0.88, "max_running_requests": null, "max_queued_requests": null, "max_total_tokens": null, "chunked_prefill_size": 8192, "enable_dynamic_chunking": false, "max_prefill_tokens": 16384, "prefill_max_requests": null, "schedule_policy": "fcfs", "enable_priority_scheduling": false, "disable_priority_preemption": false, "default_priority_value": null, "abort_on_priority_when_disabled": false, "schedule_low_priority_values_first": false, "priority_scheduling_preemption_threshold": 10, "retraction_policy": "length", "schedule_conservativeness": 1.0, "page_size": 64, "swa_full_tokens_ratio": 0.8, "disable_hybrid_swa_memory": false, "radix_eviction_policy": "lru", "prefill_only_disable_kv_cache": false, "disable_radix_cache": true, "enable_page_major_kv_layout": false, "enable_unified_memory": false, "disable_chunked_prefix_cache": true, "disable_overlap_schedule": true, "num_continuous_decode_steps": 1, "scheduler_recv_interval": 1, "enable_mixed_chunk": false, "nccl_port": null, "dist_timeout": 3600, "dist_init_addr": "174.1.60.1:20000", "nnodes": 4, "node_rank": 0, "tp_size": 4, "dcp_size": 1, "pp_size": 8, "pp_max_micro_batch_size": 6, "pp_async_batch_depth": 0, "dp_size": 1, "load_balance_method": "follow_bootstrap_room", "attn_cp_size": 1, "moe_dp_size": 1, "dcp_comm_backend": "ag_rs", "dcp_replicate_q_proj": null, "dwdp_size": 1, "enable_prefill_cp": false, "cp_strategy": null, "enable_dsa_cache_layer_split": false, "enable_dsa_prefill_context_parallel": false, "dsa_prefill_cp_mode": "round-robin-split", "enable_prefill_context_parallel": false, "prefill_cp_mode": "in-seq-split", "enable_cp_decode_attn_tp": false, "enable_dp_attention": false, "enable_dp_attention_local_control_broadcast": false, "enable_dp_lm_head": false, "enable_attn_tp_input_scattered": false, "disable_attn_tp_gather": false, "enable_p2p_check": false, "device": "cuda", "base_gpu_id": 0, "gpu_id_step": 1, "random_seed": 984607766, "watchdog_timeout": 300, "soft_watchdog_timeout": null, "sleep_on_idle": false, "use_ray": false, "numa_node": null, "gc_threshold": null, "host": "0.0.0.0", "port": 30000, "fastapi_root_path": "", "smg_grpc_mode": false, "grpc_mode": false, "grpc_port": null, "sidecar": null, "sidecar_args": null, "skip_server_warmup": false, "warmups": null, "enable_http2": false, "ssl_keyfile": null, "ssl_certfile": null, "ssl_ca_certs": null, "ssl_keyfile_password": null, "enable_ssl_refresh": false, "api_key": null, "admin_api_key": null, "served_model_name": "kimi-k3", "weight_version": "default", "chat_template": null, "hf_chat_template_name": null, "completion_template": null, "file_storage_path": "sglang_storage", "enable_cache_report": false, "reasoning_parser": null, "default_chat_template_kwargs": null, "strip_thinking_cache": false, "enable_strict_thinking": false, "tool_call_parser": null, "tool_server": null, "sampling_defaults": "model", "asr_max_buffer_seconds": 60, "asr_max_concurrent_sessions": 32, "preferred_sampling_params": null, "allow_auto_truncate": false, "stream_interval": 1, "batch_notify_size": 16, "stream_response_default_include_usage": false, "incremental_streaming_output": false, "enable_streaming_session": false, "enable_session_radix_cache": false, "log_level": "info", "log_level_http": null, "log_requests": false, "log_requests_level": 2, "log_requests_format": "text", "log_requests_target": null, "uvicorn_access_log_exclude_prefixes": [], "crash_dump_folder": null, "show_time_cost": false, "enable_metrics": false, "smg_http_sidecar_port": null, "enable_mfu_metrics": false, "enable_metrics_for_all_schedulers": false, "load_snapshot_publish_interval": 15, "tokenizer_metrics_custom_labels_header": "x-custom-labels", "tokenizer_metrics_allowed_custom_labels": null, "extra_metric_labels": null, "bucket_time_to_first_token": null, "bucket_inter_token_latency": null, "bucket_e2e_request_latency": null, "prompt_tokens_buckets": null, "generation_tokens_buckets": null, "gc_warning_threshold_secs": 0.0, "decode_log_interval": 40, "enable_request_time_stats_logging": false, "kv_events_config": null, "enable_forward_pass_metrics": false, "forward_pass_metrics_worker_id": "", "forward_pass_metrics_ipc_name": null, "enable_trace": false, "trace_modules": "request", "otlp_traces_endpoint": "localhost:4317", "export_metrics_to_file": false, "export_metrics_to_file_dir": null, "stat_loggers": null, "constrained_json_whitespace_pattern": null, "constrained_json_disable_any_whitespace": false, "attention_backend": "triton", "decode_attention_backend": null, "prefill_attention_backend": null, "sampling_backend": "flashinfer", "grammar_backend": "xgrammar", "radix_cache_backend": null, "mm_attention_backend": null, "fp8_gemm_runner_backend": "auto", "fp4_gemm_runner_backend": "auto", "bf16_gemm_backend": "auto", "dsa_prefill_backend": null, "dsa_decode_backend": null, "dsa_paged_mqa_logits_backend": "auto", "dsa_topk_backend": "sgl-kernel", "disable_flashinfer_autotune": false, "flashinfer_autotune_skip_ops": null, "mamba_backend": "triton", "cuda_graph_config": {"decode": {"backend": "disabled", "max_bs": 16, "bs": [1, 2, 4, 8, 12, 16], "tc_compiler": "eager", "full_prefill_max_req": null}, "prefill": {"backend": "disabled", "max_bs": 2048, "bs": [4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], "tc_compiler": "eager", "full_prefill_max_req": null}}, "cuda_graph_backend_decode": null, "cuda_graph_backend_prefill": null, "cuda_graph_max_bs_decode": 16, "cuda_graph_max_bs_prefill": null, "cuda_graph_bs_decode": null, "cuda_graph_bs_prefill": null, "cuda_graph_tc_compiler": null, "disable_prefill_cuda_graph": false, "disable_decode_cuda_graph": false, "disable_cuda_graph": false, "disable_cuda_graph_padding": false, "enable_profile_cuda_graph": false, "enable_cudagraph_gc": false, "debug_cuda_graph": false, "enable_layerwise_nvtx_marker": false, "enable_nccl_nvls": false, "enable_symm_mem": false, "triton_attention_reduce_in_fp32": false, "triton_attention_num_kv_splits": 8, "triton_attention_split_tile_size": null, "flashinfer_mla_disable_ragged": false, "enable_fused_qk_norm_rope": false, "enable_precise_embedding_interpolation": false, "enable_fused_moe_sum_all_reduce": false, "enable_deepseek_v4_fp4_indexer": false, "disable_custom_all_reduce": false, "enable_mscclpp": false, "enable_torch_symm_mem": false, "enable_scattered_sconv": false, "pre_warm_nccl": false, "enable_quant_communications": false, "enable_flashinfer_allreduce_fusion": false, "enforce_disable_flashinfer_allreduce_fusion": false, "flashinfer_allreduce_fusion_backend": null, "enable_aiter_allreduce_fusion": false, "enable_torch_compile": false, "enable_torch_compile_debug_mode": false, "torch_compile_max_bs": 32, "torchao_config": "", "speculative_algorithm": null, "speculative_draft_model_path": null, "speculative_draft_model_revision": null, "speculative_draft_load_format": null, "speculative_num_steps": null, "speculative_eagle_topk": null, "speculative_num_draft_tokens": null, "speculative_dflash_block_size": null, "speculative_dspark_block_size": null, "speculative_dspark_sps_table_path": null, "speculative_dspark_confidence_sts_path": null, "speculative_dspark_align_verify_tokens_to_graph_tier": false, "speculative_accept_threshold_single": 1.0, "speculative_accept_threshold_acc": 1.0, "speculative_use_rejection_sampling": false, "speculative_token_map": null, "speculative_attention_mode": "prefill", "speculative_draft_attention_backend": null, "speculative_draft_window_size": null, "speculative_moe_runner_backend": "flashinfer_mxfp4", "speculative_moe_a2a_backend": null, "speculative_draft_model_quantization": null, "speculative_skip_dp_mlp_sync": false, "enable_multi_layer_eagle": false, "speculative_adaptive": false, "speculative_adaptive_config": null, "decoupled_spec_bind_endpoint": null, "decoupled_spec_connect_endpoints": null, "decoupled_spec_rank": null, "decoupled_spec_role": "null", "spec_trace_dir": null, "speculative_ngram_min_bfs_breadth": 1, "speculative_ngram_max_bfs_breadth": 10, "speculative_ngram_match_type": "BFS", "speculative_ngram_max_trie_depth": 18, "speculative_ngram_capacity": 10000000, "speculative_ngram_external_corpus_path": null, "speculative_ngram_external_sam_budget": 0, "speculative_ngram_external_corpus_max_tokens": 10000000, "ep_size": 4, "moe_a2a_backend": "none", "moe_runner_backend": "flashinfer_mxfp4", "flashinfer_mxfp4_moe_precision": "default", "deepep_mode": "auto", "fuseep_mode": 2, "deepep_dispatcher_output_dtype": "auto", "ep_num_redundant_experts": 0, "ep_dispatch_algorithm": null, "init_expert_location": "trivial", "enable_eplb": false, "eplb_algorithm": "auto", "eplb_rebalance_num_iterations": 1000, "eplb_rebalance_layers_per_chunk": null, "eplb_min_rebalancing_utilization_threshold": 1.0, "expert_distribution_recorder_mode": null, "expert_distribution_recorder_buffer_size": 1000, "enable_expert_distribution_metrics": false, "deepep_config": null, "moe_dense_tp_size": null, "elastic_ep_backend": null, "enable_elastic_expert_backup": false, "mooncake_ib_device": null, "enable_waterfill": false, "ep_join_mode": null, "ep_join_rank_offset": 0, "elastic_ep_initial_size": null, "max_ep_size": null, "elastic_ep_scale_timeout": 600, "elastic_ep_rejoin": false, "disable_flashinfer_cutlass_moe_fp4_allgather": false, "disable_shared_experts_fusion": false, "enforce_shared_experts_fusion": false, "max_mamba_cache_size": 54, "mamba_ssm_dtype": null, "mamba_max_states_per_path": -1, "enable_mamba_cache_stochastic_rounding": false, "mamba_cache_philox_rounds": 0, "mamba_full_memory_ratio": 0.36, "mamba_radix_cache_strategy": "extra_buffer_lazy", "uses_mamba_radix_cache": false, "mamba_track_interval": 256, "enable_int8_mamba_checkpoint": false, "int8_mamba_ckpt_size": null, "linear_attn_backend": "triton", "linear_attn_decode_backend": null, "linear_attn_prefill_backend": null, "linear_attn_verify_backend": null, "enable_linear_replayssm": false, "linear_replayssm_cache_len": 16, "enable_linear_replayssm_spec": false, "enable_hierarchical_cache": false, "hicache_ratio": 2.0, "hicache_size": 0, "hicache_write_policy": "write_through", "hicache_io_backend": "kernel", "hicache_mem_layout": "page_first", "hicache_storage_backend": null, "hicache_storage_prefetch_policy": "timeout", "hicache_storage_backend_extra_config": null, "enable_hisparse": false, "hisparse_config": null, "enable_broadcast_mm_inputs_process": false, "enable_prefix_mm_cache": false, "mm_enable_dp_encoder": false, "mm_process_config": {}, "mm_processor_worker_num": 0, "mm_io_worker_num": 0, "limit_mm_data_per_request": null, "enable_mm_global_cache": false, "disable_fast_image_processor": false, "mm_feature_transport": "cpu", "keep_mm_feature_on_device": false, "enable_lora": null, "enable_lora_overlap_loading": null, "max_lora_rank": null, "lora_target_modules": null, "lora_paths": null, "max_loaded_loras": null, "max_loras_per_batch": 8, "lora_eviction_policy": "lru", "lora_backend": "csgmv", "max_lora_chunk_size": 16, "experts_shared_outer_loras": null, "lora_use_virtual_experts": false, "lora_strict_loading": false, "lora_drain_wait_threshold": 0.0, "enable_two_batch_overlap": false, "enable_single_batch_overlap": false, "tbo_token_distribution_threshold": 0.48, "cpu_offload_gb": 0, "offload_group_size": -1, "offload_num_in_group": 1, "offload_prefetch_step": 1, "offload_mode": "cpu", "enable_lmcache": false, "lmcache_config_file": null, "enable_flexkv": false, "flexkv_config_file": null, "kt_weight_path": null, "kt_method": "AMXINT4", "kt_cpuinfer": null, "kt_threadpool_count": 2, "kt_num_gpu_experts": null, "kt_max_deferred_experts_per_token": null, "dllm_algorithm": null, "dllm_algorithm_config": null, "dllm_fdfo": true, "disaggregation_mode": "prefill", "disaggregation_transfer_backend": "mooncake", "disaggregation_bootstrap_port": 28800, "disaggregation_ib_device": "mlx5_0,mlx5_1,mlx5_2,mlx5_3", "disaggregation_decode_enable_radix_cache": false, "disaggregation_decode_enable_offload_kvcache": false, "num_reserved_decode_tokens": 512, "disaggregation_decode_extra_slots": null, "disaggregation_decode_polling_interval": 1, "optimistic_prefill_attempts": 0, "encoder_only": false, "language_only": false, "encoder_transfer_backend": "zmq_to_tokenizer", "encoder_urls": [], "encoder_bootstrap_port": 8997, "encoder_register_urls": [], "enable_adaptive_dispatch_to_encoder": false, "enable_pdmux": false, "pdmux_config_path": null, "sm_group_num": 8, "custom_weight_loader": [], "weight_loader_disable_mmap": false, "weight_loader_prefetch_checkpoints": false, "weight_loader_prefetch_num_threads": 4, "weight_loader_drop_cache_after_load": false, "remote_instance_weight_loader_seed_instance_ip": null, "remote_instance_weight_loader_seed_instance_service_port": null, "remote_instance_weight_loader_send_weights_group_ports": null, "remote_instance_weight_loader_backend": "nccl", "remote_instance_weight_loader_start_seed_via_transfer_engine": false, "engine_info_bootstrap_port": 6789, "modelexpress_config": null, "download_dir": null, "model_checksum": null, "delete_ckpt_after_loading": false, "decrypted_config_file": null, "decrypted_draft_config_file": null, "checkpoint_engine_wait_weights_before_ready": false, "enable_prefill_delayer": false, "prefill_delayer_max_delay_passes": 30, "prefill_delayer_token_usage_low_watermark": null, "prefill_delayer_forward_passes_buckets": null, "prefill_delayer_wait_seconds_buckets": null, "prefill_delayer_queue_min_ratio": null, "prefill_delayer_max_delay_ms": null, "min_free_slots_delay": null, "enable_deterministic_inference": false, "rl_on_policy_target": null, "kv_canary": "none", "kv_canary_real_data": "none", "kv_canary_sweep_interval": 0, "enable_dynamic_batch_tokenizer": false, "dynamic_batch_tokenizer_batch_size": 32, "dynamic_batch_tokenizer_batch_timeout": 0.002, "enable_tokenizer_batch_encode": false, "disable_tokenizer_batch_decode": false, "debug_tensor_dump_output_folder": null, "debug_tensor_dump_layers": null, "debug_tensor_dump_input_file": null, "enable_memory_saver": false, "enable_weights_cpu_backup": false, "enable_draft_weights_cpu_backup": false, "enable_custom_logit_processor": false, "enable_return_hidden_states": false, "enable_return_routed_experts": false, "enable_return_indexer_topk": false, "disable_outlines_disk_cache": false, "enable_mis": false, "weight_cache_mode": "off", "weight_cache_socket": null, "weight_cache_timeout": 1800, "forward_hooks": null, "msprobe_dump_config": null, "_resolved_overrides": [["_sampling_backend_default", {"sampling_backend": "flashinfer"}], ["_attention_backend_default", {"attention_backend": "triton"}], ["_data_parallelism_defaults", {"enable_dp_attention": false, "enable_dp_lm_head": false}], ["_pipeline_parallel_overlap_disable", {"disable_overlap_schedule": true}], ["_speculative_moe_runner_default", {"speculative_moe_runner_backend": "flashinfer_mxfp4"}]], "grpc_worker_threads": 4, "_quantization_explicitly_unset": false, "_cuda_graph_config_locked": [["decode", "max_bs"]], "_declarations_materialized": true, "_runtime_mutations": [["model_runner.chunked_prefix_cache_gate", {"disable_chunked_prefix_cache": true}], ["mamba_pool.memory_budget", {"max_mamba_cache_size": 54}]], "_in_override": false, "_mx_config_cache": {}, "max_speculative_num_draft_tokens": null, "last_gen_throughput": 0.0, "memory_usage": {"weight": 43.09, "kvcache": 8.16, "token_capacity": 3801408, "graph": 0}, "effective_max_running_requests_per_dp": 54}], "version": "0.5.16", "kv_events": null}], "decode": [{"model_path": "/data/hf_models/Kimi-K3", "tokenizer_path": "/data/hf_models/Kimi-K3", "tokenizer_mode": "auto", "tokenizer_backend": "huggingface", "tokenizer_worker_num": 1, "detokenizer_worker_num": 1, "skip_tokenizer_init": false, "load_format": "auto", "model_loader_extra_config": "{}", "trust_remote_code": true, "context_length": null, "is_embedding": false, "enable_multimodal": null, "revision": null, "model_impl": "auto", "model_config_parser": "auto", "json_model_override_args": "{}", "dtype": "auto", "quantization": null, "quantization_param_path": null, "kv_cache_dtype": "auto", "enable_fp32_lm_head": false, "modelopt_quant": null, "modelopt_checkpoint_restore_path": null, "modelopt_checkpoint_save_path": null, "modelopt_export_path": null, "quantize_and_serve": false, "rl_quant_profile": null, "enable_tf32_matmul": false, "mem_fraction_static": 0.88, "max_running_requests": null, "max_queued_requests": null, "max_total_tokens": null, "chunked_prefill_size": 8192, "enable_dynamic_chunking": false, "max_prefill_tokens": 16384, "prefill_max_requests": null, "schedule_policy": "fcfs", "enable_priority_scheduling": false, "disable_priority_preemption": false, "default_priority_value": null, "abort_on_priority_when_disabled": false, "schedule_low_priority_values_first": false, "priority_scheduling_preemption_threshold": 10, "retraction_policy": "length", "schedule_conservativeness": 1.0, "page_size": 64, "swa_full_tokens_ratio": 0.8, "disable_hybrid_swa_memory": false, "radix_eviction_policy": "lru", "prefill_only_disable_kv_cache": false, "disable_radix_cache": true, "enable_page_major_kv_layout": false, "enable_unified_memory": false, "disable_chunked_prefix_cache": false, "disable_overlap_schedule": false, "num_continuous_decode_steps": 1, "scheduler_recv_interval": 1, "enable_mixed_chunk": false, "nccl_port": null, "dist_timeout": 3600, "dist_init_addr": "174.1.60.5:20000", "nnodes": 4, "node_rank": 0, "tp_size": 32, "dcp_size": 1, "pp_size": 1, "pp_max_micro_batch_size": null, "pp_async_batch_depth": 0, "dp_size": 1, "load_balance_method": "round_robin", "attn_cp_size": 1, "moe_dp_size": 1, "dcp_comm_backend": "ag_rs", "dcp_replicate_q_proj": null, "dwdp_size": 1, "enable_prefill_cp": false, "cp_strategy": null, "enable_dsa_cache_layer_split": false, "enable_dsa_prefill_context_parallel": false, "dsa_prefill_cp_mode": "round-robin-split", "enable_prefill_context_parallel": false, "prefill_cp_mode": "in-seq-split", "enable_cp_decode_attn_tp": false, "enable_dp_attention": false, "enable_dp_attention_local_control_broadcast": false, "enable_dp_lm_head": false, "enable_attn_tp_input_scattered": false, "disable_attn_tp_gather": false, "enable_p2p_check": false, "device": "cuda", "base_gpu_id": 0, "gpu_id_step": 1, "random_seed": 369235288, "watchdog_timeout": 300, "soft_watchdog_timeout": null, "sleep_on_idle": false, "use_ray": false, "custom_sigquit_handler": null, "numa_node": null, "gc_threshold": null, "host": "0.0.0.0", "port": 30000, "fastapi_root_path": "", "smg_grpc_mode": false, "grpc_mode": false, "grpc_port": null, "sidecar": null, "sidecar_args": null, "skip_server_warmup": false, "warmups": null, "enable_http2": false, "ssl_keyfile": null, "ssl_certfile": null, "ssl_ca_certs": null, "ssl_keyfile_password": null, "enable_ssl_refresh": false, "api_key": null, "admin_api_key": null, "served_model_name": "kimi-k3", "weight_version": "default", "chat_template": null, "hf_chat_template_name": null, "completion_template": null, "file_storage_path": "sglang_storage", "enable_cache_report": false, "reasoning_parser": null, "default_chat_template_kwargs": null, "strip_thinking_cache": false, "enable_strict_thinking": false, "tool_call_parser": null, "tool_server": null, "sampling_defaults": "model", "asr_max_buffer_seconds": 60, "asr_max_concurrent_sessions": 32, "preferred_sampling_params": null, "allow_auto_truncate": false, "stream_interval": 1, "batch_notify_size": 16, "stream_response_default_include_usage": false, "incremental_streaming_output": false, "enable_streaming_session": false, "enable_session_radix_cache": false, "log_level": "info", "log_level_http": null, "log_requests": false, "log_requests_level": 2, "log_requests_format": "text", "log_requests_target": null, "uvicorn_access_log_exclude_prefixes": [], "crash_dump_folder": null, "show_time_cost": false, "enable_metrics": false, "smg_http_sidecar_port": null, "enable_mfu_metrics": false, "enable_metrics_for_all_schedulers": false, "load_snapshot_publish_interval": 15, "tokenizer_metrics_custom_labels_header": "x-custom-labels", "tokenizer_metrics_allowed_custom_labels": null, "extra_metric_labels": null, "bucket_time_to_first_token": null, "bucket_inter_token_latency": null, "bucket_e2e_request_latency": null, "prompt_tokens_buckets": null, "generation_tokens_buckets": null, "gc_warning_threshold_secs": 0.0, "decode_log_interval": 40, "enable_request_time_stats_logging": false, "kv_events_config": null, "enable_forward_pass_metrics": false, "forward_pass_metrics_worker_id": "", "forward_pass_metrics_ipc_name": null, "enable_trace": false, "trace_modules": "request", "otlp_traces_endpoint": "localhost:4317", "export_metrics_to_file": false, "export_metrics_to_file_dir": null, "stat_loggers": null, "constrained_json_whitespace_pattern": null, "constrained_json_disable_any_whitespace": false, "attention_backend": "triton", "decode_attention_backend": null, "prefill_attention_backend": null, "sampling_backend": "flashinfer", "grammar_backend": "xgrammar", "radix_cache_backend": null, "mm_attention_backend": null, "fp8_gemm_runner_backend": "auto", "fp4_gemm_runner_backend": "auto", "bf16_gemm_backend": "auto", "dsa_prefill_backend": null, "dsa_decode_backend": null, "dsa_paged_mqa_logits_backend": "auto", "dsa_topk_backend": "sgl-kernel", "disable_flashinfer_autotune": false, "flashinfer_autotune_skip_ops": null, "mamba_backend": "triton", "cuda_graph_config": {"decode": {"backend": "full", "max_bs": 16, "bs": [1, 2, 4, 8, 12, 16], "tc_compiler": "eager", "full_prefill_max_req": null}, "prefill": {"backend": "disabled", "max_bs": 2048, "bs": [4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], "tc_compiler": "eager", "full_prefill_max_req": null}}, "cuda_graph_backend_decode": null, "cuda_graph_backend_prefill": null, "cuda_graph_max_bs_decode": 16, "cuda_graph_max_bs_prefill": null, "cuda_graph_bs_decode": null, "cuda_graph_bs_prefill": null, "cuda_graph_tc_compiler": null, "disable_prefill_cuda_graph": false, "disable_decode_cuda_graph": false, "disable_cuda_graph": false, "disable_cuda_graph_padding": false, "enable_profile_cuda_graph": false, "enable_cudagraph_gc": false, "debug_cuda_graph": false, "enable_layerwise_nvtx_marker": false, "enable_nccl_nvls": false, "enable_symm_mem": false, "triton_attention_reduce_in_fp32": false, "triton_attention_num_kv_splits": 8, "triton_attention_split_tile_size": null, "flashinfer_mla_disable_ragged": false, "enable_fused_qk_norm_rope": false, "enable_precise_embedding_interpolation": false, "enable_fused_moe_sum_all_reduce": false, "enable_deepseek_v4_fp4_indexer": false, "disable_custom_all_reduce": false, "enable_mscclpp": false, "enable_torch_symm_mem": false, "enable_scattered_sconv": false, "pre_warm_nccl": false, "enable_quant_communications": false, "enable_flashinfer_allreduce_fusion": false, "enforce_disable_flashinfer_allreduce_fusion": false, "flashinfer_allreduce_fusion_backend": null, "enable_aiter_allreduce_fusion": false, "enable_torch_compile": false, "enable_torch_compile_debug_mode": false, "torch_compile_max_bs": 32, "torchao_config": "", "speculative_algorithm": null, "speculative_draft_model_path": null, "speculative_draft_model_revision": null, "speculative_draft_load_format": null, "speculative_num_steps": null, "speculative_eagle_topk": null, "speculative_num_draft_tokens": null, "speculative_dflash_block_size": null, "speculative_dspark_block_size": null, "speculative_dspark_sps_table_path": null, "speculative_dspark_confidence_sts_path": null, "speculative_dspark_align_verify_tokens_to_graph_tier": false, "speculative_accept_threshold_single": 1.0, "speculative_accept_threshold_acc": 1.0, "speculative_use_rejection_sampling": false, "speculative_token_map": null, "speculative_attention_mode": "prefill", "speculative_draft_attention_backend": null, "speculative_draft_window_size": null, "speculative_moe_runner_backend": "marlin", "speculative_moe_a2a_backend": null, "speculative_draft_model_quantization": null, "speculative_skip_dp_mlp_sync": false, "enable_multi_layer_eagle": false, "speculative_adaptive": false, "speculative_adaptive_config": null, "decoupled_spec_bind_endpoint": null, "decoupled_spec_connect_endpoints": null, "decoupled_spec_rank": null, "decoupled_spec_role": "null", "spec_trace_dir": null, "speculative_ngram_min_bfs_breadth": 1, "speculative_ngram_max_bfs_breadth": 10, "speculative_ngram_match_type": "BFS", "speculative_ngram_max_trie_depth": 18, "speculative_ngram_capacity": 10000000, "speculative_ngram_external_corpus_path": null, "speculative_ngram_external_sam_budget": 0, "speculative_ngram_external_corpus_max_tokens": 10000000, "ep_size": 32, "moe_a2a_backend": "none", "moe_runner_backend": "marlin", "flashinfer_mxfp4_moe_precision": "default", "deepep_mode": "auto", "fuseep_mode": 2, "deepep_dispatcher_output_dtype": "auto", "ep_num_redundant_experts": 0, "ep_dispatch_algorithm": null, "init_expert_location": "trivial", "enable_eplb": false, "eplb_algorithm": "auto", "eplb_rebalance_num_iterations": 1000, "eplb_rebalance_layers_per_chunk": null, "eplb_min_rebalancing_utilization_threshold": 1.0, "expert_distribution_recorder_mode": null, "expert_distribution_recorder_buffer_size": 1000, "enable_expert_distribution_metrics": false, "deepep_config": null, "moe_dense_tp_size": null, "elastic_ep_backend": null, "enable_elastic_expert_backup": false, "mooncake_ib_device": null, "enable_waterfill": false, "ep_join_mode": null, "ep_join_rank_offset": 0, "elastic_ep_initial_size": null, "max_ep_size": null, "elastic_ep_scale_timeout": 600, "elastic_ep_rejoin": false, "disable_flashinfer_cutlass_moe_fp4_allgather": false, "disable_shared_experts_fusion": false, "enforce_shared_experts_fusion": false, "max_mamba_cache_size": null, "mamba_ssm_dtype": null, "mamba_max_states_per_path": -1, "enable_mamba_cache_stochastic_rounding": false, "mamba_cache_philox_rounds": 0, "mamba_full_memory_ratio": 0.21, "mamba_radix_cache_strategy": "extra_buffer_lazy", "uses_mamba_radix_cache": false, "mamba_track_interval": 256, "enable_int8_mamba_checkpoint": false, "int8_mamba_ckpt_size": null, "linear_attn_backend": "triton", "linear_attn_decode_backend": null, "linear_attn_prefill_backend": null, "linear_attn_verify_backend": null, "enable_linear_replayssm": false, "linear_replayssm_cache_len": 16, "enable_linear_replayssm_spec": false, "enable_hierarchical_cache": false, "hicache_ratio": 2.0, "hicache_size": 0, "hicache_write_policy": "write_through", "hicache_io_backend": "kernel", "hicache_mem_layout": "page_first", "hicache_storage_backend": null, "hicache_storage_prefetch_policy": "timeout", "hicache_storage_backend_extra_config": null, "enable_hisparse": false, "hisparse_config": null, "enable_broadcast_mm_inputs_process": false, "enable_prefix_mm_cache": false, "mm_enable_dp_encoder": false, "mm_process_config": {}, "mm_processor_worker_num": 0, "mm_io_worker_num": 0, "limit_mm_data_per_request": null, "enable_mm_global_cache": false, "disable_fast_image_processor": false, "mm_feature_transport": "cpu", "keep_mm_feature_on_device": false, "enable_lora": null, "enable_lora_overlap_loading": null, "max_lora_rank": null, "lora_target_modules": null, "lora_paths": null, "max_loaded_loras": null, "max_loras_per_batch": 8, "lora_eviction_policy": "lru", "lora_backend": "csgmv", "max_lora_chunk_size": 16, "experts_shared_outer_loras": null, "lora_use_virtual_experts": false, "lora_strict_loading": false, "lora_drain_wait_threshold": 0.0, "enable_two_batch_overlap": false, "enable_single_batch_overlap": false, "tbo_token_distribution_threshold": 0.48, "cpu_offload_gb": 0, "offload_group_size": -1, "offload_num_in_group": 1, "offload_prefetch_step": 1, "offload_mode": "cpu", "enable_lmcache": false, "lmcache_config_file": null, "enable_flexkv": false, "flexkv_config_file": null, "kt_weight_path": null, "kt_method": "AMXINT4", "kt_cpuinfer": null, "kt_threadpool_count": 2, "kt_num_gpu_experts": null, "kt_max_deferred_experts_per_token": null, "dllm_algorithm": null, "dllm_algorithm_config": null, "dllm_fdfo": true, "disaggregation_mode": "decode", "disaggregation_transfer_backend": "mooncake", "disaggregation_bootstrap_port": 28800, "disaggregation_ib_device": "mlx5_0,mlx5_1,mlx5_2,mlx5_3", "disaggregation_decode_enable_radix_cache": false, "disaggregation_decode_enable_offload_kvcache": false, "num_reserved_decode_tokens": 512, "disaggregation_decode_extra_slots": 0, "disaggregation_decode_polling_interval": 1, "optimistic_prefill_attempts": 0, "encoder_only": false, "language_only": false, "encoder_transfer_backend": "zmq_to_tokenizer", "encoder_urls": [], "encoder_bootstrap_port": 8997, "encoder_register_urls": [], "enable_adaptive_dispatch_to_encoder": false, "enable_pdmux": false, "pdmux_config_path": null, "sm_group_num": 8, "custom_weight_loader": [], "weight_loader_disable_mmap": false, "weight_loader_prefetch_checkpoints": false, "weight_loader_prefetch_num_threads": 4, "weight_loader_drop_cache_after_load": false, "remote_instance_weight_loader_seed_instance_ip": null, "remote_instance_weight_loader_seed_instance_service_port": null, "remote_instance_weight_loader_send_weights_group_ports": null, "remote_instance_weight_loader_backend": "nccl", "remote_instance_weight_loader_start_seed_via_transfer_engine": false, "engine_info_bootstrap_port": 6789, "modelexpress_config": null, "download_dir": null, "model_checksum": null, "delete_ckpt_after_loading": false, "decrypted_config_file": null, "decrypted_draft_config_file": null, "checkpoint_engine_wait_weights_before_ready": false, "enable_prefill_delayer": false, "prefill_delayer_max_delay_passes": 30, "prefill_delayer_token_usage_low_watermark": null, "prefill_delayer_forward_passes_buckets": null, "prefill_delayer_wait_seconds_buckets": null, "prefill_delayer_queue_min_ratio": null, "prefill_delayer_max_delay_ms": null, "min_free_slots_delay": null, "enable_deterministic_inference": false, "rl_on_policy_target": null, "kv_canary": "none", "kv_canary_real_data": "none", "kv_canary_sweep_interval": 0, "enable_dynamic_batch_tokenizer": false, "dynamic_batch_tokenizer_batch_size": 32, "dynamic_batch_tokenizer_batch_timeout": 0.002, "enable_tokenizer_batch_encode": false, "disable_tokenizer_batch_decode": false, "debug_tensor_dump_output_folder": null, "debug_tensor_dump_layers": null, "debug_tensor_dump_input_file": null, "enable_memory_saver": false, "enable_weights_cpu_backup": false, "enable_draft_weights_cpu_backup": false, "enable_custom_logit_processor": false, "enable_return_hidden_states": false, "enable_return_routed_experts": false, "enable_return_indexer_topk": false, "disable_outlines_disk_cache": false, "enable_mis": false, "weight_cache_mode": "off", "weight_cache_socket": null, "weight_cache_timeout": 1800, "forward_hooks": null, "msprobe_dump_config": null, "status": "ready", "max_total_num_tokens": 440256, "max_req_input_len": 440250, "internal_states": [{"model_path": "/data/hf_models/Kimi-K3", "tokenizer_path": "/data/hf_models/Kimi-K3", "tokenizer_mode": "auto", "tokenizer_backend": "huggingface", "tokenizer_worker_num": 1, "detokenizer_worker_num": 1, "skip_tokenizer_init": false, "load_format": "auto", "model_loader_extra_config": "{}", "trust_remote_code": true, "context_length": null, "is_embedding": false, "enable_multimodal": null, "revision": null, "model_impl": "auto", "model_config_parser": "auto", "json_model_override_args": "{}", "dtype": "auto", "quantization": null, "quantization_param_path": null, "kv_cache_dtype": "auto", "enable_fp32_lm_head": false, "modelopt_quant": null, "modelopt_checkpoint_restore_path": null, "modelopt_checkpoint_save_path": null, "modelopt_export_path": null, "quantize_and_serve": false, "rl_quant_profile": null, "enable_tf32_matmul": false, "mem_fraction_static": 0.88, "max_running_requests": null, "max_queued_requests": null, "max_total_tokens": null, "chunked_prefill_size": 8192, "enable_dynamic_chunking": false, "max_prefill_tokens": 16384, "prefill_max_requests": null, "schedule_policy": "fcfs", "enable_priority_scheduling": false, "disable_priority_preemption": false, "default_priority_value": null, "abort_on_priority_when_disabled": false, "schedule_low_priority_values_first": false, "priority_scheduling_preemption_threshold": 10, "retraction_policy": "length", "schedule_conservativeness": 1.0, "page_size": 64, "swa_full_tokens_ratio": 0.8, "disable_hybrid_swa_memory": false, "radix_eviction_policy": "lru", "prefill_only_disable_kv_cache": false, "disable_radix_cache": true, "enable_page_major_kv_layout": false, "enable_unified_memory": false, "disable_chunked_prefix_cache": true, "disable_overlap_schedule": false, "num_continuous_decode_steps": 1, "scheduler_recv_interval": 1, "enable_mixed_chunk": false, "nccl_port": null, "dist_timeout": 3600, "dist_init_addr": "174.1.60.5:20000", "nnodes": 4, "node_rank": 0, "tp_size": 32, "dcp_size": 1, "pp_size": 1, "pp_max_micro_batch_size": 181, "pp_async_batch_depth": 0, "dp_size": 1, "load_balance_method": "round_robin", "attn_cp_size": 1, "moe_dp_size": 1, "dcp_comm_backend": "ag_rs", "dcp_replicate_q_proj": null, "dwdp_size": 1, "enable_prefill_cp": false, "cp_strategy": null, "enable_dsa_cache_layer_split": false, "enable_dsa_prefill_context_parallel": false, "dsa_prefill_cp_mode": "round-robin-split", "enable_prefill_context_parallel": false, "prefill_cp_mode": "in-seq-split", "enable_cp_decode_attn_tp": false, "enable_dp_attention": false, "enable_dp_attention_local_control_broadcast": false, "enable_dp_lm_head": false, "enable_attn_tp_input_scattered": false, "disable_attn_tp_gather": false, "enable_p2p_check": false, "device": "cuda", "base_gpu_id": 0, "gpu_id_step": 1, "random_seed": 369235288, "watchdog_timeout": 300, "soft_watchdog_timeout": null, "sleep_on_idle": false, "use_ray": false, "numa_node": null, "gc_threshold": null, "host": "0.0.0.0", "port": 30000, "fastapi_root_path": "", "smg_grpc_mode": false, "grpc_mode": false, "grpc_port": null, "sidecar": null, "sidecar_args": null, "skip_server_warmup": false, "warmups": null, "enable_http2": false, "ssl_keyfile": null, "ssl_certfile": null, "ssl_ca_certs": null, "ssl_keyfile_password": null, "enable_ssl_refresh": false, "api_key": null, "admin_api_key": null, "served_model_name": "kimi-k3", "weight_version": "default", "chat_template": null, "hf_chat_template_name": null, "completion_template": null, "file_storage_path": "sglang_storage", "enable_cache_report": false, "reasoning_parser": null, "default_chat_template_kwargs": null, "strip_thinking_cache": false, "enable_strict_thinking": false, "tool_call_parser": null, "tool_server": null, "sampling_defaults": "model", "asr_max_buffer_seconds": 60, "asr_max_concurrent_sessions": 32, "preferred_sampling_params": null, "allow_auto_truncate": false, "stream_interval": 1, "batch_notify_size": 16, "stream_response_default_include_usage": false, "incremental_streaming_output": false, "enable_streaming_session": false, "enable_session_radix_cache": false, "log_level": "info", "log_level_http": null, "log_requests": false, "log_requests_level": 2, "log_requests_format": "text", "log_requests_target": null, "uvicorn_access_log_exclude_prefixes": [], "crash_dump_folder": null, "show_time_cost": false, "enable_metrics": false, "smg_http_sidecar_port": null, "enable_mfu_metrics": false, "enable_metrics_for_all_schedulers": false, "load_snapshot_publish_interval": 15, "tokenizer_metrics_custom_labels_header": "x-custom-labels", "tokenizer_metrics_allowed_custom_labels": null, "extra_metric_labels": null, "bucket_time_to_first_token": null, "bucket_inter_token_latency": null, "bucket_e2e_request_latency": null, "prompt_tokens_buckets": null, "generation_tokens_buckets": null, "gc_warning_threshold_secs": 0.0, "decode_log_interval": 40, "enable_request_time_stats_logging": false, "kv_events_config": null, "enable_forward_pass_metrics": false, "forward_pass_metrics_worker_id": "", "forward_pass_metrics_ipc_name": null, "enable_trace": false, "trace_modules": "request", "otlp_traces_endpoint": "localhost:4317", "export_metrics_to_file": false, "export_metrics_to_file_dir": null, "stat_loggers": null, "constrained_json_whitespace_pattern": null, "constrained_json_disable_any_whitespace": false, "attention_backend": "triton", "decode_attention_backend": null, "prefill_attention_backend": null, "sampling_backend": "flashinfer", "grammar_backend": "xgrammar", "radix_cache_backend": null, "mm_attention_backend": null, "fp8_gemm_runner_backend": "auto", "fp4_gemm_runner_backend": "auto", "bf16_gemm_backend": "auto", "dsa_prefill_backend": null, "dsa_decode_backend": null, "dsa_paged_mqa_logits_backend": "auto", "dsa_topk_backend": "sgl-kernel", "disable_flashinfer_autotune": false, "flashinfer_autotune_skip_ops": null, "mamba_backend": "triton", "cuda_graph_config": {"decode": {"backend": "full", "max_bs": 16, "bs": [1, 2, 4, 8, 12, 16], "tc_compiler": "eager", "full_prefill_max_req": null}, "prefill": {"backend": "disabled", "max_bs": 2048, "bs": [4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], "tc_compiler": "eager", "full_prefill_max_req": null}}, "cuda_graph_backend_decode": null, "cuda_graph_backend_prefill": null, "cuda_graph_max_bs_decode": 16, "cuda_graph_max_bs_prefill": null, "cuda_graph_bs_decode": null, "cuda_graph_bs_prefill": null, "cuda_graph_tc_compiler": null, "disable_prefill_cuda_graph": false, "disable_decode_cuda_graph": false, "disable_cuda_graph": false, "disable_cuda_graph_padding": false, "enable_profile_cuda_graph": false, "enable_cudagraph_gc": false, "debug_cuda_graph": false, "enable_layerwise_nvtx_marker": false, "enable_nccl_nvls": false, "enable_symm_mem": false, "triton_attention_reduce_in_fp32": false, "triton_attention_num_kv_splits": 8, "triton_attention_split_tile_size": null, "flashinfer_mla_disable_ragged": false, "enable_fused_qk_norm_rope": false, "enable_precise_embedding_interpolation": false, "enable_fused_moe_sum_all_reduce": false, "enable_deepseek_v4_fp4_indexer": false, "disable_custom_all_reduce": false, "enable_mscclpp": false, "enable_torch_symm_mem": false, "enable_scattered_sconv": false, "pre_warm_nccl": false, "enable_quant_communications": false, "enable_flashinfer_allreduce_fusion": false, "enforce_disable_flashinfer_allreduce_fusion": false, "flashinfer_allreduce_fusion_backend": null, "enable_aiter_allreduce_fusion": false, "enable_torch_compile": false, "enable_torch_compile_debug_mode": false, "torch_compile_max_bs": 32, "torchao_config": "", "speculative_algorithm": null, "speculative_draft_model_path": null, "speculative_draft_model_revision": null, "speculative_draft_load_format": null, "speculative_num_steps": null, "speculative_eagle_topk": null, "speculative_num_draft_tokens": null, "speculative_dflash_block_size": null, "speculative_dspark_block_size": null, "speculative_dspark_sps_table_path": null, "speculative_dspark_confidence_sts_path": null, "speculative_dspark_align_verify_tokens_to_graph_tier": false, "speculative_accept_threshold_single": 1.0, "speculative_accept_threshold_acc": 1.0, "speculative_use_rejection_sampling": false, "speculative_token_map": null, "speculative_attention_mode": "prefill", "speculative_draft_attention_backend": null, "speculative_draft_window_size": null, "speculative_moe_runner_backend": "marlin", "speculative_moe_a2a_backend": null, "speculative_draft_model_quantization": null, "speculative_skip_dp_mlp_sync": false, "enable_multi_layer_eagle": false, "speculative_adaptive": false, "speculative_adaptive_config": null, "decoupled_spec_bind_endpoint": null, "decoupled_spec_connect_endpoints": null, "decoupled_spec_rank": null, "decoupled_spec_role": "null", "spec_trace_dir": null, "speculative_ngram_min_bfs_breadth": 1, "speculative_ngram_max_bfs_breadth": 10, "speculative_ngram_match_type": "BFS", "speculative_ngram_max_trie_depth": 18, "speculative_ngram_capacity": 10000000, "speculative_ngram_external_corpus_path": null, "speculative_ngram_external_sam_budget": 0, "speculative_ngram_external_corpus_max_tokens": 10000000, "ep_size": 32, "moe_a2a_backend": "none", "moe_runner_backend": "marlin", "flashinfer_mxfp4_moe_precision": "default", "deepep_mode": "auto", "fuseep_mode": 2, "deepep_dispatcher_output_dtype": "auto", "ep_num_redundant_experts": 0, "ep_dispatch_algorithm": null, "init_expert_location": "trivial", "enable_eplb": false, "eplb_algorithm": "auto", "eplb_rebalance_num_iterations": 1000, "eplb_rebalance_layers_per_chunk": null, "eplb_min_rebalancing_utilization_threshold": 1.0, "expert_distribution_recorder_mode": null, "expert_distribution_recorder_buffer_size": 1000, "enable_expert_distribution_metrics": false, "deepep_config": null, "moe_dense_tp_size": null, "elastic_ep_backend": null, "enable_elastic_expert_backup": false, "mooncake_ib_device": null, "enable_waterfill": false, "ep_join_mode": null, "ep_join_rank_offset": 0, "elastic_ep_initial_size": null, "max_ep_size": null, "elastic_ep_scale_timeout": 600, "elastic_ep_rejoin": false, "disable_flashinfer_cutlass_moe_fp4_allgather": false, "disable_shared_experts_fusion": false, "enforce_shared_experts_fusion": false, "max_mamba_cache_size": 181, "mamba_ssm_dtype": null, "mamba_max_states_per_path": -1, "enable_mamba_cache_stochastic_rounding": false, "mamba_cache_philox_rounds": 0, "mamba_full_memory_ratio": 0.21, "mamba_radix_cache_strategy": "extra_buffer_lazy", "uses_mamba_radix_cache": false, "mamba_track_interval": 256, "enable_int8_mamba_checkpoint": false, "int8_mamba_ckpt_size": null, "linear_attn_backend": "triton", "linear_attn_decode_backend": null, "linear_attn_prefill_backend": null, "linear_attn_verify_backend": null, "enable_linear_replayssm": false, "linear_replayssm_cache_len": 16, "enable_linear_replayssm_spec": false, "enable_hierarchical_cache": false, "hicache_ratio": 2.0, "hicache_size": 0, "hicache_write_policy": "write_through", "hicache_io_backend": "kernel", "hicache_mem_layout": "page_first", "hicache_storage_backend": null, "hicache_storage_prefetch_policy": "timeout", "hicache_storage_backend_extra_config": null, "enable_hisparse": false, "hisparse_config": null, "enable_broadcast_mm_inputs_process": false, "enable_prefix_mm_cache": false, "mm_enable_dp_encoder": false, "mm_process_config": {}, "mm_processor_worker_num": 0, "mm_io_worker_num": 0, "limit_mm_data_per_request": null, "enable_mm_global_cache": false, "disable_fast_image_processor": false, "mm_feature_transport": "cpu", "keep_mm_feature_on_device": false, "enable_lora": null, "enable_lora_overlap_loading": null, "max_lora_rank": null, "lora_target_modules": null, "lora_paths": null, "max_loaded_loras": null, "max_loras_per_batch": 8, "lora_eviction_policy": "lru", "lora_backend": "csgmv", "max_lora_chunk_size": 16, "experts_shared_outer_loras": null, "lora_use_virtual_experts": false, "lora_strict_loading": false, "lora_drain_wait_threshold": 0.0, "enable_two_batch_overlap": false, "enable_single_batch_overlap": false, "tbo_token_distribution_threshold": 0.48, "cpu_offload_gb": 0, "offload_group_size": -1, "offload_num_in_group": 1, "offload_prefetch_step": 1, "offload_mode": "cpu", "enable_lmcache": false, "lmcache_config_file": null, "enable_flexkv": false, "flexkv_config_file": null, "kt_weight_path": null, "kt_method": "AMXINT4", "kt_cpuinfer": null, "kt_threadpool_count": 2, "kt_num_gpu_experts": null, "kt_max_deferred_experts_per_token": null, "dllm_algorithm": null, "dllm_algorithm_config": null, "dllm_fdfo": true, "disaggregation_mode": "decode", "disaggregation_transfer_backend": "mooncake", "disaggregation_bootstrap_port": 28800, "disaggregation_ib_device": "mlx5_0,mlx5_1,mlx5_2,mlx5_3", "disaggregation_decode_enable_radix_cache": false, "disaggregation_decode_enable_offload_kvcache": false, "num_reserved_decode_tokens": 512, "disaggregation_decode_extra_slots": 0, "disaggregation_decode_polling_interval": 1, "optimistic_prefill_attempts": 0, "encoder_only": false, "language_only": false, "encoder_transfer_backend": "zmq_to_tokenizer", "encoder_urls": [], "encoder_bootstrap_port": 8997, "encoder_register_urls": [], "enable_adaptive_dispatch_to_encoder": false, "enable_pdmux": false, "pdmux_config_path": null, "sm_group_num": 8, "custom_weight_loader": [], "weight_loader_disable_mmap": false, "weight_loader_prefetch_checkpoints": false, "weight_loader_prefetch_num_threads": 4, "weight_loader_drop_cache_after_load": false, "remote_instance_weight_loader_seed_instance_ip": null, "remote_instance_weight_loader_seed_instance_service_port": null, "remote_instance_weight_loader_send_weights_group_ports": null, "remote_instance_weight_loader_backend": "nccl", "remote_instance_weight_loader_start_seed_via_transfer_engine": false, "engine_info_bootstrap_port": 6789, "modelexpress_config": null, "download_dir": null, "model_checksum": null, "delete_ckpt_after_loading": false, "decrypted_config_file": null, "decrypted_draft_config_file": null, "checkpoint_engine_wait_weights_before_ready": false, "enable_prefill_delayer": false, "prefill_delayer_max_delay_passes": 30, "prefill_delayer_token_usage_low_watermark": null, "prefill_delayer_forward_passes_buckets": null, "prefill_delayer_wait_seconds_buckets": null, "prefill_delayer_queue_min_ratio": null, "prefill_delayer_max_delay_ms": null, "min_free_slots_delay": null, "enable_deterministic_inference": false, "rl_on_policy_target": null, "kv_canary": "none", "kv_canary_real_data": "none", "kv_canary_sweep_interval": 0, "enable_dynamic_batch_tokenizer": false, "dynamic_batch_tokenizer_batch_size": 32, "dynamic_batch_tokenizer_batch_timeout": 0.002, "enable_tokenizer_batch_encode": false, "disable_tokenizer_batch_decode": false, "debug_tensor_dump_output_folder": null, "debug_tensor_dump_layers": null, "debug_tensor_dump_input_file": null, "enable_memory_saver": false, "enable_weights_cpu_backup": false, "enable_draft_weights_cpu_backup": false, "enable_custom_logit_processor": false, "enable_return_hidden_states": false, "enable_return_routed_experts": false, "enable_return_indexer_topk": false, "disable_outlines_disk_cache": false, "enable_mis": false, "weight_cache_mode": "off", "weight_cache_socket": null, "weight_cache_timeout": 1800, "forward_hooks": null, "msprobe_dump_config": null, "_resolved_overrides": [["_sampling_backend_default", {"sampling_backend": "flashinfer"}], ["_attention_backend_default", {"attention_backend": "triton"}], ["_data_parallelism_defaults", {"enable_dp_attention": false, "enable_dp_lm_head": false}], ["_speculative_moe_runner_default", {"speculative_moe_runner_backend": "marlin"}]], "grpc_worker_threads": 4, "_quantization_explicitly_unset": false, "_cuda_graph_config_locked": [["decode", "max_bs"]], "_declarations_materialized": true, "_runtime_mutations": [["model_runner.chunked_prefix_cache_gate", {"disable_chunked_prefix_cache": true}], ["mamba_pool.memory_budget", {"max_mamba_cache_size": 181}]], "_in_override": false, "_mx_config_cache": {}, "max_speculative_num_draft_tokens": null, "last_gen_throughput": 0.09006957280512767, "memory_usage": {"weight": 58.57, "kvcache": 11.34, "token_capacity": 440256, "graph": 0.6}, "effective_max_running_requests_per_dp": 181}], "version": "0.5.16", "kv_events": null}]}, "duration": 1.517340388149023, "completed": 1, "total_input_tokens": 64, "total_input_text_tokens": 64, "total_input_vision_tokens": 0, "total_output_tokens": 8, "total_output_tokens_retokenized": 7, "request_throughput": 0.6590479023760004, "input_throughput": 42.17906575206403, "output_throughput": 5.272383219008003, "total_throughput": 47.45144897107203, "mean_e2e_latency_ms": 1449.7034070082009, "median_e2e_latency_ms": 1449.7034070082009, "std_e2e_latency_ms": 0.0, "p90_e2e_latency_ms": 1449.7034070082009, "p95_e2e_latency_ms": 1449.7034070082009, "p99_e2e_latency_ms": 1449.7034070082009, "mean_ttft_ms": 1216.788083082065, "median_ttft_ms": 1216.788083082065, "std_ttft_ms": 0.0, "p90_ttft_ms": 1216.788083082065, "p95_ttft_ms": 1216.788083082065, "p99_ttft_ms": 1216.788083082065, "mean_tpot_ms": 33.2736177037337, "median_tpot_ms": 33.2736177037337, "std_tpot_ms": 0.0, "p90_tpot_ms": 33.2736177037337, "p95_tpot_ms": 33.2736177037337, "p99_tpot_ms": 33.2736177037337, "mean_itl_ms": 38.8128844788298, "median_itl_ms": 38.78513048402965, "std_itl_ms": 0.26855067899252305, "p90_itl_ms": 39.12828944157809, "p95_itl_ms": 39.21070724027231, "p99_itl_ms": 39.27664147922769, "concurrency": 0.9554239894560961, "accept_length": null, "max_output_tokens_per_s": 7.0, "max_concurrent_requests": 1, "input_lens": [64], "output_lens": [8], "ttfts": [1.216788083082065], "itls": [[0.03885094984434545, 0.03871931112371385, 0.03856838191859424, 0.038482085103169084, 0.038963453844189644, 0.039293125038966537]], "generated_texts": ["\ufffd\n+\t\t// \t"], "errors": [""]}
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/bench/smoke_64_8_c1.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/bench/smoke_64_8_c1.log
new file mode 100644
index 0000000..11831ee
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/bench/smoke_64_8_c1.log
@@ -0,0 +1,71 @@
+/usr/local/lib/python3.12/dist-packages/torchao/quantization/quant_api.py:1731: SyntaxWarning: invalid escape sequence '\.'
+  """Configuration class for applying different quantization configs to modules or parameters based on their fully qualified names (FQNs).
+/sgl-workspace/sglang/python/sglang/kernels/ops/attention/fla/utils.py:223: UserWarning: Triton is not supported on current platform, roll back to CPU.
+  warnings.warn(
+/sgl-workspace/sglang/python/sglang/srt/layers/quantization/awq/awq.py:52: UserWarning: Only CUDA, HIP and XPU support AWQ currently.
+  warnings.warn(f"Only CUDA, HIP and XPU support AWQ currently.")
+/sgl-workspace/sglang/python/sglang/srt/layers/quantization/gguf.py:69: UserWarning: Only CUDA, MUSA and NPU support GGUF quantization currently.
+  warnings.warn(f"Only CUDA, MUSA and NPU support GGUF quantization currently.")
+benchmark_args=Namespace(backend='sglang', base_url=None, host='174.1.60.5', port=31000, ready_check_timeout_sec=60, dataset_name='random-ids', dataset_path='', dataset_offset=0, agentic_max_turns=None, speed_bench_category=None, speed_bench_output_len=512, model=None, served_model_name=None, tokenizer='/data/hf_models/Kimi-K3', num_prompts=1, sharegpt_output_len=None, sharegpt_context_len=None, random_input_len=64, random_output_len=8, random_range_ratio=1.0, image_count=1, image_resolution='1080p', random_image_count=False, image_format='jpeg', image_content='random', request_rate=10000.0, use_trace_timestamps=False, max_concurrency=1, output_file='/data/hzy/sskj/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/bench/smoke_64_8_c1.jsonl', output_details=True, print_requests=False, disable_tqdm=True, disable_stream=False, return_logprob=False, top_logprobs_num=0, token_ids_logprob=None, logprob_start_len=-1, return_routed_experts=False, cache_report=False, seed=42, disable_ignore_eos=False, temperature=0.0, top_p=1.0, extra_request_body=None, apply_chat_template=False, profile=False, plot_throughput=False, profile_activities=['CPU', 'GPU'], profile_start_step=None, profile_steps=None, profile_num_steps=None, profile_by_stage=False, profile_stages=None, profile_output_dir=None, profile_prefix=None, lora_name=None, lora_request_distribution='uniform', lora_zipf_alpha=1.5, prompt_suffix='', pd_separated=False, profile_prefill_url=None, profile_decode_url=None, flush_cache=False, warmup_requests=0, tokenize_prompt=False, gsp_num_groups=64, gsp_prompts_per_group=16, gsp_system_prompt_len=2048, gsp_question_len=128, gsp_output_len=256, gsp_range_ratio=1.0, gsp_fast_prepare=False, gsp_send_routing_key=False, gsp_num_turns=1, gsp_ordered=False, gsp_group_distribution='uniform', gsp_zipf_alpha=None, mooncake_slowdown_factor=1.0, mooncake_num_rounds=1, mooncake_workload='conversation', fake_prefill=False, tag=None, header=None)
+Waiting up to 60s for http://174.1.60.5:31000/v1/models to become ready...
+Server ready in 0.0s.
+'[Errno 101] Network is unreachable' thrown while requesting HEAD https://huggingface.co/kimi-k3/resolve/main/config.json
+Retrying in 1s [Retry 1/5].
+'[Errno 101] Network is unreachable' thrown while requesting HEAD https://huggingface.co/kimi-k3/resolve/main/config.json
+Retrying in 1s [Retry 1/5].
+Fail to load tokenizer config with error=Can't load the configuration of 'kimi-k3'. If you were trying to load it from 'https://huggingface.co/models', make sure you don't have a local directory with the same name. Otherwise, make sure 'kimi-k3' is the correct path to a directory containing a config.json file
+
+WARNING It is recommended to use the `Chat` or `Instruct` model for benchmarking.
+Because when the tokenizer counts the output tokens, if there is gibberish, it might count incorrectly.
+
+Namespace(backend='sglang', base_url=None, host='174.1.60.5', port=31000, ready_check_timeout_sec=60, dataset_name='random-ids', dataset_path='', dataset_offset=0, agentic_max_turns=None, speed_bench_category=None, speed_bench_output_len=512, model='kimi-k3', served_model_name=None, tokenizer='/data/hf_models/Kimi-K3', num_prompts=1, sharegpt_output_len=None, sharegpt_context_len=None, random_input_len=64, random_output_len=8, random_range_ratio=1.0, image_count=1, image_resolution='1080p', random_image_count=False, image_format='jpeg', image_content='random', request_rate=10000.0, use_trace_timestamps=False, max_concurrency=1, output_file='/data/hzy/sskj/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/bench/smoke_64_8_c1.jsonl', output_details=True, print_requests=False, disable_tqdm=True, disable_stream=False, return_logprob=False, top_logprobs_num=0, token_ids_logprob=None, logprob_start_len=-1, return_routed_experts=False, cache_report=False, seed=42, disable_ignore_eos=False, temperature=0.0, top_p=1.0, extra_request_body=None, apply_chat_template=False, profile=False, plot_throughput=False, profile_activities=['CPU', 'GPU'], profile_start_step=None, profile_steps=None, profile_num_steps=None, profile_by_stage=False, profile_stages=None, profile_output_dir=None, profile_prefix=None, lora_name=None, lora_request_distribution='uniform', lora_zipf_alpha=1.5, prompt_suffix='', pd_separated=False, profile_prefill_url=None, profile_decode_url=None, flush_cache=False, warmup_requests=0, tokenize_prompt=False, gsp_num_groups=64, gsp_prompts_per_group=16, gsp_system_prompt_len=2048, gsp_question_len=128, gsp_output_len=256, gsp_range_ratio=1.0, gsp_fast_prepare=False, gsp_send_routing_key=False, gsp_num_turns=1, gsp_ordered=False, gsp_group_distribution='uniform', gsp_zipf_alpha=None, mooncake_slowdown_factor=1.0, mooncake_num_rounds=1, mooncake_workload='conversation', fake_prefill=False, tag=None, header=None)
+
+#Input tokens: 64
+#Output tokens: 8
+Starting warmup with 0 sequences...
+Warmup completed with 0 sequences. Starting main benchmark run...
+Calling super().encode with {'add_special_tokens': False}
+
+============ Serving Benchmark Result ============
+Backend:                                 sglang    
+Traffic request rate:                    10000.0   
+Max request concurrency:                 1         
+Successful requests:                     1         
+Benchmark duration (s):                  1.52      
+Total input tokens:                      64        
+Total input text tokens:                 64        
+Total generated tokens:                  8         
+Total generated tokens (retokenized):    7         
+Request throughput (req/s):              0.66      
+Input token throughput (tok/s):          42.18     
+Output token throughput (tok/s):         5.27      
+Peak output token throughput (tok/s):    7.00      
+Peak concurrent requests:                1         
+Total token throughput (tok/s):          47.45     
+Concurrency:                             0.96      
+----------------End-to-End Latency----------------
+Mean E2E Latency (ms):                   1449.70   
+Median E2E Latency (ms):                 1449.70   
+P90 E2E Latency (ms):                    1449.70   
+P95 E2E Latency (ms):                    1449.70   
+P99 E2E Latency (ms):                    1449.70   
+---------------Time to First Token----------------
+Mean TTFT (ms):                          1216.79   
+Median TTFT (ms):                        1216.79   
+P90 TTFT (ms):                           1216.79   
+P95 TTFT (ms):                           1216.79   
+P99 TTFT (ms):                           1216.79   
+-----Time per Output Token (excl. 1st token)------
+Mean TPOT (ms):                          33.27     
+Median TPOT (ms):                        33.27     
+P90 TPOT (ms):                           33.27     
+P95 TPOT (ms):                           33.27     
+P99 TPOT (ms):                           33.27     
+---------------Inter-Token Latency----------------
+Mean ITL (ms):                           38.81     
+Median ITL (ms):                         38.79     
+P90 ITL (ms):                            39.13     
+P95 ITL (ms):                            39.21     
+P99 ITL (ms):                            39.28     
+Max ITL (ms):                            39.29     
+==================================================
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/logs/manual_d_node0.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/logs/manual_d_node0.log
new file mode 100644
index 0000000..3a4204b
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/logs/manual_d_node0.log
@@ -0,0 +1,1625 @@
+/usr/local/lib/python3.12/dist-packages/torchao/quantization/quant_api.py:1731: SyntaxWarning: invalid escape sequence '\.'
+  """Configuration class for applying different quantization configs to modules or parameters based on their fully qualified names (FQNs).
+/sgl-workspace/sglang/python/sglang/launch_server.py:56: UserWarning: 'python -m sglang.launch_server' is still supported, but 'sglang serve' is the recommended entrypoint.
+  Example: sglang serve --model-path <model> [options]
+  warnings.warn(
+[2026-08-27 04:36:08] KV cache is forced as chunk cache for decode server
+[2026-08-27 04:36:11] Breakable CUDA graph is incompatible with MLA attention; disabling prefill CUDA graph.
+[2026-08-27 04:36:12] Attention backend not specified. Use triton backend by default.
+[2026-08-27 04:36:14] server_args=ServerArgs(model_path='/data/hf_models/Kimi-K3', tokenizer_path='/data/hf_models/Kimi-K3', tokenizer_mode='auto', tokenizer_backend='huggingface', tokenizer_worker_num=1, detokenizer_worker_num=1, skip_tokenizer_init=False, load_format='auto', model_loader_extra_config='{}', trust_remote_code=True, context_length=None, is_embedding=False, enable_multimodal=None, revision=None, model_impl='auto', model_config_parser='auto', json_model_override_args='{}', dtype='auto', quantization=None, quantization_param_path=None, kv_cache_dtype='auto', enable_fp32_lm_head=False, modelopt_quant=None, modelopt_checkpoint_restore_path=None, modelopt_checkpoint_save_path=None, modelopt_export_path=None, quantize_and_serve=False, rl_quant_profile=None, enable_tf32_matmul=False, mem_fraction_static=0.88, max_running_requests=None, max_queued_requests=None, max_total_tokens=None, chunked_prefill_size=8192, enable_dynamic_chunking=False, max_prefill_tokens=16384, prefill_max_requests=None, schedule_policy='fcfs', enable_priority_scheduling=False, disable_priority_preemption=False, default_priority_value=None, abort_on_priority_when_disabled=False, schedule_low_priority_values_first=False, priority_scheduling_preemption_threshold=10, retraction_policy='length', schedule_conservativeness=1.0, page_size=64, swa_full_tokens_ratio=0.8, disable_hybrid_swa_memory=False, radix_eviction_policy='lru', prefill_only_disable_kv_cache=False, disable_radix_cache=True, enable_page_major_kv_layout=False, enable_unified_memory=False, disable_chunked_prefix_cache=False, disable_overlap_schedule=False, num_continuous_decode_steps=1, scheduler_recv_interval=1, enable_mixed_chunk=False, nccl_port=None, dist_timeout=3600, dist_init_addr='174.1.60.5:20000', nnodes=4, node_rank=0, tp_size=32, dcp_size=1, pp_size=1, pp_max_micro_batch_size=None, pp_async_batch_depth=0, dp_size=1, load_balance_method='round_robin', attn_cp_size=1, moe_dp_size=1, dcp_comm_backend='ag_rs', dcp_replicate_q_proj=None, dwdp_size=1, enable_prefill_cp=False, cp_strategy=None, enable_dsa_cache_layer_split=False, enable_dsa_prefill_context_parallel=False, dsa_prefill_cp_mode='round-robin-split', enable_prefill_context_parallel=False, prefill_cp_mode='in-seq-split', enable_cp_decode_attn_tp=False, enable_dp_attention=False, enable_dp_attention_local_control_broadcast=False, enable_dp_lm_head=False, enable_attn_tp_input_scattered=False, disable_attn_tp_gather=False, enable_p2p_check=False, device='cuda', base_gpu_id=0, gpu_id_step=1, random_seed=369235288, watchdog_timeout=300, soft_watchdog_timeout=None, sleep_on_idle=False, use_ray=False, custom_sigquit_handler=None, numa_node=None, gc_threshold=None, host='0.0.0.0', port=30000, fastapi_root_path='', smg_grpc_mode=False, grpc_mode=False, grpc_port=None, sidecar=None, sidecar_args=None, skip_server_warmup=False, warmups=None, enable_http2=False, ssl_keyfile=None, ssl_certfile=None, ssl_ca_certs=None, ssl_keyfile_password=None, enable_ssl_refresh=False, api_key=None, admin_api_key=None, served_model_name='kimi-k3', weight_version='default', chat_template=None, hf_chat_template_name=None, completion_template=None, file_storage_path='sglang_storage', enable_cache_report=False, reasoning_parser=None, default_chat_template_kwargs=None, strip_thinking_cache=False, enable_strict_thinking=False, tool_call_parser=None, tool_server=None, sampling_defaults='model', asr_max_buffer_seconds=60, asr_max_concurrent_sessions=32, preferred_sampling_params=None, allow_auto_truncate=False, stream_interval=1, batch_notify_size=16, stream_response_default_include_usage=False, incremental_streaming_output=False, enable_streaming_session=False, enable_session_radix_cache=False, log_level='info', log_level_http=None, log_requests=False, log_requests_level=2, log_requests_format='text', log_requests_target=None, uvicorn_access_log_exclude_prefixes=[], crash_dump_folder=None, show_time_cost=False, enable_metrics=False, smg_http_sidecar_port=None, enable_mfu_metrics=False, enable_metrics_for_all_schedulers=False, load_snapshot_publish_interval=15, tokenizer_metrics_custom_labels_header='x-custom-labels', tokenizer_metrics_allowed_custom_labels=None, extra_metric_labels=None, bucket_time_to_first_token=None, bucket_inter_token_latency=None, bucket_e2e_request_latency=None, prompt_tokens_buckets=None, generation_tokens_buckets=None, gc_warning_threshold_secs=0.0, decode_log_interval=40, enable_request_time_stats_logging=False, kv_events_config=None, enable_forward_pass_metrics=False, forward_pass_metrics_worker_id='', forward_pass_metrics_ipc_name=None, enable_trace=False, trace_modules='request', otlp_traces_endpoint='localhost:4317', export_metrics_to_file=False, export_metrics_to_file_dir=None, stat_loggers=None, constrained_json_whitespace_pattern=None, constrained_json_disable_any_whitespace=False, attention_backend='triton', decode_attention_backend=None, prefill_attention_backend=None, sampling_backend='flashinfer', grammar_backend='xgrammar', radix_cache_backend=None, mm_attention_backend=None, fp8_gemm_runner_backend='auto', fp4_gemm_runner_backend='auto', bf16_gemm_backend='auto', dsa_prefill_backend=None, dsa_decode_backend=None, dsa_paged_mqa_logits_backend='auto', dsa_topk_backend='sgl-kernel', disable_flashinfer_autotune=False, flashinfer_autotune_skip_ops=None, mamba_backend='triton', cuda_graph_config=CudaGraphConfig(decode=PhaseConfig(backend='full', max_bs=16, bs=[1, 2, 4, 8, 12, 16], tc_compiler='eager', full_prefill_max_req=None), prefill=PhaseConfig(backend='disabled', max_bs=2048, bs=[4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], tc_compiler='eager', full_prefill_max_req=None)), cuda_graph_backend_decode=None, cuda_graph_backend_prefill=None, cuda_graph_max_bs_decode=16, cuda_graph_max_bs_prefill=None, cuda_graph_bs_decode=None, cuda_graph_bs_prefill=None, cuda_graph_tc_compiler=None, disable_prefill_cuda_graph=False, disable_decode_cuda_graph=False, disable_cuda_graph=False, disable_cuda_graph_padding=False, enable_profile_cuda_graph=False, enable_cudagraph_gc=False, debug_cuda_graph=False, enable_layerwise_nvtx_marker=False, enable_nccl_nvls=False, enable_symm_mem=False, triton_attention_reduce_in_fp32=False, triton_attention_num_kv_splits=8, triton_attention_split_tile_size=None, flashinfer_mla_disable_ragged=False, enable_fused_qk_norm_rope=False, enable_precise_embedding_interpolation=False, enable_fused_moe_sum_all_reduce=False, enable_deepseek_v4_fp4_indexer=False, disable_custom_all_reduce=False, enable_mscclpp=False, enable_torch_symm_mem=False, enable_scattered_sconv=False, pre_warm_nccl=False, enable_quant_communications=False, enable_flashinfer_allreduce_fusion=False, enforce_disable_flashinfer_allreduce_fusion=False, flashinfer_allreduce_fusion_backend=None, enable_aiter_allreduce_fusion=False, enable_torch_compile=False, enable_torch_compile_debug_mode=False, torch_compile_max_bs=32, torchao_config='', speculative_algorithm=None, speculative_draft_model_path=None, speculative_draft_model_revision=None, speculative_draft_load_format=None, speculative_num_steps=None, speculative_eagle_topk=None, speculative_num_draft_tokens=None, speculative_dflash_block_size=None, speculative_dspark_block_size=None, speculative_dspark_sps_table_path=None, speculative_dspark_confidence_sts_path=None, speculative_dspark_align_verify_tokens_to_graph_tier=False, speculative_accept_threshold_single=1.0, speculative_accept_threshold_acc=1.0, speculative_use_rejection_sampling=False, speculative_token_map=None, speculative_attention_mode='prefill', speculative_draft_attention_backend=None, speculative_draft_window_size=None, speculative_moe_runner_backend='marlin', speculative_moe_a2a_backend=None, speculative_draft_model_quantization=None, speculative_skip_dp_mlp_sync=False, enable_multi_layer_eagle=False, speculative_adaptive=False, speculative_adaptive_config=None, decoupled_spec_bind_endpoint=None, decoupled_spec_connect_endpoints=None, decoupled_spec_rank=None, decoupled_spec_role='null', spec_trace_dir=None, speculative_ngram_min_bfs_breadth=1, speculative_ngram_max_bfs_breadth=10, speculative_ngram_match_type='BFS', speculative_ngram_max_trie_depth=18, speculative_ngram_capacity=10000000, speculative_ngram_external_corpus_path=None, speculative_ngram_external_sam_budget=0, speculative_ngram_external_corpus_max_tokens=10000000, ep_size=32, moe_a2a_backend='none', moe_runner_backend='marlin', flashinfer_mxfp4_moe_precision='default', deepep_mode='auto', fuseep_mode=2, deepep_dispatcher_output_dtype='auto', ep_num_redundant_experts=0, ep_dispatch_algorithm=None, init_expert_location='trivial', enable_eplb=False, eplb_algorithm='auto', eplb_rebalance_num_iterations=1000, eplb_rebalance_layers_per_chunk=None, eplb_min_rebalancing_utilization_threshold=1.0, expert_distribution_recorder_mode=None, expert_distribution_recorder_buffer_size=1000, enable_expert_distribution_metrics=False, deepep_config=None, moe_dense_tp_size=None, elastic_ep_backend=None, enable_elastic_expert_backup=False, mooncake_ib_device=None, enable_waterfill=False, ep_join_mode=None, ep_join_rank_offset=0, elastic_ep_initial_size=None, max_ep_size=None, elastic_ep_scale_timeout=600, elastic_ep_rejoin=False, disable_flashinfer_cutlass_moe_fp4_allgather=False, disable_shared_experts_fusion=False, enforce_shared_experts_fusion=False, max_mamba_cache_size=None, mamba_ssm_dtype=None, mamba_max_states_per_path=-1, enable_mamba_cache_stochastic_rounding=False, mamba_cache_philox_rounds=0, mamba_full_memory_ratio=0.21, mamba_radix_cache_strategy='extra_buffer_lazy', uses_mamba_radix_cache=False, mamba_track_interval=256, enable_int8_mamba_checkpoint=False, int8_mamba_ckpt_size=None, linear_attn_backend='triton', linear_attn_decode_backend=None, linear_attn_prefill_backend=None, linear_attn_verify_backend=None, enable_linear_replayssm=False, linear_replayssm_cache_len=16, enable_linear_replayssm_spec=False, enable_hierarchical_cache=False, hicache_ratio=2.0, hicache_size=0, hicache_write_policy='write_through', hicache_io_backend='kernel', hicache_mem_layout='page_first', hicache_storage_backend=None, hicache_storage_prefetch_policy='timeout', hicache_storage_backend_extra_config=None, enable_hisparse=False, hisparse_config=None, enable_broadcast_mm_inputs_process=False, enable_prefix_mm_cache=False, mm_enable_dp_encoder=False, mm_process_config={}, mm_processor_worker_num=0, mm_io_worker_num=0, limit_mm_data_per_request=None, enable_mm_global_cache=False, disable_fast_image_processor=False, mm_feature_transport='cpu', keep_mm_feature_on_device=False, enable_lora=None, enable_lora_overlap_loading=None, max_lora_rank=None, lora_target_modules=None, lora_paths=None, max_loaded_loras=None, max_loras_per_batch=8, lora_eviction_policy='lru', lora_backend='csgmv', max_lora_chunk_size=16, experts_shared_outer_loras=None, lora_use_virtual_experts=False, lora_strict_loading=False, lora_drain_wait_threshold=0.0, enable_two_batch_overlap=False, enable_single_batch_overlap=False, tbo_token_distribution_threshold=0.48, cpu_offload_gb=0, offload_group_size=-1, offload_num_in_group=1, offload_prefetch_step=1, offload_mode='cpu', enable_lmcache=False, lmcache_config_file=None, enable_flexkv=False, flexkv_config_file=None, kt_weight_path=None, kt_method='AMXINT4', kt_cpuinfer=None, kt_threadpool_count=2, kt_num_gpu_experts=None, kt_max_deferred_experts_per_token=None, dllm_algorithm=None, dllm_algorithm_config=None, dllm_fdfo=True, disaggregation_mode='decode', disaggregation_transfer_backend='mooncake', disaggregation_bootstrap_port=28800, disaggregation_ib_device='mlx5_0,mlx5_1,mlx5_2,mlx5_3', disaggregation_decode_enable_radix_cache=False, disaggregation_decode_enable_offload_kvcache=False, num_reserved_decode_tokens=512, disaggregation_decode_extra_slots=0, disaggregation_decode_polling_interval=1, optimistic_prefill_attempts=0, encoder_only=False, language_only=False, encoder_transfer_backend='zmq_to_tokenizer', encoder_urls=[], encoder_bootstrap_port=8997, encoder_register_urls=[], enable_adaptive_dispatch_to_encoder=False, enable_pdmux=False, pdmux_config_path=None, sm_group_num=8, custom_weight_loader=[], weight_loader_disable_mmap=False, weight_loader_prefetch_checkpoints=False, weight_loader_prefetch_num_threads=4, weight_loader_drop_cache_after_load=False, remote_instance_weight_loader_seed_instance_ip=None, remote_instance_weight_loader_seed_instance_service_port=None, remote_instance_weight_loader_send_weights_group_ports=None, remote_instance_weight_loader_backend='nccl', remote_instance_weight_loader_start_seed_via_transfer_engine=False, engine_info_bootstrap_port=6789, modelexpress_config=None, download_dir=None, model_checksum=None, delete_ckpt_after_loading=False, decrypted_config_file=None, decrypted_draft_config_file=None, checkpoint_engine_wait_weights_before_ready=False, enable_prefill_delayer=False, prefill_delayer_max_delay_passes=30, prefill_delayer_token_usage_low_watermark=None, prefill_delayer_forward_passes_buckets=None, prefill_delayer_wait_seconds_buckets=None, prefill_delayer_queue_min_ratio=None, prefill_delayer_max_delay_ms=None, min_free_slots_delay=None, enable_deterministic_inference=False, rl_on_policy_target=None, kv_canary='none', kv_canary_real_data='none', kv_canary_sweep_interval=0, enable_dynamic_batch_tokenizer=False, dynamic_batch_tokenizer_batch_size=32, dynamic_batch_tokenizer_batch_timeout=0.002, enable_tokenizer_batch_encode=False, disable_tokenizer_batch_decode=False, debug_tensor_dump_output_folder=None, debug_tensor_dump_layers=None, debug_tensor_dump_input_file=None, enable_memory_saver=False, enable_weights_cpu_backup=False, enable_draft_weights_cpu_backup=False, enable_custom_logit_processor=False, enable_return_hidden_states=False, enable_return_routed_experts=False, enable_return_indexer_topk=False, disable_outlines_disk_cache=False, enable_mis=False, weight_cache_mode='off', weight_cache_socket=None, weight_cache_timeout=1800, forward_hooks=None, msprobe_dump_config=None)
+[2026-08-27 04:36:14] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:14] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:14] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:14] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:14] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:14] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:14] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:14] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:17] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:17] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:18] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:19] Multimodal processor concurrency enabled with 2 isolated worker threads (auto).
+[2026-08-27 04:36:19] No HuggingFace chat template found
+[2026-08-27 04:36:19] No chat template found, defaulting to 'string' content format
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:34 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:34 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:34 TP4 EP4] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:34 TP4 EP4] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:34 TP6 EP6] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:34 TP6 EP6] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:34] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:34] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:34 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:34 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+/sgl-workspace/sglang/python/sglang/srt/utils/hf_transformers/tokenizer.py:423: UserWarning: Using a slow tokenizer. This might cause a significant slowdown. Consider using a fast tokenizer instead.
+  warnings.warn(
+[2026-08-27 04:36:34] Applying special tokens cache patch for Kimi tokenizer: <class 'transformers_modules.Kimi_hyphen_K3.caf40b2773f9029e.tokenization_kimi.TikTokenTokenizer'>
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:34 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:34 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:34 TP7 EP7] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:34 TP7 EP7] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:34 TP5 EP5] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:34 TP5 EP5] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:35 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:35 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:36 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:36 TP6 EP6] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:36 TP4 EP4] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:36 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:36 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:36 TP7 EP7] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:36 TP5 EP5] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:37 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:37.190651   236 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:37.190778   236 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:37.190796   236 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.5 port: 12001
+I0827 04:36:37.190856   236 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.5:16500
+I0827 04:36:37.191038   236 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:37.194572   241 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:37.194692   241 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:37.194712   241 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.5 port: 12001
+I0827 04:36:37.194767   241 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.5:15678
+I0827 04:36:37.194965   241 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:37.219420   236 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:37.223201   241 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:37.227941   236 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:37.231628   241 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:37.236233   236 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:37.239850   241 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:37.244647   236 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:37.248145   241 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:37.248418   236 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:37.248464   236 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:37.248525   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.250921   241 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:37.250947   241 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:37.251004   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.255973   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:37.257668   236 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:05
+I0827 04:36:37.257705   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.283210   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:37.283238   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:37.284873   236 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:05
+I0827 04:36:37.284902   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.284904   241 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:05
+I0827 04:36:37.284934   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.293713   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:37.295406   241 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:05
+I0827 04:36:37.295433   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.311285   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:37.312896   236 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:05
+I0827 04:36:37.312925   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.327399   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:37.328944   241 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:05
+I0827 04:36:37.328972   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.339579   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:37.341301   236 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:05
+I0827 04:36:37.341434   236 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:37 TP1 EP1] Init torch distributed begin.
+I0827 04:36:37.359290   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:37.360814   241 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:05
+I0827 04:36:37.360936   241 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:37 TP6 EP6] Init torch distributed begin.
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:37.366210   239 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:37.366365   239 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:37.366390   239 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.5 port: 12001
+I0827 04:36:37.366467   239 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.5:15787
+I0827 04:36:37.366704   239 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:37.395485   239 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:37.402652   239 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:37.409775   239 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:37.434130   239 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:37.436257   239 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:37.436282   239 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:37.436331   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.442425   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:37.443729   239 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:05
+I0827 04:36:37.443749   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.448421   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:37.449750   239 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:05
+I0827 04:36:37.449769   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.454622   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:37.455952   239 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:05
+I0827 04:36:37.455972   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.463356   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:37.464902   239 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:05
+I0827 04:36:37.465021   239 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:37 TP4 EP4] Init torch distributed begin.
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:37.510618   238 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:37.510744   238 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:37.510763   238 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.5 port: 12001
+I0827 04:36:37.510824   238 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.5:16205
+I0827 04:36:37.511003   238 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:37.524268   242 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:37.524391   242 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:37.524411   242 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.5 port: 12001
+I0827 04:36:37.524524   242 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.5:16605
+I0827 04:36:37.524709   242 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:37.528693   235 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:37.528817   235 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:37.528836   235 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.5 port: 12001
+I0827 04:36:37.528893   235 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.5:16672
+I0827 04:36:37.529069   235 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:37.535192   238 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:37.535425   235 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:37.541376   235 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:37.542625   238 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:37.547586   235 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:37.549854   238 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:37.551366   242 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:37.553905   235 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:37.556406   235 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:37.556468   235 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:37.556523   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.558794   242 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:37.563313   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:37.566102   242 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:37.573271   242 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:37.574895   238 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:37.576385   242 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:37.576412   242 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:37.576478   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.577947   238 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:37.577976   238 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:37.578033   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.583841   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:37.585472   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:37.585623   242 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:05
+I0827 04:36:37.585656   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.587072   238 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:05
+I0827 04:36:37.587105   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.594069   235 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:05
+I0827 04:36:37.594128   235 rdma_context.cpp:205] Using SIEVE endpoint store
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:37.598918   240 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:37.599033   240 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:37.599050   240 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.5 port: 12001
+I0827 04:36:37.599107   240 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.5:15428
+I0827 04:36:37.599270   240 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:37.605341   240 topology.cpp:127] Device mlx5_0 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:37.606741   237 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:37.606841   237 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:37.606855   237 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.5 port: 12001
+I0827 04:36:37.606901   237 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.5:16531
+I0827 04:36:37.607028   237 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:37.611215   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:37.611989   237 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:37.612797   242 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:05
+I0827 04:36:37.612828   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.615110   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:37.616685   238 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:05
+I0827 04:36:37.616714   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.619258   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:37.620560   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:37.620851   235 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:05
+I0827 04:36:37.620885   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.622239   242 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:05
+I0827 04:36:37.622268   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.624459   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:37.625998   238 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:05
+I0827 04:36:37.626025   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.630085   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:37.630686   240 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:37.631680   242 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:05
+I0827 04:36:37.631806   242 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:37 TP7 EP7] Init torch distributed begin.
+I0827 04:36:37.633723   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:37.635268   238 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:05
+I0827 04:36:37.635390   238 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:37 TP3 EP3] Init torch distributed begin.
+I0827 04:36:37.642561   237 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:37.649834   237 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:37.651238   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:37.652786   235 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:05
+I0827 04:36:37.652813   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.657148   237 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:37.658787   240 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:37.659991   237 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:37.660017   237 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:37.660071   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.667523   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:37.669195   240 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:37.669226   237 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:05
+I0827 04:36:37.669257   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.672511   240 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:37.672547   240 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:37.672606   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.677067   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:37.678661   237 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:05
+I0827 04:36:37.678689   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.682178   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:37.683274   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:37.683740   240 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:05
+I0827 04:36:37.683773   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.684834   235 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:05
+I0827 04:36:37.684954   235 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:37 TP0 EP0] Init torch distributed begin.
+I0827 04:36:37.703369   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:37.704975   237 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:05
+I0827 04:36:37.705003   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.711591   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:37.713003   240 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:05
+I0827 04:36:37.713030   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.735574   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:37.737313   237 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:05
+I0827 04:36:37.737439   237 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:37 TP2 EP2] Init torch distributed begin.
+I0827 04:36:37.740485   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:37.741588   240 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:05
+I0827 04:36:37.741605   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:37.766494   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:37.767982   240 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:05
+I0827 04:36:37.768137   240 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:37 TP5 EP5] Init torch distributed begin.
+[2026-08-27 04:36:46 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:36:48 TP0 EP0] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP1 EP1] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP2 EP2] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP5 EP5] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP4 EP4] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP3 EP3] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP6 EP6] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP7 EP7] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP0 EP0] Init torch distributed ends. elapsed=10.55 s, mem usage=0.40 GB
+[2026-08-27 04:36:48 TP7 EP7] Init torch distributed ends. elapsed=10.61 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP6 EP6] Init torch distributed ends. elapsed=10.88 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP4 EP4] Init torch distributed ends. elapsed=10.77 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP3 EP3] Init torch distributed ends. elapsed=10.60 s, mem usage=0.40 GB
+[2026-08-27 04:36:48 TP1 EP1] Init torch distributed ends. elapsed=10.90 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP5 EP5] Init torch distributed ends. elapsed=10.47 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP2 EP2] Init torch distributed ends. elapsed=10.50 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP7 EP7] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP4 EP4] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP1 EP1] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP6 EP6] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP7 EP7] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP4 EP4] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP5 EP5] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP6 EP6] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP3 EP3] Load weight begin. avail mem=82.19 GB
+[2026-08-27 04:36:48 TP0 EP0] Load weight begin. avail mem=82.19 GB
+[2026-08-27 04:36:48 TP5 EP5] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP2 EP2] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP4 EP4] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP7 EP7] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP6 EP6] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP4 EP4] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP7 EP7] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP6 EP6] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP5 EP5] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP5 EP5] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:50 TP1 EP1] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP0 EP0] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP2 EP2] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP3 EP3] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP5 EP5] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP4 EP4] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP6 EP6] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP7 EP7] multimem all-gather disabled because the TP group spans across nodes.
+
Multi-thread loading shards:   0% Completed | 0/96 [00:00<?, ?it/s]
Multi-thread loading shards:   1% Completed | 1/96 [00:00<00:24,  3.85it/s]
Multi-thread loading shards:   2% Completed | 2/96 [00:07<07:13,  4.61s/it]
Multi-thread loading shards:   3% Completed | 3/96 [00:09<04:44,  3.06s/it]
Multi-thread loading shards:   4% Completed | 4/96 [00:10<03:33,  2.32s/it]
Multi-thread loading shards:   5% Completed | 5/96 [00:11<02:54,  1.92s/it]
Multi-thread loading shards:   6% Completed | 6/96 [00:12<02:29,  1.66s/it]
Multi-thread loading shards:   7% Completed | 7/96 [00:13<02:13,  1.50s/it]
Multi-thread loading shards:   8% Completed | 8/96 [00:15<02:04,  1.41s/it]
Multi-thread loading shards:   9% Completed | 9/96 [00:16<01:56,  1.34s/it]
Multi-thread loading shards:  10% Completed | 10/96 [00:17<01:47,  1.25s/it]
Multi-thread loading shards:  11% Completed | 11/96 [00:18<01:44,  1.23s/it]
Multi-thread loading shards:  12% Completed | 12/96 [00:19<01:43,  1.23s/it]
Multi-thread loading shards:  14% Completed | 13/96 [00:21<01:59,  1.44s/it]
Multi-thread loading shards:  15% Completed | 14/96 [00:22<01:51,  1.36s/it]
Multi-thread loading shards:  16% Completed | 15/96 [00:24<01:45,  1.30s/it]
Multi-thread loading shards:  17% Completed | 16/96 [00:25<01:42,  1.28s/it]
Multi-thread loading shards:  18% Completed | 17/96 [00:26<01:39,  1.26s/it]
Multi-thread loading shards:  19% Completed | 18/96 [00:27<01:36,  1.24s/it]
Multi-thread loading shards:  20% Completed | 19/96 [00:28<01:32,  1.20s/it]
Multi-thread loading shards:  21% Completed | 20/96 [00:29<01:30,  1.19s/it]
Multi-thread loading shards:  22% Completed | 21/96 [00:31<01:29,  1.19s/it]
Multi-thread loading shards:  23% Completed | 22/96 [00:32<01:27,  1.18s/it]
Multi-thread loading shards:  24% Completed | 23/96 [00:33<01:26,  1.18s/it]
Multi-thread loading shards:  25% Completed | 24/96 [00:34<01:23,  1.16s/it]
Multi-thread loading shards:  26% Completed | 25/96 [00:35<01:20,  1.14s/it]
Multi-thread loading shards:  27% Completed | 26/96 [00:36<01:18,  1.13s/it]
Multi-thread loading shards:  28% Completed | 27/96 [00:37<01:19,  1.15s/it]
Multi-thread loading shards:  29% Completed | 28/96 [00:39<01:18,  1.16s/it]
Multi-thread loading shards:  30% Completed | 29/96 [00:40<01:17,  1.16s/it]
Multi-thread loading shards:  31% Completed | 30/96 [00:41<01:17,  1.18s/it]
Multi-thread loading shards:  32% Completed | 31/96 [00:42<01:16,  1.18s/it]
Multi-thread loading shards:  33% Completed | 32/96 [00:44<01:30,  1.42s/it]
Multi-thread loading shards:  34% Completed | 33/96 [00:45<01:23,  1.33s/it]
Multi-thread loading shards:  35% Completed | 34/96 [00:46<01:17,  1.26s/it]
Multi-thread loading shards:  36% Completed | 35/96 [00:48<01:14,  1.23s/it]
Multi-thread loading shards:  38% Completed | 36/96 [00:49<01:13,  1.22s/it]
Multi-thread loading shards:  39% Completed | 37/96 [00:50<01:11,  1.21s/it]
Multi-thread loading shards:  40% Completed | 38/96 [00:51<01:09,  1.20s/it]
Multi-thread loading shards:  41% Completed | 39/96 [00:52<01:07,  1.18s/it]
Multi-thread loading shards:  42% Completed | 40/96 [00:53<01:05,  1.18s/it]
Multi-thread loading shards:  43% Completed | 41/96 [00:55<01:04,  1.17s/it]
Multi-thread loading shards:  44% Completed | 42/96 [00:56<01:02,  1.17s/it]
Multi-thread loading shards:  45% Completed | 43/96 [00:57<01:03,  1.19s/it]
Multi-thread loading shards:  46% Completed | 44/96 [00:58<01:01,  1.18s/it]
Multi-thread loading shards:  47% Completed | 45/96 [00:59<00:57,  1.13s/it]
Multi-thread loading shards:  48% Completed | 46/96 [01:00<00:57,  1.15s/it]
Multi-thread loading shards:  49% Completed | 47/96 [01:02<00:57,  1.17s/it]
Multi-thread loading shards:  50% Completed | 48/96 [01:03<00:55,  1.15s/it]
Multi-thread loading shards:  51% Completed | 49/96 [01:04<00:53,  1.13s/it]
Multi-thread loading shards:  52% Completed | 50/96 [01:06<01:01,  1.34s/it]
Multi-thread loading shards:  53% Completed | 51/96 [01:07<00:58,  1.29s/it]
Multi-thread loading shards:  54% Completed | 52/96 [01:08<00:55,  1.27s/it]
Multi-thread loading shards:  55% Completed | 53/96 [01:09<00:53,  1.24s/it]
Multi-thread loading shards:  56% Completed | 54/96 [01:10<00:50,  1.21s/it]
Multi-thread loading shards:  57% Completed | 55/96 [01:12<00:49,  1.22s/it]
Multi-thread loading shards:  58% Completed | 56/96 [01:13<00:48,  1.22s/it]
Multi-thread loading shards:  59% Completed | 57/96 [01:14<00:47,  1.21s/it]
Multi-thread loading shards:  60% Completed | 58/96 [01:15<00:45,  1.20s/it]
Multi-thread loading shards:  61% Completed | 59/96 [01:16<00:44,  1.19s/it]
Multi-thread loading shards:  62% Completed | 60/96 [01:17<00:43,  1.20s/it]
Multi-thread loading shards:  64% Completed | 61/96 [01:19<00:42,  1.21s/it]
Multi-thread loading shards:  65% Completed | 62/96 [01:20<00:40,  1.19s/it]
Multi-thread loading shards:  66% Completed | 63/96 [01:21<00:40,  1.21s/it]
Multi-thread loading shards:  67% Completed | 64/96 [01:22<00:38,  1.20s/it]
Multi-thread loading shards:  68% Completed | 65/96 [01:24<00:37,  1.20s/it]
Multi-thread loading shards:  69% Completed | 66/96 [01:25<00:36,  1.20s/it]
Multi-thread loading shards:  70% Completed | 67/96 [01:26<00:35,  1.22s/it]
Multi-thread loading shards:  71% Completed | 68/96 [01:28<00:41,  1.48s/it]
Multi-thread loading shards:  72% Completed | 69/96 [01:29<00:37,  1.39s/it]
Multi-thread loading shards:  73% Completed | 70/96 [01:30<00:34,  1.34s/it]
Multi-thread loading shards:  74% Completed | 71/96 [01:32<00:32,  1.29s/it]
Multi-thread loading shards:  75% Completed | 72/96 [01:33<00:31,  1.30s/it]
Multi-thread loading shards:  76% Completed | 73/96 [01:34<00:29,  1.29s/it]
Multi-thread loading shards:  77% Completed | 74/96 [01:35<00:27,  1.27s/it]
Multi-thread loading shards:  78% Completed | 75/96 [01:37<00:26,  1.26s/it]
Multi-thread loading shards:  79% Completed | 76/96 [01:38<00:24,  1.24s/it]
Multi-thread loading shards:  80% Completed | 77/96 [01:39<00:23,  1.23s/it]
Multi-thread loading shards:  81% Completed | 78/96 [01:40<00:21,  1.19s/it]
Multi-thread loading shards:  82% Completed | 79/96 [01:41<00:20,  1.20s/it]
Multi-thread loading shards:  83% Completed | 80/96 [01:43<00:19,  1.20s/it]
Multi-thread loading shards:  84% Completed | 81/96 [01:44<00:18,  1.21s/it]
Multi-thread loading shards:  85% Completed | 82/96 [01:45<00:16,  1.20s/it]
Multi-thread loading shards:  86% Completed | 83/96 [01:46<00:15,  1.17s/it]
Multi-thread loading shards:  88% Completed | 84/96 [01:47<00:14,  1.17s/it]
Multi-thread loading shards:  89% Completed | 85/96 [01:48<00:12,  1.10s/it]
Multi-thread loading shards:  90% Completed | 86/96 [01:49<00:10,  1.05s/it]
Multi-thread loading shards:  91% Completed | 87/96 [01:50<00:09,  1.02s/it]
Multi-thread loading shards:  92% Completed | 88/96 [01:51<00:07,  1.01it/s]
Multi-thread loading shards:  93% Completed | 89/96 [01:52<00:06,  1.03it/s]
Multi-thread loading shards:  94% Completed | 90/96 [01:53<00:05,  1.04it/s]
Multi-thread loading shards:  95% Completed | 91/96 [01:54<00:04,  1.06it/s]
Multi-thread loading shards:  96% Completed | 92/96 [01:55<00:03,  1.06it/s]
Multi-thread loading shards:  97% Completed | 93/96 [01:56<00:02,  1.08it/s]
Multi-thread loading shards:  98% Completed | 94/96 [01:56<00:01,  1.46it/s]
Multi-thread loading shards: 100% Completed | 96/96 [01:56<00:00,  2.28it/s]
Multi-thread loading shards: 100% Completed | 96/96 [01:56<00:00,  1.21s/it]
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:48 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (1152, 4), A_log (1, 1, 3, 1), dt_bias (384,))
+[2026-08-27 04:38:58 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP6 EP6] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP0 EP0] Load weight end. elapsed=130.28 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP6 EP6] Load weight end. elapsed=130.29 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP3 EP3] Load weight end. elapsed=130.28 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP2 EP2] Load weight end. elapsed=130.32 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP5 EP5] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP5 EP5] Load weight end. elapsed=130.38 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP4 EP4] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP4 EP4] Load weight end. elapsed=130.48 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP7 EP7] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP7 EP7] Load weight end. elapsed=130.48 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:39:01 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:39:01 TP1 EP1] Load weight end. elapsed=132.81 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:39:01 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP6 EP6] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP6 EP6] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP5 EP5] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP5 EP5] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP7 EP7] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP7 EP7] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP4 EP4] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP4 EP4] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:02 TP0 EP0] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP7 EP7] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP6 EP6] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP5 EP5] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP1 EP1] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP4 EP4] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP3 EP3] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP2 EP2] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP7 EP7] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP4 EP4] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP6 EP6] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP5 EP5] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP0 EP0] Memory pool end. avail mem=9.18 GB
+[2026-08-27 04:39:02 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP1 EP1] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP6 EP6] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP5 EP5] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP7 EP7] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP6 EP6] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP5 EP5] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP7 EP7] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP4 EP4] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP4 EP4] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP2 EP2] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP3 EP3] Memory pool end. avail mem=9.18 GB
+[2026-08-27 04:39:02 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:39:02 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:39:02 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP0 EP0] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.17 GB
+[2026-08-27 04:39:02 TP6 EP6] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP6 EP6] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP1 EP1] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP5 EP5] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP7 EP7] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP5 EP5] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP7 EP7] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:39:02 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP3 EP3] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.17 GB
+[2026-08-27 04:39:02 TP4 EP4] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP4 EP4] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:39:02 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP2 EP2] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+
  0%|          | 0/6 [00:00<?, ?it/s]
Capturing batches (bs=16 avail_mem=9.00 GB):   0%|          | 0/6 [00:00<?, ?it/s]
Capturing batches (bs=16 avail_mem=9.00 GB):  17%|█▋        | 1/6 [03:43<18:38, 223.66s/it]
Capturing batches (bs=12 avail_mem=8.84 GB):  17%|█▋        | 1/6 [03:43<18:38, 223.66s/it]
Capturing batches (bs=12 avail_mem=8.84 GB):  33%|███▎      | 2/6 [03:44<06:11, 92.84s/it] 
Capturing batches (bs=8 avail_mem=8.78 GB):  33%|███▎      | 2/6 [03:44<06:11, 92.84s/it] 
Capturing batches (bs=8 avail_mem=8.78 GB):  50%|█████     | 3/6 [03:49<02:36, 52.32s/it]
Capturing batches (bs=4 avail_mem=8.73 GB):  50%|█████     | 3/6 [03:49<02:36, 52.32s/it]
Capturing batches (bs=4 avail_mem=8.73 GB):  67%|██████▋   | 4/6 [03:49<01:03, 32.00s/it]
Capturing batches (bs=2 avail_mem=8.68 GB):  67%|██████▋   | 4/6 [03:49<01:03, 32.00s/it]
Capturing batches (bs=2 avail_mem=8.68 GB):  83%|████████▎ | 5/6 [03:50<00:20, 20.77s/it]
Capturing batches (bs=1 avail_mem=8.62 GB):  83%|████████▎ | 5/6 [03:50<00:20, 20.77s/it]
Capturing batches (bs=1 avail_mem=8.62 GB): 100%|██████████| 6/6 [03:52<00:00, 14.36s/it]
Capturing batches (bs=1 avail_mem=8.62 GB): 100%|██████████| 6/6 [03:52<00:00, 38.78s/it]
+[2026-08-27 04:42:56 TP4 EP4] Capture target decode CUDA graph end. elapsed=233.75 s, mem usage=0.60 GB, avail mem=8.52 GB.
+[2026-08-27 04:42:56 TP4 EP4] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP3 EP3] Capture target decode CUDA graph end. elapsed=233.75 s, mem usage=0.62 GB, avail mem=8.55 GB.
+[2026-08-27 04:42:56 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP0 EP0] Capture target decode CUDA graph end. elapsed=233.77 s, mem usage=0.60 GB, avail mem=8.56 GB.
+[2026-08-27 04:42:56 TP0 EP0] max_total_num_tokens=440256, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=181, context_len=1048576, available_gpu_mem=8.56 GB
+[2026-08-27 04:42:56 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP7 EP7] Capture target decode CUDA graph end. elapsed=233.77 s, mem usage=0.62 GB, avail mem=8.51 GB.
+[2026-08-27 04:42:56 TP7 EP7] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP6 EP6] Capture target decode CUDA graph end. elapsed=233.86 s, mem usage=0.56 GB, avail mem=8.59 GB.
+[2026-08-27 04:42:56 TP6 EP6] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP1 EP1] Capture target decode CUDA graph end. elapsed=233.86 s, mem usage=0.58 GB, avail mem=8.58 GB.
+[2026-08-27 04:42:56 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP2 EP2] Capture target decode CUDA graph end. elapsed=233.86 s, mem usage=0.56 GB, avail mem=8.59 GB.
+[2026-08-27 04:42:56 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP5 EP5] Capture target decode CUDA graph end. elapsed=233.90 s, mem usage=0.58 GB, avail mem=8.58 GB.
+[2026-08-27 04:42:56 TP5 EP5] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+W0827 04:42:57.550421  2569 memory_location.cpp:73] Failed to get NUMA node, addr: 0x599a0800, len: 23168: Operation not permitted [1]
+W0827 04:42:57.550521  2570 memory_location.cpp:73] Failed to get NUMA node, addr: 0x599cb380, len: 23168: Operation not permitted [1]
+W0827 04:42:57.550542  2571 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59e1b540, len: 23168: Operation not permitted [1]
+W0827 04:42:57.550628  2572 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59c8a2c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.550725  2573 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5be7a240, len: 185344: Operation not permitted [1]
+W0827 04:42:57.550769  2574 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5bea7680, len: 185344: Operation not permitted [1]
+W0827 04:42:57.550900  2575 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59cc48c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.551011  2576 memory_location.cpp:73] Failed to get NUMA node, addr: 0x45f39fc0, len: 46336: Operation not permitted [1]
+W0827 04:42:57.551070  2577 memory_location.cpp:73] Failed to get NUMA node, addr: 0x599bfbc0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.551105  2578 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59c69c80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.580384  2676 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59c988c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.580493  2678 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59a30c40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.580636  2677 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59d06340, len: 23168: Operation not permitted [1]
+W0827 04:42:57.580636  2679 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5bd91640, len: 185344: Operation not permitted [1]
+W0827 04:42:57.580685  2675 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59e30d80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.580811  2681 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59f53440, len: 23168: Operation not permitted [1]
+W0827 04:42:57.580950  2680 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5bdbea80, len: 185344: Operation not permitted [1]
+W0827 04:42:57.581368  2686 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59dca000, len: 23168: Operation not permitted [1]
+W0827 04:42:57.581421  2682 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5987fc00, len: 46336: Operation not permitted [1]
+W0827 04:42:57.581741  2684 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59e10480, len: 23168: Operation not permitted [1]
+W0827 04:42:57.597819  2845 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59224900, len: 23168: Operation not permitted [1]
+W0827 04:42:57.598558  2839 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5917e740, len: 23168: Operation not permitted [1]
+W0827 04:42:57.598562  2849 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58e6ce00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.598762  2838 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58e982c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.598834  2841 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5b32ec00, len: 185344: Operation not permitted [1]
+W0827 04:42:57.598897  2834 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5927b680, len: 23168: Operation not permitted [1]
+W0827 04:42:57.598922  2843 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5b35c040, len: 185344: Operation not permitted [1]
+W0827 04:42:57.598953  2836 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5939fc40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.599097  2850 memory_location.cpp:73] Failed to get NUMA node, addr: 0x592d3fc0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.599287  2847 memory_location.cpp:73] Failed to get NUMA node, addr: 0x453f4c40, len: 46336: Operation not permitted [1]
+W0827 04:42:57.612057  2918 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59f4a140, len: 23168: Operation not permitted [1]
+W0827 04:42:57.612175  2919 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59f098c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.612243  2920 memory_location.cpp:73] Failed to get NUMA node, addr: 0x46045dc0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.612710  2925 memory_location.cpp:73] Failed to get NUMA node, addr: 0x46106480, len: 46336: Operation not permitted [1]
+W0827 04:42:57.612715  2924 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59a7ab40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.612880  2922 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5bf25300, len: 185344: Operation not permitted [1]
+W0827 04:42:57.612898  2927 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59e8a9c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.612931  2921 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59fec840, len: 23168: Operation not permitted [1]
+W0827 04:42:57.612948  2923 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5bf52740, len: 185344: Operation not permitted [1]
+W0827 04:42:57.612996  2926 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59edf040, len: 23168: Operation not permitted [1]
+W0827 04:42:57.618498  2975 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a4cda80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.618582  2974 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a3878c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.618582  2973 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a3abc80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.618669  2976 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5c40da00, len: 185344: Operation not permitted [1]
+W0827 04:42:57.618669  2978 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59f66a40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.618741  2977 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5c43ae40, len: 185344: Operation not permitted [1]
+W0827 04:42:57.618854  2979 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59ce6c00, len: 46336: Operation not permitted [1]
+W0827 04:42:57.618898  2980 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a2ebe80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.619125  2982 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a37d0c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.632592  2972 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a371340, len: 23168: Operation not permitted [1]
+W0827 04:42:57.776260  3310 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58ee7080, len: 23168: Operation not permitted [1]
+W0827 04:42:57.776290  3311 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5854f480, len: 23168: Operation not permitted [1]
+W0827 04:42:57.776294  3312 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58edd900, len: 23168: Operation not permitted [1]
+W0827 04:42:57.776240  3309 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58f70e40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.776324  3313 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a63db40, len: 185344: Operation not permitted [1]
+W0827 04:42:57.776337  3314 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a66af80, len: 185344: Operation not permitted [1]
+W0827 04:42:57.776417  3315 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58e5ec40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.776444  3316 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58109800, len: 46336: Operation not permitted [1]
+W0827 04:42:57.776451  3317 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58ed2cc0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.776649  3318 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58316800, len: 23168: Operation not permitted [1]
+W0827 04:42:57.812515  3457 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58919a80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.812533  3458 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5892c840, len: 23168: Operation not permitted [1]
+W0827 04:42:57.812597  3460 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58063380, len: 23168: Operation not permitted [1]
+W0827 04:42:57.812603  3459 memory_location.cpp:73] Failed to get NUMA node, addr: 0x588ed280, len: 23168: Operation not permitted [1]
+W0827 04:42:57.812654  3461 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59fc0080, len: 185344: Operation not permitted [1]
+W0827 04:42:57.812734  3462 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59fed4c0, len: 185344: Operation not permitted [1]
+W0827 04:42:57.812816  3463 memory_location.cpp:73] Failed to get NUMA node, addr: 0x57ebf300, len: 23168: Operation not permitted [1]
+W0827 04:42:57.812877  3464 memory_location.cpp:73] Failed to get NUMA node, addr: 0x44a36bc0, len: 46336: Operation not permitted [1]
+W0827 04:42:57.812991  3466 memory_location.cpp:73] Failed to get NUMA node, addr: 0x44982040, len: 23168: Operation not permitted [1]
+W0827 04:42:57.812996  3465 memory_location.cpp:73] Failed to get NUMA node, addr: 0x588ba900, len: 23168: Operation not permitted [1]
+W0827 04:42:57.851508  3605 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5888adc0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.851550  3606 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58775700, len: 23168: Operation not permitted [1]
+W0827 04:42:57.851619  3607 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58aca800, len: 23168: Operation not permitted [1]
+W0827 04:42:57.851691  3608 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5886ea80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.851799  3609 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5aaacb40, len: 185344: Operation not permitted [1]
+W0827 04:42:57.851871  3611 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58ad7b00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.851959  3610 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5aad9f80, len: 185344: Operation not permitted [1]
+W0827 04:42:57.851996  3612 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58aa8840, len: 46336: Operation not permitted [1]
+W0827 04:42:57.852005  3614 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58ca01c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.852105  3613 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5879a9c0, len: 23168: Operation not permitted [1]
+[2026-08-27 04:43:00] INFO:     Started server process [1]
+[2026-08-27 04:43:00] INFO:     Waiting for application startup.
+[2026-08-27 04:43:00] Calling super().encode with {'add_special_tokens': False}
+[2026-08-27 04:43:00] Calling super().encode with {'add_special_tokens': False}
+[2026-08-27 04:43:00] INFO:     Application startup complete.
+[2026-08-27 04:43:01] INFO:     174.1.60.1:55082 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:43:01] INFO:     127.0.0.1:38530 - "GET /model_info HTTP/1.1" 200 OK
+[2026-08-27 04:43:02] INFO:     174.1.60.1:55098 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:43:03] INFO:     127.0.0.1:38536 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 04:43:04] INFO:     174.1.60.1:55110 - "GET /health HTTP/1.1" 200 OK
+[2026-08-27 04:43:25] INFO:     174.1.60.1:41170 - "GET /health HTTP/1.1" 200 OK
+[2026-08-27 04:48:11] INFO:     174.1.60.5:38682 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 04:43:00] INFO:     Uvicorn running on http://0.0.0.0:30000 (Press CTRL+C to quit)
+[2026-08-27 04:43:01] Start of pd disaggregation warmup ...
+[2026-08-27 04:43:03] Disaggregation warmup requests completed for all 1 DP ranks
+[2026-08-27 04:43:03] End of disaggregation warmup
+[2026-08-27 04:43:03] The server is fired up and ready to roll!
+I0827 04:48:13.997220  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:13.997509  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.013371  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.013677  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.042984  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.043259  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.046793  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.047015  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.057778  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.067556  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.076747  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.095343  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.099277  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.105573  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.105803  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.106056  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.109412  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.109645  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.109865  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.110114  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.110388  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.110597  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.110783  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.114174  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.114495  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.120798  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.136216  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.142400  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.142663  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.153999  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.154243  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.154534  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.154755  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.154990  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.158114  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.158325  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.170076  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.170320  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.170512  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.170779  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.185034  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.197149  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.197443  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.197665  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.197889  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.203205  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.203492  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.203737  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.203975  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.204202  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.215955  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.216298  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.216583  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.216789  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.222101  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.223455  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.232357  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.233336  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.233599  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.233934  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.248603  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.248853  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.249085  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.250525  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.252888  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.253137  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.253366  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.253639  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.256778  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.258342  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.259446  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.259649  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.259908  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.260114  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.260375  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.260576  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.262905  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.263399  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.263654  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.263896  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.264135  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.264392  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.264699  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.264971  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.276409  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.276680  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.280252  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.281527  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.289309  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.289914  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.290587  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.299579  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.299823  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.303650  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.303859  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.304064  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.307004  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.307231  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.307541  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.307801  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.308017  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.308123  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.308235  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.308485  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.308728  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.308988  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.309187  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.309360  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.309405  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.309656  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.309904  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.310132  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.310344  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.312103  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.312309  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.314198  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.315554  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.316985  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.320315  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.327411  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.337924  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.339138  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.344900  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.346967  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.348446  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.361662  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.361877  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.362105  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.365690  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.366845  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.373291  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.373983  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.374544  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.391561  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.391779  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.392053  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.392257  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.392505  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.392726  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.393024  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.393260  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.393772  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.395025  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.396075  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.396298  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.396525  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.396822  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.397025  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.397225  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.403301  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.404704  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.407682  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.417869  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.418104  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.418273  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.421886  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.422092  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.422293  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.422566  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.422824  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.424790  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.427910  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.429622  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.433352  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.433625  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.433885  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.434249  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.434464  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.436031  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.444931  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.445168  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.445376  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.445645  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.448356  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.448545  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.448810  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.449036  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.449236  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.450745  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.453661  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.455060  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.456887  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.457075  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.458276  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.463183  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.464545  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.473290  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.473594  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.473889  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.474606  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.478461  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.478672  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.478849  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.485463  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.485714  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.486359  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.486591  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.486837  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.486868  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.487639  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.492800  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.494125  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.496006  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.496238  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.496584  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.504025  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.504601  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.504830  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.505040  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.505227  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.505416  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.505596  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.505838  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.506070  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.506418  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.506714  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.506938  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.507185  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.507364  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.507618  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.507850  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.508181  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.509447  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.510848  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.510802  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.511059  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.511301  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.512179  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.513973  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.520185  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.521516  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.528406  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.528625  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.528868  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.532759  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.535007  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.535238  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.535440  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.535620  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.536624  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.538493  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.541709  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.541929  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.543128  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.543134  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.543255  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.543396  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.543586  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.543852  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.544047  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.544274  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.544514  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.544595  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.544750  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.544992  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.545264  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.545506  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.545732  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.549163  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.556982  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.557228  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.557428  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.558394  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.559183  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.563414  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.564719  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.564847  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.566068  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.568861  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.574138  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.575349  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.583217  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.583469  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.583667  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.584623  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.588591  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.588935  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.589197  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.590059  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.590425  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.591902  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.599382  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.599529  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.599663  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.599762  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.600073  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.600317  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.600823  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.605185  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.605659  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.605909  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.609126  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.610448  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.611255  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.614173  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.616300  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.616987  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.617319  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.617666  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.620188  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.620463  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.623186  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.623404  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.626320  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.629146  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.632516  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.632800  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.633054  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.634070  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.634683  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.637434  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.640316  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.640353  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.640614  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.641824  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.643349  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.646051  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.658253  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.662520  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.663751  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.663796  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.668447  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.674060  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.683264  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.683470  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.683725  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.684523  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.684583  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.684775  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.685092  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.685344  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.685595  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.685825  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.686066  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.694078  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.694272  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.694502  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.694731  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.696010  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.696235  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.696461  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.697254  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.702469  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.702711  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.702946  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.703189  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.712618  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.716657  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.717875  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.718667  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.718958  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.719236  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.722149  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.724561  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.724784  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.724995  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.730228  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.730405  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.730648  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.730953  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.731205  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.731441  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.736352  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.736553  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.736785  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.737946  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.739337  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.739604  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.743849  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.745306  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.747385  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.750119  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.755378  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.755673  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.755957  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.756151  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.756417  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.756649  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.756873  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.757055  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.757272  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.762418  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.762653  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.767700  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.767921  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.768116  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.769114  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.805779  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.806072  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.806393  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.816823  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.817051  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.817302  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.817519  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.822906  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.823174  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.823410  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.823618  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.828871  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.831584  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.834362  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.834635  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.845114  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.847841  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.853120  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.853320  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.853577  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.853799  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.856580  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.859321  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.859536  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.864822  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.875097  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.895438  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.895658  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.903518  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.903801  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.904052  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.904273  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.909722  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.911178  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.912268  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.912475  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.912720  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.912957  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.913183  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.913419  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.913677  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.913909  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.914156  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.914402  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.914625  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.917008  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.919989  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.920225  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.922858  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.923017  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.935199  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.935660  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.935937  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.939235  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.941046  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.943796  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.944716  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.946557  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.947388  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.949306  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.949542  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.949780  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.950085  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.950371  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.950598  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.950831  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.951049  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.951236  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.951502  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.955791  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.956029  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.964305  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.964589  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.964807  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.965030  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.965714  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.969712  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.971029  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.976545  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.978840  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.982290  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.993181  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.993423  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.993613  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.995042  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.995337  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.995555  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.996627  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.007792  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.008044  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.008275  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.009122  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.013960  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.014259  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.014477  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.014698  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.014875  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.015080  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.015322  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.015535  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.023283  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.024626  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.028729  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.028869  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.029064  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.029294  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.029582  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.030203  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.035056  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.047416  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.047739  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.047941  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.048739  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.054301  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.055744  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.056360  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.056833  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.057034  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.057294  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.057492  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.057713  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.057981  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.058207  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.058401  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.058617  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.069922  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.071324  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.075461  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.081681  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.093765  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.093961  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.094169  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.095338  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.101928  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.102181  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.102407  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.102629  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.102851  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.103041  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.103250  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.103484  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.103757  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.103947  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.104204  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.104463  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.104683  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.104951  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.105253  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.105489  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.105674  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.105890  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.106104  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.106318  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.107211  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.120297  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.124440  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.125627  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.131145  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.131403  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.131698  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.131965  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.132176  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.132416  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.132588  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.132663  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.132896  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.133141  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.133297  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.146152  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.146390  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.146620  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.146828  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.147426  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.156941  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.157151  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.178486  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.178771  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.179106  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.179323  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.208334  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.208683  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.208947  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.209216  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.209455  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.209815  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.210053  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.210210  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.210412  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.210608  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.210822  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.221958  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.222180  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.222395  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.223281  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.225317  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.226866  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.230994  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.237720  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.242835  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.243105  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.243345  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.243551  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.243834  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.244047  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.244271  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.250334  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.250538  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.250738  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.251941  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.257243  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.258617  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.273547  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.277629  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.278937  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.281635  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.281862  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.282078  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.282316  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.282523  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.282783  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.283066  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.283294  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.283550  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.283828  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.284119  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.284336  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.284598  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.285751  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.287382  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.298511  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.298730  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.298920  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.299866  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.301535  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.302943  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.305897  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.306208  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.306404  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.306623  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.306824  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.307075  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.307155  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.307289  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.307511  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.310241  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.310458  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.310705  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.311606  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.313308  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.320852  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.321092  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.321306  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.321498  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.322360  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.323149  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.323376  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.344715  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.362433  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.365234  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.365482  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.365670  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.365831  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.366075  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.366329  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.366532  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.369321  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.379413  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.379633  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.379848  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.380110  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.380908  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.388870  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.389055  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.393224  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.396981  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.397784  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.398291  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.404456  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.405587  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.405781  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.405988  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.416702  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.416770  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.416965  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.417004  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.417227  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.418077  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.419777  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.425088  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.425359  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.425642  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.425822  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.428227  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.428457  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.428719  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.429623  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.431046  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.433805  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.434010  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.441807  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.442006  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.442276  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.447495  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.447705  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.447944  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.453174  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.453361  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.456089  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.456282  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.457403  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.487757  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.487957  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.488181  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.488385  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.488687  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.489209  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.497211  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.497412  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.498611  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.506928  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.507129  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.508323  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.533521  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.533809  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.533984  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.534192  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.534411  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.534649  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.534850  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.535123  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.535336  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.538051  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.553102  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.555838  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.556046  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.556244  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.556519  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.556715  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.556900  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.559612  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.567348  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.570070  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.570282  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.570564  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.571498  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+[2026-08-27 04:48:16] INFO:     174.1.60.5:49356 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 04:48:16] INFO:     174.1.60.5:49370 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 04:57:13] INFO:     174.1.60.5:34722 - "POST /generate HTTP/1.1" 200 OK
+I0827 04:57:14.461794  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.462142  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.463397  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.464907  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.466337  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.472958  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.474522  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.477468  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.477753  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.479032  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.480643  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.498212  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.498531  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.498838  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.499125  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.499409  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.499795  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.505314  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.517073  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.518641  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.527722  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.529148  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+[2026-08-27 04:57:14] INFO:     174.1.60.5:34732 - "GET /server_info HTTP/1.1" 200 OK
+I0827 04:57:14.534837  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+[2026-08-27 04:57:14] INFO:     174.1.60.5:34748 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:00:07] INFO:     174.1.60.5:40816 - "POST /generate HTTP/1.1" 200 OK
+I0827 05:04:21.299574  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.301399  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.307986  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.310060  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.311648  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.317925  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.321113  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.321422  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.322642  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.325122  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.326697  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.332526  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.332682  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.334165  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.340161  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.341799  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.343374  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.345765  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.347450  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.349627  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.350939  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.352005  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.355481  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.355765  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.357254  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.357306  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.358845  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.363112  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.363651  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.368119  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.373224  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.374758  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.379626  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.379845  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.380074  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.381069  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.382846  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.384393  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.386190  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.387002  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.387748  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.388473  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.388860  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.399398  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.399649  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.400830  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.402611  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.402901  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.404011  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.405441  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.408586  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.409112  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.414987  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.416474  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.418708  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.420038  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.421981  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.424405  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.437287  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.437530  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.438683  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.443255  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+[2026-08-27 05:04:22] INFO:     174.1.60.5:33422 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:04:22] INFO:     174.1.60.5:33424 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:04:22 TP0 EP0] Decode batch, #running-req: 1, #full token: 0, full token usage: 0.00, mamba num: 0, mamba usage: 0.00, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 0.03, #queue-req: 0
+I0827 05:04:41.398729  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.399127  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.400347  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.401475  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.401768  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.403021  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.413998  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.415591  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+[2026-08-27 05:04:42] INFO:     174.1.60.5:58198 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:05:38] INFO:     174.1.60.5:48352 - "POST /generate HTTP/1.1" 200 OK
+I0827 05:04:41.421803  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.576606  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.576977  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.578258  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.600133  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.600483  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.600749  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.601066  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.601655  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.607283  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.615131  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.620574  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.620859  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.621142  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.622053  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+[2026-08-27 05:07:16] INFO:     174.1.60.5:48172 - "POST /generate HTTP/1.1" 200 OK
+I0827 05:05:39.626708  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.148217  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.148566  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.149849  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+[2026-08-27 05:07:33] INFO:     174.1.60.5:48616 - "POST /generate HTTP/1.1" 200 OK
+I0827 05:07:33.155894  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+[2026-08-27 05:07:40] INFO:     174.1.60.5:48620 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:07:48] INFO:     174.1.60.5:33014 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:07:56] INFO:     174.1.60.5:48046 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:08:03] INFO:     174.1.60.5:43662 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:08:11] INFO:     174.1.60.5:50736 - "POST /generate HTTP/1.1" 200 OK
+I0827 05:08:18.657902  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.659292  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.665387  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.682516  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.682862  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.683116  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.683411  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.683794  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.684041  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.690148  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.690267  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.691737  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.698587  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.699856  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.700106  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.701548  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.704351  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.705502  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.705786  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.706010  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.706919  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.708040  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.713306  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.714720  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.720657  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.722337  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+[2026-08-27 05:08:18] INFO:     174.1.60.5:50748 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:08:26] INFO:     174.1.60.5:48502 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:08:26] INFO:     174.1.60.5:48506 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:11:44] INFO:     174.1.60.5:54256 - "POST /generate HTTP/1.1" 200 OK
+I0827 05:08:18.726626  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+[2026-08-27 05:11:46] INFO:     174.1.60.5:54260 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:11:46] INFO:     174.1.60.5:54274 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:11:46 TP0 EP0] Decode batch, #running-req: 1, #full token: 0, full token usage: 0.00, mamba num: 0, mamba usage: 0.00, pre-allocated usage: 0.00, #prealloc-req: 0, #transfer-req: 0, #retracted-req: 0, cuda graph: True, gen throughput (token/s): 0.09, #queue-req: 0
+I0827 05:14:42.491809  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+[2026-08-27 05:14:43] INFO:     174.1.60.5:52586 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:14:43] INFO:     174.1.60.5:52600 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:14:43] INFO:     174.1.60.5:52616 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:17:40] INFO:     174.1.60.5:47526 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:17:47] INFO:     174.1.60.5:33870 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:17:55] INFO:     174.1.60.5:33666 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:18:02] INFO:     174.1.60.5:59970 - "POST /generate HTTP/1.1" 200 OK
+I0827 05:14:42.493278  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.499468  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+[2026-08-27 05:18:10] INFO:     174.1.60.5:59984 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:18:18] INFO:     174.1.60.5:35200 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:18:25] INFO:     174.1.60.5:47324 - "POST /generate HTTP/1.1" 200 OK
+I0827 05:18:33.158722  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.159102  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.160380  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.167616  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.169173  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.173859  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.185045  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.185313  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.186507  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+[2026-08-27 05:18:33] INFO:     174.1.60.5:60276 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:18:40] INFO:     174.1.60.5:60282 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:18:40] INFO:     174.1.60.5:60286 - "GET /server_info HTTP/1.1" 200 OK
+I0827 05:18:33.192727  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+[2026-08-27 05:21:37] INFO:     174.1.60.5:39028 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:37] INFO:     174.1.60.5:39042 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:37] INFO:     174.1.60.5:39052 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:37] INFO:     174.1.60.5:39066 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:37] INFO:     174.1.60.5:39070 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:37] INFO:     174.1.60.5:39084 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:37] INFO:     174.1.60.5:39092 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:37] INFO:     174.1.60.5:39102 - "POST /generate HTTP/1.1" 200 OK
+I0827 05:21:47.564407  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:47.575759  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:47.585628  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:47.597601  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:47.599035  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:48.554785  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:48.565164  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:49.333541  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:49.343402  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:49.352372  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:49.361497  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:50.265654  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:50.274874  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:50.283715  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.110808  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.124187  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.162127  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.173472  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+[2026-08-27 05:21:51] INFO:     174.1.60.5:33930 - "POST /generate HTTP/1.1" 200 OK
+I0827 05:21:52.005925  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.016026  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.024489  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.033867  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.856267  1534 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+[2026-08-27 05:21:53] INFO:     174.1.60.5:33942 - "POST /generate HTTP/1.1" 200 OK
+I0827 05:21:52.867897  1521 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:53.712086  1469 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:53.724234  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:54.551172  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:54.564998  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:55.326069  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:55.338083  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+[2026-08-27 05:21:55] INFO:     174.1.60.5:33946 - "POST /generate HTTP/1.1" 200 OK
+I0827 05:21:56.186933  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:56.188442  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:56.200929  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:57.056439  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:57.056681  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:57.058399  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+[2026-08-27 05:21:57] INFO:     174.1.60.5:33950 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:57] INFO:     174.1.60.5:33960 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:57] INFO:     174.1.60.5:33968 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:59] INFO:     174.1.60.5:33980 - "POST /generate HTTP/1.1" 200 OK
+I0827 05:21:57.654704  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+[2026-08-27 05:22:00] INFO:     174.1.60.5:33990 - "POST /generate HTTP/1.1" 200 OK
+I0827 05:22:05.685521  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:05.687002  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:06.720834  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:07.458693  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:08.386162  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:08.387703  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+[2026-08-27 05:22:11] INFO:     174.1.60.5:51774 - "POST /generate HTTP/1.1" 200 OK
+I0827 05:22:08.395984  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:10.976925  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+[2026-08-27 05:22:13] INFO:     174.1.60.5:51778 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:15] INFO:     174.1.60.5:51794 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:17] INFO:     174.1.60.5:51800 - "POST /generate HTTP/1.1" 200 OK
+I0827 05:22:16.233013  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:17.990159  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:17.991626  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:17.999158  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:18.227823  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:18.250644  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+[2026-08-27 05:22:18] INFO:     174.1.60.5:51812 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:18] INFO:     174.1.60.5:51824 - "POST /generate HTTP/1.1" 200 OK
+I0827 05:22:18.360224  1557 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+[2026-08-27 05:22:20] INFO:     174.1.60.5:51834 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:21] INFO:     174.1.60.5:46458 - "POST /generate HTTP/1.1" 200 OK
+I0827 05:22:28.065888  1533 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+[2026-08-27 05:22:31] INFO:     174.1.60.5:36970 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:34] INFO:     174.1.60.5:36980 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:35] INFO:     174.1.60.5:36990 - "POST /generate HTTP/1.1" 200 OK
+I0827 05:22:36.003260  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+[2026-08-27 05:22:37] INFO:     174.1.60.5:36996 - "POST /generate HTTP/1.1" 200 OK
+I0827 05:22:36.977586  1468 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:38.816133  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+[2026-08-27 05:22:38] INFO:     174.1.60.5:37004 - "POST /generate HTTP/1.1" 200 OK
+I0827 05:22:38.914274  1551 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:39.463928  1501 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+[2026-08-27 05:22:39] INFO:     174.1.60.5:37012 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:40] INFO:     174.1.60.5:37020 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:42] INFO:     174.1.60.5:36076 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:52] INFO:     174.1.60.5:50774 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:54] INFO:     174.1.60.5:50786 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:56] INFO:     174.1.60.5:50788 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:58] INFO:     174.1.60.5:50790 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:59] INFO:     174.1.60.5:50806 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:23:00] INFO:     174.1.60.5:50818 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:23:01] INFO:     174.1.60.5:49800 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:23:03] INFO:     174.1.60.5:49808 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:23:20] INFO:     174.1.60.5:35588 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:23:20] INFO:     174.1.60.5:35590 - "GET /server_info HTTP/1.1" 200 OK
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/logs/manual_d_node1.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/logs/manual_d_node1.log
new file mode 100644
index 0000000..4ae8fab
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/logs/manual_d_node1.log
@@ -0,0 +1,1458 @@
+/usr/local/lib/python3.12/dist-packages/torchao/quantization/quant_api.py:1731: SyntaxWarning: invalid escape sequence '\.'
+  """Configuration class for applying different quantization configs to modules or parameters based on their fully qualified names (FQNs).
+/sgl-workspace/sglang/python/sglang/launch_server.py:56: UserWarning: 'python -m sglang.launch_server' is still supported, but 'sglang serve' is the recommended entrypoint.
+  Example: sglang serve --model-path <model> [options]
+  warnings.warn(
+[2026-08-27 04:35:58] KV cache is forced as chunk cache for decode server
+[2026-08-27 04:36:02] Breakable CUDA graph is incompatible with MLA attention; disabling prefill CUDA graph.
+[2026-08-27 04:36:02] Attention backend not specified. Use triton backend by default.
+[2026-08-27 04:36:05] server_args=ServerArgs(model_path='/data/hf_models/Kimi-K3', tokenizer_path='/data/hf_models/Kimi-K3', tokenizer_mode='auto', tokenizer_backend='huggingface', tokenizer_worker_num=1, detokenizer_worker_num=1, skip_tokenizer_init=False, load_format='auto', model_loader_extra_config='{}', trust_remote_code=True, context_length=None, is_embedding=False, enable_multimodal=None, revision=None, model_impl='auto', model_config_parser='auto', json_model_override_args='{}', dtype='auto', quantization=None, quantization_param_path=None, kv_cache_dtype='auto', enable_fp32_lm_head=False, modelopt_quant=None, modelopt_checkpoint_restore_path=None, modelopt_checkpoint_save_path=None, modelopt_export_path=None, quantize_and_serve=False, rl_quant_profile=None, enable_tf32_matmul=False, mem_fraction_static=0.88, max_running_requests=None, max_queued_requests=None, max_total_tokens=None, chunked_prefill_size=8192, enable_dynamic_chunking=False, max_prefill_tokens=16384, prefill_max_requests=None, schedule_policy='fcfs', enable_priority_scheduling=False, disable_priority_preemption=False, default_priority_value=None, abort_on_priority_when_disabled=False, schedule_low_priority_values_first=False, priority_scheduling_preemption_threshold=10, retraction_policy='length', schedule_conservativeness=1.0, page_size=64, swa_full_tokens_ratio=0.8, disable_hybrid_swa_memory=False, radix_eviction_policy='lru', prefill_only_disable_kv_cache=False, disable_radix_cache=True, enable_page_major_kv_layout=False, enable_unified_memory=False, disable_chunked_prefix_cache=False, disable_overlap_schedule=False, num_continuous_decode_steps=1, scheduler_recv_interval=1, enable_mixed_chunk=False, nccl_port=None, dist_timeout=3600, dist_init_addr='174.1.60.5:20000', nnodes=4, node_rank=1, tp_size=32, dcp_size=1, pp_size=1, pp_max_micro_batch_size=None, pp_async_batch_depth=0, dp_size=1, load_balance_method='round_robin', attn_cp_size=1, moe_dp_size=1, dcp_comm_backend='ag_rs', dcp_replicate_q_proj=None, dwdp_size=1, enable_prefill_cp=False, cp_strategy=None, enable_dsa_cache_layer_split=False, enable_dsa_prefill_context_parallel=False, dsa_prefill_cp_mode='round-robin-split', enable_prefill_context_parallel=False, prefill_cp_mode='in-seq-split', enable_cp_decode_attn_tp=False, enable_dp_attention=False, enable_dp_attention_local_control_broadcast=False, enable_dp_lm_head=False, enable_attn_tp_input_scattered=False, disable_attn_tp_gather=False, enable_p2p_check=False, device='cuda', base_gpu_id=0, gpu_id_step=1, random_seed=434112381, watchdog_timeout=300, soft_watchdog_timeout=None, sleep_on_idle=False, use_ray=False, custom_sigquit_handler=None, numa_node=None, gc_threshold=None, host='0.0.0.0', port=30000, fastapi_root_path='', smg_grpc_mode=False, grpc_mode=False, grpc_port=None, sidecar=None, sidecar_args=None, skip_server_warmup=False, warmups=None, enable_http2=False, ssl_keyfile=None, ssl_certfile=None, ssl_ca_certs=None, ssl_keyfile_password=None, enable_ssl_refresh=False, api_key=None, admin_api_key=None, served_model_name='kimi-k3', weight_version='default', chat_template=None, hf_chat_template_name=None, completion_template=None, file_storage_path='sglang_storage', enable_cache_report=False, reasoning_parser=None, default_chat_template_kwargs=None, strip_thinking_cache=False, enable_strict_thinking=False, tool_call_parser=None, tool_server=None, sampling_defaults='model', asr_max_buffer_seconds=60, asr_max_concurrent_sessions=32, preferred_sampling_params=None, allow_auto_truncate=False, stream_interval=1, batch_notify_size=16, stream_response_default_include_usage=False, incremental_streaming_output=False, enable_streaming_session=False, enable_session_radix_cache=False, log_level='info', log_level_http=None, log_requests=False, log_requests_level=2, log_requests_format='text', log_requests_target=None, uvicorn_access_log_exclude_prefixes=[], crash_dump_folder=None, show_time_cost=False, enable_metrics=False, smg_http_sidecar_port=None, enable_mfu_metrics=False, enable_metrics_for_all_schedulers=False, load_snapshot_publish_interval=15, tokenizer_metrics_custom_labels_header='x-custom-labels', tokenizer_metrics_allowed_custom_labels=None, extra_metric_labels=None, bucket_time_to_first_token=None, bucket_inter_token_latency=None, bucket_e2e_request_latency=None, prompt_tokens_buckets=None, generation_tokens_buckets=None, gc_warning_threshold_secs=0.0, decode_log_interval=40, enable_request_time_stats_logging=False, kv_events_config=None, enable_forward_pass_metrics=False, forward_pass_metrics_worker_id='', forward_pass_metrics_ipc_name=None, enable_trace=False, trace_modules='request', otlp_traces_endpoint='localhost:4317', export_metrics_to_file=False, export_metrics_to_file_dir=None, stat_loggers=None, constrained_json_whitespace_pattern=None, constrained_json_disable_any_whitespace=False, attention_backend='triton', decode_attention_backend=None, prefill_attention_backend=None, sampling_backend='flashinfer', grammar_backend='xgrammar', radix_cache_backend=None, mm_attention_backend=None, fp8_gemm_runner_backend='auto', fp4_gemm_runner_backend='auto', bf16_gemm_backend='auto', dsa_prefill_backend=None, dsa_decode_backend=None, dsa_paged_mqa_logits_backend='auto', dsa_topk_backend='sgl-kernel', disable_flashinfer_autotune=False, flashinfer_autotune_skip_ops=None, mamba_backend='triton', cuda_graph_config=CudaGraphConfig(decode=PhaseConfig(backend='full', max_bs=16, bs=[1, 2, 4, 8, 12, 16], tc_compiler='eager', full_prefill_max_req=None), prefill=PhaseConfig(backend='disabled', max_bs=2048, bs=[4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], tc_compiler='eager', full_prefill_max_req=None)), cuda_graph_backend_decode=None, cuda_graph_backend_prefill=None, cuda_graph_max_bs_decode=16, cuda_graph_max_bs_prefill=None, cuda_graph_bs_decode=None, cuda_graph_bs_prefill=None, cuda_graph_tc_compiler=None, disable_prefill_cuda_graph=False, disable_decode_cuda_graph=False, disable_cuda_graph=False, disable_cuda_graph_padding=False, enable_profile_cuda_graph=False, enable_cudagraph_gc=False, debug_cuda_graph=False, enable_layerwise_nvtx_marker=False, enable_nccl_nvls=False, enable_symm_mem=False, triton_attention_reduce_in_fp32=False, triton_attention_num_kv_splits=8, triton_attention_split_tile_size=None, flashinfer_mla_disable_ragged=False, enable_fused_qk_norm_rope=False, enable_precise_embedding_interpolation=False, enable_fused_moe_sum_all_reduce=False, enable_deepseek_v4_fp4_indexer=False, disable_custom_all_reduce=False, enable_mscclpp=False, enable_torch_symm_mem=False, enable_scattered_sconv=False, pre_warm_nccl=False, enable_quant_communications=False, enable_flashinfer_allreduce_fusion=False, enforce_disable_flashinfer_allreduce_fusion=False, flashinfer_allreduce_fusion_backend=None, enable_aiter_allreduce_fusion=False, enable_torch_compile=False, enable_torch_compile_debug_mode=False, torch_compile_max_bs=32, torchao_config='', speculative_algorithm=None, speculative_draft_model_path=None, speculative_draft_model_revision=None, speculative_draft_load_format=None, speculative_num_steps=None, speculative_eagle_topk=None, speculative_num_draft_tokens=None, speculative_dflash_block_size=None, speculative_dspark_block_size=None, speculative_dspark_sps_table_path=None, speculative_dspark_confidence_sts_path=None, speculative_dspark_align_verify_tokens_to_graph_tier=False, speculative_accept_threshold_single=1.0, speculative_accept_threshold_acc=1.0, speculative_use_rejection_sampling=False, speculative_token_map=None, speculative_attention_mode='prefill', speculative_draft_attention_backend=None, speculative_draft_window_size=None, speculative_moe_runner_backend='marlin', speculative_moe_a2a_backend=None, speculative_draft_model_quantization=None, speculative_skip_dp_mlp_sync=False, enable_multi_layer_eagle=False, speculative_adaptive=False, speculative_adaptive_config=None, decoupled_spec_bind_endpoint=None, decoupled_spec_connect_endpoints=None, decoupled_spec_rank=None, decoupled_spec_role='null', spec_trace_dir=None, speculative_ngram_min_bfs_breadth=1, speculative_ngram_max_bfs_breadth=10, speculative_ngram_match_type='BFS', speculative_ngram_max_trie_depth=18, speculative_ngram_capacity=10000000, speculative_ngram_external_corpus_path=None, speculative_ngram_external_sam_budget=0, speculative_ngram_external_corpus_max_tokens=10000000, ep_size=32, moe_a2a_backend='none', moe_runner_backend='marlin', flashinfer_mxfp4_moe_precision='default', deepep_mode='auto', fuseep_mode=2, deepep_dispatcher_output_dtype='auto', ep_num_redundant_experts=0, ep_dispatch_algorithm=None, init_expert_location='trivial', enable_eplb=False, eplb_algorithm='auto', eplb_rebalance_num_iterations=1000, eplb_rebalance_layers_per_chunk=None, eplb_min_rebalancing_utilization_threshold=1.0, expert_distribution_recorder_mode=None, expert_distribution_recorder_buffer_size=1000, enable_expert_distribution_metrics=False, deepep_config=None, moe_dense_tp_size=None, elastic_ep_backend=None, enable_elastic_expert_backup=False, mooncake_ib_device=None, enable_waterfill=False, ep_join_mode=None, ep_join_rank_offset=0, elastic_ep_initial_size=None, max_ep_size=None, elastic_ep_scale_timeout=600, elastic_ep_rejoin=False, disable_flashinfer_cutlass_moe_fp4_allgather=False, disable_shared_experts_fusion=False, enforce_shared_experts_fusion=False, max_mamba_cache_size=None, mamba_ssm_dtype=None, mamba_max_states_per_path=-1, enable_mamba_cache_stochastic_rounding=False, mamba_cache_philox_rounds=0, mamba_full_memory_ratio=0.21, mamba_radix_cache_strategy='extra_buffer_lazy', uses_mamba_radix_cache=False, mamba_track_interval=256, enable_int8_mamba_checkpoint=False, int8_mamba_ckpt_size=None, linear_attn_backend='triton', linear_attn_decode_backend=None, linear_attn_prefill_backend=None, linear_attn_verify_backend=None, enable_linear_replayssm=False, linear_replayssm_cache_len=16, enable_linear_replayssm_spec=False, enable_hierarchical_cache=False, hicache_ratio=2.0, hicache_size=0, hicache_write_policy='write_through', hicache_io_backend='kernel', hicache_mem_layout='page_first', hicache_storage_backend=None, hicache_storage_prefetch_policy='timeout', hicache_storage_backend_extra_config=None, enable_hisparse=False, hisparse_config=None, enable_broadcast_mm_inputs_process=False, enable_prefix_mm_cache=False, mm_enable_dp_encoder=False, mm_process_config={}, mm_processor_worker_num=0, mm_io_worker_num=0, limit_mm_data_per_request=None, enable_mm_global_cache=False, disable_fast_image_processor=False, mm_feature_transport='cpu', keep_mm_feature_on_device=False, enable_lora=None, enable_lora_overlap_loading=None, max_lora_rank=None, lora_target_modules=None, lora_paths=None, max_loaded_loras=None, max_loras_per_batch=8, lora_eviction_policy='lru', lora_backend='csgmv', max_lora_chunk_size=16, experts_shared_outer_loras=None, lora_use_virtual_experts=False, lora_strict_loading=False, lora_drain_wait_threshold=0.0, enable_two_batch_overlap=False, enable_single_batch_overlap=False, tbo_token_distribution_threshold=0.48, cpu_offload_gb=0, offload_group_size=-1, offload_num_in_group=1, offload_prefetch_step=1, offload_mode='cpu', enable_lmcache=False, lmcache_config_file=None, enable_flexkv=False, flexkv_config_file=None, kt_weight_path=None, kt_method='AMXINT4', kt_cpuinfer=None, kt_threadpool_count=2, kt_num_gpu_experts=None, kt_max_deferred_experts_per_token=None, dllm_algorithm=None, dllm_algorithm_config=None, dllm_fdfo=True, disaggregation_mode='decode', disaggregation_transfer_backend='mooncake', disaggregation_bootstrap_port=28800, disaggregation_ib_device='mlx5_0,mlx5_1,mlx5_2,mlx5_3', disaggregation_decode_enable_radix_cache=False, disaggregation_decode_enable_offload_kvcache=False, num_reserved_decode_tokens=512, disaggregation_decode_extra_slots=0, disaggregation_decode_polling_interval=1, optimistic_prefill_attempts=0, encoder_only=False, language_only=False, encoder_transfer_backend='zmq_to_tokenizer', encoder_urls=[], encoder_bootstrap_port=8997, encoder_register_urls=[], enable_adaptive_dispatch_to_encoder=False, enable_pdmux=False, pdmux_config_path=None, sm_group_num=8, custom_weight_loader=[], weight_loader_disable_mmap=False, weight_loader_prefetch_checkpoints=False, weight_loader_prefetch_num_threads=4, weight_loader_drop_cache_after_load=False, remote_instance_weight_loader_seed_instance_ip=None, remote_instance_weight_loader_seed_instance_service_port=None, remote_instance_weight_loader_send_weights_group_ports=None, remote_instance_weight_loader_backend='nccl', remote_instance_weight_loader_start_seed_via_transfer_engine=False, engine_info_bootstrap_port=6789, modelexpress_config=None, download_dir=None, model_checksum=None, delete_ckpt_after_loading=False, decrypted_config_file=None, decrypted_draft_config_file=None, checkpoint_engine_wait_weights_before_ready=False, enable_prefill_delayer=False, prefill_delayer_max_delay_passes=30, prefill_delayer_token_usage_low_watermark=None, prefill_delayer_forward_passes_buckets=None, prefill_delayer_wait_seconds_buckets=None, prefill_delayer_queue_min_ratio=None, prefill_delayer_max_delay_ms=None, min_free_slots_delay=None, enable_deterministic_inference=False, rl_on_policy_target=None, kv_canary='none', kv_canary_real_data='none', kv_canary_sweep_interval=0, enable_dynamic_batch_tokenizer=False, dynamic_batch_tokenizer_batch_size=32, dynamic_batch_tokenizer_batch_timeout=0.002, enable_tokenizer_batch_encode=False, disable_tokenizer_batch_decode=False, debug_tensor_dump_output_folder=None, debug_tensor_dump_layers=None, debug_tensor_dump_input_file=None, enable_memory_saver=False, enable_weights_cpu_backup=False, enable_draft_weights_cpu_backup=False, enable_custom_logit_processor=False, enable_return_hidden_states=False, enable_return_routed_experts=False, enable_return_indexer_topk=False, disable_outlines_disk_cache=False, enable_mis=False, weight_cache_mode='off', weight_cache_socket=None, weight_cache_timeout=1800, forward_hooks=None, msprobe_dump_config=None)
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:22 TP15 EP15] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:22 TP15 EP15] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:23 TP9 EP9] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:23 TP9 EP9] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:23 TP12 EP12] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:23 TP12 EP12] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:23 TP13 EP13] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:23 TP13 EP13] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:23 TP14 EP14] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:23 TP14 EP14] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:23 TP8 EP8] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:23 TP8 EP8] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:23 TP10 EP10] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:23 TP10 EP10] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:23 TP11 EP11] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:23 TP11 EP11] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:24 TP15 EP15] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:24 TP9 EP9] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:24 TP13 EP13] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:24 TP12 EP12] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:24 TP14 EP14] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:25 TP10 EP10] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:25 TP8 EP8] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:25 TP11 EP11] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:26.087226   242 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:26.087333   242 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:26.087350   242 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.6 port: 12001
+I0827 04:36:26.087409   242 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.6:16298
+I0827 04:36:26.087569   242 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:26.095995   236 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:26.096091   236 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:26.096107   236 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.6 port: 12001
+I0827 04:36:26.096150   236 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.6:15958
+I0827 04:36:26.096269   236 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:26.096457   240 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:26.096560   240 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:26.096577   240 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.6 port: 12001
+I0827 04:36:26.096626   240 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.6:16447
+I0827 04:36:26.096773   240 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:26.096876   241 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:26.096975   241 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:26.096992   241 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.6 port: 12001
+I0827 04:36:26.097035   241 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.6:16035
+I0827 04:36:26.097170   241 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:26.103180   241 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:26.107820   240 topology.cpp:127] Device mlx5_0 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:26.109326   239 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:26.109416   239 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:26.109428   239 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.6 port: 12001
+I0827 04:36:26.109464   239 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.6:16843
+I0827 04:36:26.109562   239 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:26.114297   242 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:26.114653   239 topology.cpp:127] Device mlx5_0 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:26.115540   237 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:26.115630   237 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:26.115645   237 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.6 port: 12001
+I0827 04:36:26.115686   237 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.6:15544
+I0827 04:36:26.115813   237 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:26.118220   240 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:26.121481   242 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:26.122418   237 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:26.126497   236 topology.cpp:127] Device mlx5_0 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:26.126574   235 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:26.126684   235 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:26.126703   235 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.6 port: 12001
+I0827 04:36:26.126758   235 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.6:15006
+I0827 04:36:26.126933   235 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:26.127116   240 topology.cpp:127] Device mlx5_2 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:26.127162   238 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:26.127270   238 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:26.127285   238 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.6 port: 12001
+I0827 04:36:26.127342   238 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.6:16140
+I0827 04:36:26.127489   238 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:26.128777   237 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:26.129559   242 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:26.130432   241 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:26.133821   238 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:26.134943   235 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:26.135799   240 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:26.136080   237 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:26.138710   241 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:26.139173   242 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:26.139978   240 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:26.140033   240 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:26.140090   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.140504   235 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:26.142936   242 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:26.142961   242 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:26.143014   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.144655   237 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:26.146834   235 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:26.147805   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:26.147867   241 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:26.148357   239 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:26.148370   237 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:26.148394   237 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:26.148444   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.149503   240 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:06
+I0827 04:36:26.149536   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.150290   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:26.150381   236 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:26.151773   242 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:06
+I0827 04:36:26.151803   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.152055   241 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:26.152087   241 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:26.152148   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.153594   235 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:26.156044   235 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:26.156075   235 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:26.156124   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.157910   238 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:26.159448   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:26.159547   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:26.160912   242 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:06
+I0827 04:36:26.160940   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.161144   241 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:06
+I0827 04:36:26.161176   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.162882   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:26.164348   235 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:06
+I0827 04:36:26.164378   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.165159   238 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:26.168772   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:26.170400   241 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:06
+I0827 04:36:26.170428   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.172459   238 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:26.174376   239 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:26.175243   238 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:26.175272   238 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:26.175325   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.178375   236 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:26.178763   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:26.178785   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:26.180411   237 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:06
+I0827 04:36:26.180462   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.180604   240 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:06
+I0827 04:36:26.180672   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.181643   239 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:26.182818   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:26.184404   238 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:06
+I0827 04:36:26.184437   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.184495   239 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:26.184537   239 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:26.184595   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.185601   236 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:26.186753   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:26.188311   242 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:06
+I0827 04:36:26.188344   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.188390   236 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:26.188416   236 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:26.188470   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.190752   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:26.192051   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:26.192406   235 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:06
+I0827 04:36:26.192444   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.193856   240 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:06
+I0827 04:36:26.193897   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.194197   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:26.195964   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:26.195997   238 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:06
+I0827 04:36:26.196039   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.196040   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:26.197574   242 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:06
+I0827 04:36:26.197599   236 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:06
+I0827 04:36:26.197629   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.197691   242 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:26 TP15 EP15] Init torch distributed begin.
+I0827 04:36:26.198544   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:26.200238   241 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:06
+I0827 04:36:26.200268   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.204000   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:26.205737   238 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:06
+I0827 04:36:26.205766   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.210784   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:26.212373   237 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:06
+I0827 04:36:26.212417   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.218751   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:26.220386   235 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:06
+I0827 04:36:26.220422   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.222854   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:26.223088   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:26.223217   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:26.223345   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:26.224500   241 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:06
+I0827 04:36:26.224536   240 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:06
+I0827 04:36:26.224612   241 transfer_engine_impl.cpp:388] installTransport, type=rdma
+I0827 04:36:26.224682   240 transfer_engine_impl.cpp:388] installTransport, type=rdma
+I0827 04:36:26.224973   236 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:06
+I0827 04:36:26.225018   236 rdma_context.cpp:205] Using SIEVE endpoint store
+[2026-08-27 04:36:26 TP14 EP14] Init torch distributed begin.
+[2026-08-27 04:36:26 TP13 EP13] Init torch distributed begin.
+I0827 04:36:26.225459   239 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:06
+I0827 04:36:26.225523   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.230858   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:26.232468   238 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:06
+I0827 04:36:26.232614   238 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:26 TP11 EP11] Init torch distributed begin.
+I0827 04:36:26.233065   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:26.234685   236 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:06
+I0827 04:36:26.234727   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.238961   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:26.240710   237 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:06
+I0827 04:36:26.240758   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.242511   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:26.244089   236 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:06
+I0827 04:36:26.244236   236 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:26 TP9 EP9] Init torch distributed begin.
+I0827 04:36:26.247059   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:26.248764   235 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:06
+I0827 04:36:26.248920   235 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:26 TP8 EP8] Init torch distributed begin.
+I0827 04:36:26.250787   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:26.252344   239 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:06
+I0827 04:36:26.252393   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.266752   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:26.268360   237 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:06
+I0827 04:36:26.268492   237 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:26 TP10 EP10] Init torch distributed begin.
+I0827 04:36:26.278649   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:26.280236   239 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:06
+I0827 04:36:26.280287   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:26.306850   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:26.308480   239 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:06
+I0827 04:36:26.308605   239 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:26 TP12 EP12] Init torch distributed begin.
+[2026-08-27 04:36:48 TP14 EP14] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP15 EP15] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP8 EP8] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP11 EP11] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP9 EP9] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP13 EP13] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP10 EP10] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP12 EP12] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP15 EP15] Init torch distributed ends. elapsed=22.03 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP14 EP14] Init torch distributed ends. elapsed=22.01 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP13 EP13] Init torch distributed ends. elapsed=22.01 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP12 EP12] Init torch distributed ends. elapsed=21.92 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP11 EP11] Init torch distributed ends. elapsed=22.00 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP9 EP9] Init torch distributed ends. elapsed=21.99 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP10 EP10] Init torch distributed ends. elapsed=21.96 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP8 EP8] Init torch distributed ends. elapsed=21.98 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP14 EP14] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP15 EP15] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP9 EP9] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP13 EP13] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP8 EP8] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP10 EP10] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP12 EP12] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP14 EP14] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP15 EP15] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP9 EP9] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP10 EP10] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP13 EP13] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP8 EP8] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP12 EP12] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP11 EP11] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP11 EP11] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP13 EP13] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP14 EP14] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP15 EP15] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP12 EP12] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP10 EP10] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP8 EP8] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP9 EP9] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP13 EP13] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP12 EP12] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP14 EP14] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP15 EP15] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP10 EP10] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP8 EP8] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP9 EP9] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP11 EP11] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP11 EP11] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:50 TP8 EP8] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP9 EP9] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP11 EP11] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP12 EP12] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP13 EP13] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP14 EP14] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP10 EP10] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP15 EP15] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:38:57 TP14 EP14] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:57 TP14 EP14] Load weight end. elapsed=129.27 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP9 EP9] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP9 EP9] Load weight end. elapsed=129.52 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP10 EP10] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP10 EP10] Load weight end. elapsed=129.56 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP8 EP8] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP8 EP8] Load weight end. elapsed=129.57 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP13 EP13] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP13 EP13] Load weight end. elapsed=129.58 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP11 EP11] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP11 EP11] Load weight end. elapsed=129.64 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP15 EP15] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP15 EP15] Load weight end. elapsed=129.77 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:58 TP12 EP12] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:58 TP12 EP12] Load weight end. elapsed=129.90 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:39:01 TP14 EP14] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP14 EP14] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP11 EP11] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP11 EP11] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP8 EP8] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP8 EP8] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP13 EP13] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP13 EP13] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP9 EP9] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP9 EP9] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP15 EP15] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP15 EP15] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP12 EP12] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP12 EP12] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP10 EP10] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP10 EP10] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:02 TP14 EP14] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP15 EP15] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP12 EP12] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP13 EP13] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP10 EP10] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP9 EP9] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP11 EP11] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP8 EP8] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP15 EP15] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP12 EP12] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP10 EP10] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP8 EP8] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP11 EP11] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP13 EP13] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP9 EP9] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP14 EP14] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP9 EP9] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP12 EP12] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP9 EP9] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP10 EP10] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP12 EP12] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP13 EP13] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP15 EP15] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP10 EP10] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP11 EP11] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP13 EP13] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP14 EP14] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP15 EP15] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP11 EP11] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP8 EP8] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP14 EP14] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP8 EP8] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP14 EP14] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP14 EP14] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP9 EP9] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP9 EP9] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP12 EP12] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP12 EP12] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:39:02 TP10 EP10] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP10 EP10] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP8 EP8] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP8 EP8] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:39:02 TP11 EP11] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP11 EP11] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:39:02 TP13 EP13] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP13 EP13] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP15 EP15] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP15 EP15] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:42:56 TP8 EP8] Capture target decode CUDA graph end. elapsed=233.78 s, mem usage=0.60 GB, avail mem=8.52 GB.
+[2026-08-27 04:42:56 TP13 EP13] Capture target decode CUDA graph end. elapsed=233.78 s, mem usage=0.58 GB, avail mem=8.58 GB.
+[2026-08-27 04:42:56 TP8 EP8] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP13 EP13] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP14 EP14] Capture target decode CUDA graph end. elapsed=233.81 s, mem usage=0.56 GB, avail mem=8.59 GB.
+[2026-08-27 04:42:56 TP10 EP10] Capture target decode CUDA graph end. elapsed=233.79 s, mem usage=0.56 GB, avail mem=8.59 GB.
+[2026-08-27 04:42:56 TP14 EP14] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP12 EP12] Capture target decode CUDA graph end. elapsed=233.80 s, mem usage=0.60 GB, avail mem=8.52 GB.
+[2026-08-27 04:42:56 TP10 EP10] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP12 EP12] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP11 EP11] Capture target decode CUDA graph end. elapsed=233.80 s, mem usage=0.62 GB, avail mem=8.51 GB.
+[2026-08-27 04:42:56 TP11 EP11] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP9 EP9] Capture target decode CUDA graph end. elapsed=233.91 s, mem usage=0.58 GB, avail mem=8.58 GB.
+[2026-08-27 04:42:56 TP9 EP9] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP15 EP15] Capture target decode CUDA graph end. elapsed=233.94 s, mem usage=0.62 GB, avail mem=8.51 GB.
+[2026-08-27 04:42:56 TP15 EP15] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+W0827 04:42:57.337543  2432 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5972aec0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.337602  2433 memory_location.cpp:73] Failed to get NUMA node, addr: 0x594a1440, len: 23168: Operation not permitted [1]
+W0827 04:42:57.337656  2434 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59f51680, len: 23168: Operation not permitted [1]
+W0827 04:42:57.337674  2435 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59ebed40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.337788  2436 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5b797200, len: 185344: Operation not permitted [1]
+W0827 04:42:57.337911  2437 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5b7c4640, len: 185344: Operation not permitted [1]
+W0827 04:42:57.337891  2438 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59639800, len: 23168: Operation not permitted [1]
+W0827 04:42:57.337962  2439 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59435700, len: 23168: Operation not permitted [1]
+W0827 04:42:57.338008  2440 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59660c80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.338057  2441 memory_location.cpp:73] Failed to get NUMA node, addr: 0x594c9400, len: 46336: Operation not permitted [1]
+W0827 04:42:57.338071  2442 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59bb1f00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.338107  2443 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59f5fc80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.338161  2445 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5953b700, len: 23168: Operation not permitted [1]
+W0827 04:42:57.338165  2444 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59812080, len: 23168: Operation not permitted [1]
+W0827 04:42:57.338204  2446 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5baddac0, len: 185344: Operation not permitted [1]
+W0827 04:42:57.338325  2448 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59994940, len: 23168: Operation not permitted [1]
+W0827 04:42:57.338340  2447 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5bb0af00, len: 185344: Operation not permitted [1]
+W0827 04:42:57.338538  2449 memory_location.cpp:73] Failed to get NUMA node, addr: 0x45ba2780, len: 46336: Operation not permitted [1]
+W0827 04:42:57.338596  2450 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59ac5f00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.338649  2451 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59984000, len: 23168: Operation not permitted [1]
+W0827 04:42:57.381100  2729 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59997640, len: 23168: Operation not permitted [1]
+W0827 04:42:57.381177  2728 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59a855c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.381280  2730 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59953e40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.381347  2732 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5b87eac0, len: 185344: Operation not permitted [1]
+W0827 04:42:57.381466  2731 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59971000, len: 23168: Operation not permitted [1]
+W0827 04:42:57.381542  2733 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5b8abf00, len: 185344: Operation not permitted [1]
+W0827 04:42:57.381620  2734 memory_location.cpp:73] Failed to get NUMA node, addr: 0x599b8540, len: 23168: Operation not permitted [1]
+W0827 04:42:57.381628  2735 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5927e3c0, len: 46336: Operation not permitted [1]
+W0827 04:42:57.381863  2736 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59900b40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.381878  2737 memory_location.cpp:73] Failed to get NUMA node, addr: 0x598b4f80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.421048  2876 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58bb5680, len: 23168: Operation not permitted [1]
+W0827 04:42:57.421196  2878 memory_location.cpp:73] Failed to get NUMA node, addr: 0x592f7a00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.421193  2877 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59321cc0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.421262  2881 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ab93340, len: 185344: Operation not permitted [1]
+W0827 04:42:57.421321  2879 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5929c2c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.421382  2880 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ab65f00, len: 185344: Operation not permitted [1]
+W0827 04:42:57.421423  2882 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58aa4dc0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.421440  2883 memory_location.cpp:73] Failed to get NUMA node, addr: 0x585179c0, len: 46336: Operation not permitted [1]
+W0827 04:42:57.421434  2884 memory_location.cpp:73] Failed to get NUMA node, addr: 0x588b0240, len: 23168: Operation not permitted [1]
+W0827 04:42:57.421594  2885 memory_location.cpp:73] Failed to get NUMA node, addr: 0x588d0800, len: 23168: Operation not permitted [1]
+W0827 04:42:57.423333  2900 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58bc9040, len: 23168: Operation not permitted [1]
+W0827 04:42:57.423418  2903 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58a57e40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.423974  2913 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ae07a80, len: 185344: Operation not permitted [1]
+W0827 04:42:57.424072  2911 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5adda640, len: 185344: Operation not permitted [1]
+W0827 04:42:57.424093  2908 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58f8c900, len: 23168: Operation not permitted [1]
+W0827 04:42:57.424189  2924 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58d795c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.424269  2906 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58e657c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.424296  2915 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58c46ac0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.424311  2922 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58e39680, len: 23168: Operation not permitted [1]
+W0827 04:42:57.424809  2930 memory_location.cpp:73] Failed to get NUMA node, addr: 0x591a7f40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.424866  2932 memory_location.cpp:73] Failed to get NUMA node, addr: 0x593dee00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.425096  2934 memory_location.cpp:73] Failed to get NUMA node, addr: 0x596af5c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.425186  2935 memory_location.cpp:73] Failed to get NUMA node, addr: 0x595b2ac0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.425271  2938 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5b4f5f80, len: 185344: Operation not permitted [1]
+W0827 04:42:57.425292  2937 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5b4c8b40, len: 185344: Operation not permitted [1]
+W0827 04:42:57.425374  2939 memory_location.cpp:73] Failed to get NUMA node, addr: 0x594e07c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.425433  2942 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58ec6dc0, len: 46336: Operation not permitted [1]
+W0827 04:42:57.425568  2944 memory_location.cpp:73] Failed to get NUMA node, addr: 0x596a8180, len: 23168: Operation not permitted [1]
+W0827 04:42:57.425750  2919 memory_location.cpp:73] Failed to get NUMA node, addr: 0x45850900, len: 46336: Operation not permitted [1]
+W0827 04:42:57.425791  2946 memory_location.cpp:73] Failed to get NUMA node, addr: 0x592c6f00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.427150  2998 memory_location.cpp:73] Failed to get NUMA node, addr: 0x599c15c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.427410  2989 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59750440, len: 23168: Operation not permitted [1]
+W0827 04:42:57.427572  3010 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5bad0cc0, len: 185344: Operation not permitted [1]
+W0827 04:42:57.427641  3014 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a39cd80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.427647  3024 memory_location.cpp:73] Failed to get NUMA node, addr: 0x464ff140, len: 23168: Operation not permitted [1]
+W0827 04:42:57.427655  3002 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a377380, len: 23168: Operation not permitted [1]
+W0827 04:42:57.427663  2993 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59876080, len: 23168: Operation not permitted [1]
+W0827 04:42:57.427681  3007 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5baa3880, len: 185344: Operation not permitted [1]
+W0827 04:42:57.427703  3025 memory_location.cpp:73] Failed to get NUMA node, addr: 0x463cd480, len: 23168: Operation not permitted [1]
+W0827 04:42:57.427706  3018 memory_location.cpp:73] Failed to get NUMA node, addr: 0x464e6b00, len: 46336: Operation not permitted [1]
+W0827 04:42:57.586287  3470 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a4fe500, len: 23168: Operation not permitted [1]
+W0827 04:42:57.586316  3471 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a8129c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.586287  3468 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a530a00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.586329  3469 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a72a280, len: 23168: Operation not permitted [1]
+W0827 04:42:57.586364  3472 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5c756140, len: 185344: Operation not permitted [1]
+W0827 04:42:57.586422  3474 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a287d00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.586429  3473 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5c783580, len: 185344: Operation not permitted [1]
+W0827 04:42:57.586477  3476 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a731080, len: 23168: Operation not permitted [1]
+W0827 04:42:57.586475  3475 memory_location.cpp:73] Failed to get NUMA node, addr: 0x46a69900, len: 46336: Operation not permitted [1]
+W0827 04:42:57.586767  3477 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a2cd580, len: 23168: Operation not permitted [1]
+[2026-08-27 04:42:59] Dummy health check server started in background thread at 0.0.0.0:30000
+I0827 04:48:13.995834  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:13.996141  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.010807  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.011080  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.031366  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.039373  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.042748  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.043025  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.057039  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.060467  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.074244  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.093904  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.097144  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.097398  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.100539  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.103674  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.106997  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.107286  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.107595  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.107802  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.108084  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.108319  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.111485  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.114651  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.114880  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.118037  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.127089  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.130350  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.136305  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.145159  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.145349  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.145601  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.148783  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.161775  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.162027  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.165246  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.165525  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.165712  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.171312  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.182595  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.182830  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.183091  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.183334  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.197396  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.202782  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.202994  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.211761  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.212010  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.212227  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.212466  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.218309  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.218585  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.218844  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.219072  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.219323  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.221719  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.223042  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.230329  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.230581  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.233455  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.235435  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.239619  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.239846  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.243142  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.248512  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.248768  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.249244  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.249539  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.249773  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.257058  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.259995  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.260232  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.260426  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.260643  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.263525  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.263789  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.264045  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.270218  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.272343  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.272601  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.272828  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.273056  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.273649  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.273857  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.274083  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.274338  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.274350  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.274650  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.274857  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.280273  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.281651  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.283277  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.288719  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.291597  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.291850  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.294591  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.297427  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.297646  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.297865  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.298131  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.298350  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.298571  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.300019  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.300335  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.301384  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.303872  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.304067  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.306974  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.307229  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.307489  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.308355  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.311895  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.312120  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.312359  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.312578  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.313377  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.317476  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.317679  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.319007  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.323084  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.323379  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.323665  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.333745  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.333979  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.334159  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.334322  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.337862  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.351508  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.351702  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.351871  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.352072  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.352252  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.355370  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.364216  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.367312  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.367552  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.367769  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.368913  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.375226  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.375468  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.375686  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.375947  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.376187  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.376428  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.376683  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.376904  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.377125  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.380695  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.382211  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.384042  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.384289  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.384522  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.390187  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.390600  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.391651  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.392228  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.398043  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.404222  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.404719  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.404901  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.405130  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.405673  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.405961  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.406175  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.406375  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.406561  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.406761  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.406991  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.407531  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.411696  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.411933  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.412912  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.416914  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.417143  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.417341  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.418166  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.418262  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.423719  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.425009  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.431804  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.432065  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.432313  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.432547  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.432744  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.432958  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.433151  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.433398  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.433662  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.433943  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.435259  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.439250  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.440708  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.443462  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.444864  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.446599  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.453094  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.457774  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.458002  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.458236  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.461966  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.463891  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.464155  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.465241  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.467888  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.468128  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.472088  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.473263  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.474021  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.474298  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.475373  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.479022  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.479259  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.479537  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.479795  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.480019  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.480264  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.480456  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.480696  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.480752  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.480891  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.481091  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.481326  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.481606  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.481833  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.482064  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.482175  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.482331  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.482553  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.482764  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.491822  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.493273  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.497345  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.498744  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.501601  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.502969  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.502988  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.509712  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.516039  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.516237  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.516439  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.519500  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.519733  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.519950  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.520223  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.520452  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.520711  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.520921  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.521140  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.521364  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.521629  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.521858  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.522101  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.522328  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.522553  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.522855  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.523010  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.523072  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.523293  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.523304  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.524191  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.524235  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.528468  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.535353  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.535600  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.535769  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.535825  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.536702  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.537087  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.542068  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.543424  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.545435  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.546203  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.546453  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.546715  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.546816  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.546933  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.547155  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.547389  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.547611  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.547857  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.548101  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.551676  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.553035  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.553618  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.554762  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.555172  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.556280  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.558871  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.565316  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.568676  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.569031  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.569290  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.569475  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.570151  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.572939  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.574455  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.575088  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.575302  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.576429  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.578527  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.578720  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.580932  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.582515  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.583039  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.583253  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.583478  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.583778  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.583832  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.584945  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.585202  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.586391  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.591900  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.593173  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.593271  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.593396  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.593664  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.596908  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.601369  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.602131  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.602725  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.604811  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.605091  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.606220  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.608003  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.610472  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.610689  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.610699  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.610924  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.611121  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.611315  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.611521  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.611792  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.612001  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.612202  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.612247  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.612334  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.612481  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.613652  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.615125  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.619858  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.620216  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.621151  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.624950  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.625170  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.625332  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.625414  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.625568  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.625614  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.625829  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.626011  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.626319  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.631613  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.632304  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.632570  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.632856  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.633095  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.633306  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.633574  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.634308  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.637017  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.637279  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.637583  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.638610  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.639593  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.640806  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.640900  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.641085  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.643601  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.644850  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.645069  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.646034  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.646308  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.646575  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.647399  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.648111  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.649897  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.651033  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.651146  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.653847  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.655079  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.659225  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.662467  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.663832  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.664858  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.665107  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.665309  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.666141  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.669166  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.673557  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.673764  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.673919  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.674009  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.674209  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.674503  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.674647  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.674710  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.674896  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.674955  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.675308  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.675336  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.675500  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.675552  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.675832  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.676005  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.676035  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.676223  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.676425  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.676607  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.680454  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.681128  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.685020  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.685290  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.686386  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.686959  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.694617  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.694845  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.695986  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.697674  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.698963  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.699177  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.699368  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.701349  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.701767  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.702576  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.702847  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.703204  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.706663  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.706753  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.706854  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.707039  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.707223  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.707465  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.707770  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.708069  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.708083  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.708248  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.708472  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.708722  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.708920  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.709123  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.710752  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.710954  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.712257  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.719622  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.719846  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.720033  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.720993  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.725260  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.725533  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.725705  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.725924  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.726168  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.726495  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.726595  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.727191  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.728030  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.728752  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.732066  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.734992  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.740563  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.746973  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.750195  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.750399  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.750788  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.752028  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.753922  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.756233  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.756840  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.759518  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.760934  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.766074  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.767369  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.769282  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.769475  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.769667  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.770648  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.774008  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.774199  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.774485  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.774672  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.774865  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.775053  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.775226  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.777894  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.778448  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.778666  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.779709  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.780627  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.781956  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.783814  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.785252  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.785427  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.785633  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.786656  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.787170  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.792593  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.797327  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.797559  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.797796  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.798789  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.801191  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.804433  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.804607  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.804807  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.805258  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.805953  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.806377  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.810664  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.811244  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.812598  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.824426  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.824638  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.824924  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.825793  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.831593  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.833058  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.833375  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.833634  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.833873  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.834122  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.834323  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.834601  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.834823  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.835043  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.835248  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.835427  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.835631  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.835824  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.836011  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.838608  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.838699  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.854580  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.858219  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.859412  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.862950  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.864025  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.865686  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.865891  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.866111  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.866333  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.866657  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.866854  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.867084  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.867278  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.867471  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.867661  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.872786  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.876446  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.876659  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.876896  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.877848  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.880227  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.880436  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.883129  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.883190  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.883363  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.883567  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.884621  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.884654  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.888731  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.926294  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.926565  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.926801  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.926995  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.927213  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.927462  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.927661  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.927874  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.937942  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.938158  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.938393  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.939355  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.949986  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.950210  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.950419  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.951480  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.985823  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.986092  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.986321  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.986529  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.986739  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.986924  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.987135  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.987387  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.987639  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.987892  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.988083  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.988299  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.988473  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.988667  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.012990  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.013230  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.013480  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.013671  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.013873  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.014089  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.014310  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.014530  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.014705  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.014922  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.029297  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.029497  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.029675  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.029886  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.030081  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.030721  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.075974  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.076195  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.076392  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.076560  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.076789  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.076989  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.077189  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.077405  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.086928  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.087107  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.087375  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.088318  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.098666  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.098937  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.099119  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.100116  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.134526  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.134721  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.134938  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.135179  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.135361  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.135597  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.135814  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.136029  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.136237  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.136453  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.136703  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.136907  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.137085  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.139734  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.164215  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.166940  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.167219  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.167455  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.167656  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.167850  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.168042  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.168241  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.168471  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.168686  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.168931  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.169119  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.183355  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.183578  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.183832  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.184046  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.184273  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.184649  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.231909  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.232179  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.232437  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.232667  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.232929  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.233191  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.233395  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.236016  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.241250  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.243832  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.244118  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.246883  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.252156  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.254787  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.257608  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.265419  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.273245  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.283512  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.283768  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.286342  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.291661  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.291878  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.292133  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.292326  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.292569  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.292784  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.293030  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.298199  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.298434  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.298682  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.311589  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.311841  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.319455  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.319670  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.319875  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.320118  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.320379  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.323133  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.323354  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.323570  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.331285  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.336499  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.336781  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.336977  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.337196  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.337783  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.375766  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.375975  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.376159  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.376356  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.376547  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.376745  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.376950  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.379487  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.384498  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.387058  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.387269  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.389864  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.394851  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.397480  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.397692  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.407253  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.424064  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.424232  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.424441  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.424628  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.432085  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.432324  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.432541  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.432767  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.432996  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.433199  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.433389  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.438321  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.438519  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.438731  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.450695  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.450888  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.458081  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.458277  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.458539  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.458791  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.458984  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.461611  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.461808  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.462006  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.469441  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.474292  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.474494  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.474697  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.474973  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.475522  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.457798  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.458177  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.459460  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.463181  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.464798  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.469339  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.469461  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.470870  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.475363  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.479321  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.480788  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.481803  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.483294  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.485772  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.494769  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.496369  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.500792  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.299279  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.299625  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.301414  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.309536  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.311267  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.312475  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.314007  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.319712  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.324035  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.325511  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.330739  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.330972  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.331518  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.332414  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.337031  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.339011  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.339269  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.340338  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.342085  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.343655  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.344139  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.345620  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.348776  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.349041  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.349591  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.350200  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.360395  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.360633  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.361879  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.362035  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.362210  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.363549  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.366465  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.370512  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.371948  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.373085  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.373347  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.374653  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.376161  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.378137  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.378718  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.378968  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.379657  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.384268  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.385306  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.385583  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.386750  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.386739  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.386987  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.388157  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.394881  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.398934  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.400285  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.402129  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.402400  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.403517  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.406072  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.406335  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.406554  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.406766  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.407377  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.416814  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.417057  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.418246  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.426201  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.426426  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.426638  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.426827  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.427093  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.427573  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.430253  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.430498  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.431689  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.391251  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.393167  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.399513  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.593394  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.593744  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.595059  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.599748  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.607136  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.608918  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.613543  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:40.735651  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:40.737267  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:40.743311  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:40.749338  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:40.750874  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:48.322273  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:48.323827  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:48.330022  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:55.890759  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:55.891063  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:55.892275  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:55.900822  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:55.902331  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:55.907302  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:55.915869  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:55.917403  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.662242  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.663878  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.676903  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.677158  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.677393  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.678658  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.690680  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.691715  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.692193  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.693166  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.699034  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.706213  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.707717  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.708302  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.710057  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.710531  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.725970  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.726244  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.727491  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.460754  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.462435  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.468483  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.477201  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.477489  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.478799  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.514531  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.514875  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.516037  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.170460  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.172108  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.178090  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.187213  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.187461  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.188678  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.201197  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.201504  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.202839  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.215639  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.215888  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.217134  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:47.594174  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:47.607700  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:48.578758  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:48.588865  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:48.597946  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:49.367239  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:49.380311  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:50.292033  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:50.302100  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:50.303560  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:50.312023  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:50.320772  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.137686  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.152026  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.162048  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.026280  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.032133  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.038511  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.041432  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.047593  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.050129  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.058367  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.826843  1342 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.838438  1356 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.847219  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:53.739216  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:54.582928  1340 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:55.353269  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:55.481009  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:56.209721  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:57.074601  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:57.089560  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:57.620945  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:57.639796  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:57.643188  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:57.764745  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:05.707906  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:05.720738  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:06.742728  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:06.752833  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:07.478389  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:08.273787  1341 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:09.240689  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:09.253865  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:10.164781  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:10.170979  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:11.012079  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:11.870416  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:18.023379  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:18.250713  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:18.926174  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:30.803402  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:30.804934  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:32.463899  1332 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:52.158495  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:56.508555  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:56.518028  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:58.313472  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/logs/manual_d_node2.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/logs/manual_d_node2.log
new file mode 100644
index 0000000..cf2064d
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/logs/manual_d_node2.log
@@ -0,0 +1,1470 @@
+/usr/local/lib/python3.12/dist-packages/torchao/quantization/quant_api.py:1731: SyntaxWarning: invalid escape sequence '\.'
+  """Configuration class for applying different quantization configs to modules or parameters based on their fully qualified names (FQNs).
+/sgl-workspace/sglang/python/sglang/launch_server.py:56: UserWarning: 'python -m sglang.launch_server' is still supported, but 'sglang serve' is the recommended entrypoint.
+  Example: sglang serve --model-path <model> [options]
+  warnings.warn(
+[2026-08-27 04:35:59] KV cache is forced as chunk cache for decode server
+[2026-08-27 04:36:03] Breakable CUDA graph is incompatible with MLA attention; disabling prefill CUDA graph.
+[2026-08-27 04:36:03] Attention backend not specified. Use triton backend by default.
+[2026-08-27 04:36:06] server_args=ServerArgs(model_path='/data/hf_models/Kimi-K3', tokenizer_path='/data/hf_models/Kimi-K3', tokenizer_mode='auto', tokenizer_backend='huggingface', tokenizer_worker_num=1, detokenizer_worker_num=1, skip_tokenizer_init=False, load_format='auto', model_loader_extra_config='{}', trust_remote_code=True, context_length=None, is_embedding=False, enable_multimodal=None, revision=None, model_impl='auto', model_config_parser='auto', json_model_override_args='{}', dtype='auto', quantization=None, quantization_param_path=None, kv_cache_dtype='auto', enable_fp32_lm_head=False, modelopt_quant=None, modelopt_checkpoint_restore_path=None, modelopt_checkpoint_save_path=None, modelopt_export_path=None, quantize_and_serve=False, rl_quant_profile=None, enable_tf32_matmul=False, mem_fraction_static=0.88, max_running_requests=None, max_queued_requests=None, max_total_tokens=None, chunked_prefill_size=8192, enable_dynamic_chunking=False, max_prefill_tokens=16384, prefill_max_requests=None, schedule_policy='fcfs', enable_priority_scheduling=False, disable_priority_preemption=False, default_priority_value=None, abort_on_priority_when_disabled=False, schedule_low_priority_values_first=False, priority_scheduling_preemption_threshold=10, retraction_policy='length', schedule_conservativeness=1.0, page_size=64, swa_full_tokens_ratio=0.8, disable_hybrid_swa_memory=False, radix_eviction_policy='lru', prefill_only_disable_kv_cache=False, disable_radix_cache=True, enable_page_major_kv_layout=False, enable_unified_memory=False, disable_chunked_prefix_cache=False, disable_overlap_schedule=False, num_continuous_decode_steps=1, scheduler_recv_interval=1, enable_mixed_chunk=False, nccl_port=None, dist_timeout=3600, dist_init_addr='174.1.60.5:20000', nnodes=4, node_rank=2, tp_size=32, dcp_size=1, pp_size=1, pp_max_micro_batch_size=None, pp_async_batch_depth=0, dp_size=1, load_balance_method='round_robin', attn_cp_size=1, moe_dp_size=1, dcp_comm_backend='ag_rs', dcp_replicate_q_proj=None, dwdp_size=1, enable_prefill_cp=False, cp_strategy=None, enable_dsa_cache_layer_split=False, enable_dsa_prefill_context_parallel=False, dsa_prefill_cp_mode='round-robin-split', enable_prefill_context_parallel=False, prefill_cp_mode='in-seq-split', enable_cp_decode_attn_tp=False, enable_dp_attention=False, enable_dp_attention_local_control_broadcast=False, enable_dp_lm_head=False, enable_attn_tp_input_scattered=False, disable_attn_tp_gather=False, enable_p2p_check=False, device='cuda', base_gpu_id=0, gpu_id_step=1, random_seed=212084647, watchdog_timeout=300, soft_watchdog_timeout=None, sleep_on_idle=False, use_ray=False, custom_sigquit_handler=None, numa_node=None, gc_threshold=None, host='0.0.0.0', port=30000, fastapi_root_path='', smg_grpc_mode=False, grpc_mode=False, grpc_port=None, sidecar=None, sidecar_args=None, skip_server_warmup=False, warmups=None, enable_http2=False, ssl_keyfile=None, ssl_certfile=None, ssl_ca_certs=None, ssl_keyfile_password=None, enable_ssl_refresh=False, api_key=None, admin_api_key=None, served_model_name='kimi-k3', weight_version='default', chat_template=None, hf_chat_template_name=None, completion_template=None, file_storage_path='sglang_storage', enable_cache_report=False, reasoning_parser=None, default_chat_template_kwargs=None, strip_thinking_cache=False, enable_strict_thinking=False, tool_call_parser=None, tool_server=None, sampling_defaults='model', asr_max_buffer_seconds=60, asr_max_concurrent_sessions=32, preferred_sampling_params=None, allow_auto_truncate=False, stream_interval=1, batch_notify_size=16, stream_response_default_include_usage=False, incremental_streaming_output=False, enable_streaming_session=False, enable_session_radix_cache=False, log_level='info', log_level_http=None, log_requests=False, log_requests_level=2, log_requests_format='text', log_requests_target=None, uvicorn_access_log_exclude_prefixes=[], crash_dump_folder=None, show_time_cost=False, enable_metrics=False, smg_http_sidecar_port=None, enable_mfu_metrics=False, enable_metrics_for_all_schedulers=False, load_snapshot_publish_interval=15, tokenizer_metrics_custom_labels_header='x-custom-labels', tokenizer_metrics_allowed_custom_labels=None, extra_metric_labels=None, bucket_time_to_first_token=None, bucket_inter_token_latency=None, bucket_e2e_request_latency=None, prompt_tokens_buckets=None, generation_tokens_buckets=None, gc_warning_threshold_secs=0.0, decode_log_interval=40, enable_request_time_stats_logging=False, kv_events_config=None, enable_forward_pass_metrics=False, forward_pass_metrics_worker_id='', forward_pass_metrics_ipc_name=None, enable_trace=False, trace_modules='request', otlp_traces_endpoint='localhost:4317', export_metrics_to_file=False, export_metrics_to_file_dir=None, stat_loggers=None, constrained_json_whitespace_pattern=None, constrained_json_disable_any_whitespace=False, attention_backend='triton', decode_attention_backend=None, prefill_attention_backend=None, sampling_backend='flashinfer', grammar_backend='xgrammar', radix_cache_backend=None, mm_attention_backend=None, fp8_gemm_runner_backend='auto', fp4_gemm_runner_backend='auto', bf16_gemm_backend='auto', dsa_prefill_backend=None, dsa_decode_backend=None, dsa_paged_mqa_logits_backend='auto', dsa_topk_backend='sgl-kernel', disable_flashinfer_autotune=False, flashinfer_autotune_skip_ops=None, mamba_backend='triton', cuda_graph_config=CudaGraphConfig(decode=PhaseConfig(backend='full', max_bs=16, bs=[1, 2, 4, 8, 12, 16], tc_compiler='eager', full_prefill_max_req=None), prefill=PhaseConfig(backend='disabled', max_bs=2048, bs=[4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], tc_compiler='eager', full_prefill_max_req=None)), cuda_graph_backend_decode=None, cuda_graph_backend_prefill=None, cuda_graph_max_bs_decode=16, cuda_graph_max_bs_prefill=None, cuda_graph_bs_decode=None, cuda_graph_bs_prefill=None, cuda_graph_tc_compiler=None, disable_prefill_cuda_graph=False, disable_decode_cuda_graph=False, disable_cuda_graph=False, disable_cuda_graph_padding=False, enable_profile_cuda_graph=False, enable_cudagraph_gc=False, debug_cuda_graph=False, enable_layerwise_nvtx_marker=False, enable_nccl_nvls=False, enable_symm_mem=False, triton_attention_reduce_in_fp32=False, triton_attention_num_kv_splits=8, triton_attention_split_tile_size=None, flashinfer_mla_disable_ragged=False, enable_fused_qk_norm_rope=False, enable_precise_embedding_interpolation=False, enable_fused_moe_sum_all_reduce=False, enable_deepseek_v4_fp4_indexer=False, disable_custom_all_reduce=False, enable_mscclpp=False, enable_torch_symm_mem=False, enable_scattered_sconv=False, pre_warm_nccl=False, enable_quant_communications=False, enable_flashinfer_allreduce_fusion=False, enforce_disable_flashinfer_allreduce_fusion=False, flashinfer_allreduce_fusion_backend=None, enable_aiter_allreduce_fusion=False, enable_torch_compile=False, enable_torch_compile_debug_mode=False, torch_compile_max_bs=32, torchao_config='', speculative_algorithm=None, speculative_draft_model_path=None, speculative_draft_model_revision=None, speculative_draft_load_format=None, speculative_num_steps=None, speculative_eagle_topk=None, speculative_num_draft_tokens=None, speculative_dflash_block_size=None, speculative_dspark_block_size=None, speculative_dspark_sps_table_path=None, speculative_dspark_confidence_sts_path=None, speculative_dspark_align_verify_tokens_to_graph_tier=False, speculative_accept_threshold_single=1.0, speculative_accept_threshold_acc=1.0, speculative_use_rejection_sampling=False, speculative_token_map=None, speculative_attention_mode='prefill', speculative_draft_attention_backend=None, speculative_draft_window_size=None, speculative_moe_runner_backend='marlin', speculative_moe_a2a_backend=None, speculative_draft_model_quantization=None, speculative_skip_dp_mlp_sync=False, enable_multi_layer_eagle=False, speculative_adaptive=False, speculative_adaptive_config=None, decoupled_spec_bind_endpoint=None, decoupled_spec_connect_endpoints=None, decoupled_spec_rank=None, decoupled_spec_role='null', spec_trace_dir=None, speculative_ngram_min_bfs_breadth=1, speculative_ngram_max_bfs_breadth=10, speculative_ngram_match_type='BFS', speculative_ngram_max_trie_depth=18, speculative_ngram_capacity=10000000, speculative_ngram_external_corpus_path=None, speculative_ngram_external_sam_budget=0, speculative_ngram_external_corpus_max_tokens=10000000, ep_size=32, moe_a2a_backend='none', moe_runner_backend='marlin', flashinfer_mxfp4_moe_precision='default', deepep_mode='auto', fuseep_mode=2, deepep_dispatcher_output_dtype='auto', ep_num_redundant_experts=0, ep_dispatch_algorithm=None, init_expert_location='trivial', enable_eplb=False, eplb_algorithm='auto', eplb_rebalance_num_iterations=1000, eplb_rebalance_layers_per_chunk=None, eplb_min_rebalancing_utilization_threshold=1.0, expert_distribution_recorder_mode=None, expert_distribution_recorder_buffer_size=1000, enable_expert_distribution_metrics=False, deepep_config=None, moe_dense_tp_size=None, elastic_ep_backend=None, enable_elastic_expert_backup=False, mooncake_ib_device=None, enable_waterfill=False, ep_join_mode=None, ep_join_rank_offset=0, elastic_ep_initial_size=None, max_ep_size=None, elastic_ep_scale_timeout=600, elastic_ep_rejoin=False, disable_flashinfer_cutlass_moe_fp4_allgather=False, disable_shared_experts_fusion=False, enforce_shared_experts_fusion=False, max_mamba_cache_size=None, mamba_ssm_dtype=None, mamba_max_states_per_path=-1, enable_mamba_cache_stochastic_rounding=False, mamba_cache_philox_rounds=0, mamba_full_memory_ratio=0.21, mamba_radix_cache_strategy='extra_buffer_lazy', uses_mamba_radix_cache=False, mamba_track_interval=256, enable_int8_mamba_checkpoint=False, int8_mamba_ckpt_size=None, linear_attn_backend='triton', linear_attn_decode_backend=None, linear_attn_prefill_backend=None, linear_attn_verify_backend=None, enable_linear_replayssm=False, linear_replayssm_cache_len=16, enable_linear_replayssm_spec=False, enable_hierarchical_cache=False, hicache_ratio=2.0, hicache_size=0, hicache_write_policy='write_through', hicache_io_backend='kernel', hicache_mem_layout='page_first', hicache_storage_backend=None, hicache_storage_prefetch_policy='timeout', hicache_storage_backend_extra_config=None, enable_hisparse=False, hisparse_config=None, enable_broadcast_mm_inputs_process=False, enable_prefix_mm_cache=False, mm_enable_dp_encoder=False, mm_process_config={}, mm_processor_worker_num=0, mm_io_worker_num=0, limit_mm_data_per_request=None, enable_mm_global_cache=False, disable_fast_image_processor=False, mm_feature_transport='cpu', keep_mm_feature_on_device=False, enable_lora=None, enable_lora_overlap_loading=None, max_lora_rank=None, lora_target_modules=None, lora_paths=None, max_loaded_loras=None, max_loras_per_batch=8, lora_eviction_policy='lru', lora_backend='csgmv', max_lora_chunk_size=16, experts_shared_outer_loras=None, lora_use_virtual_experts=False, lora_strict_loading=False, lora_drain_wait_threshold=0.0, enable_two_batch_overlap=False, enable_single_batch_overlap=False, tbo_token_distribution_threshold=0.48, cpu_offload_gb=0, offload_group_size=-1, offload_num_in_group=1, offload_prefetch_step=1, offload_mode='cpu', enable_lmcache=False, lmcache_config_file=None, enable_flexkv=False, flexkv_config_file=None, kt_weight_path=None, kt_method='AMXINT4', kt_cpuinfer=None, kt_threadpool_count=2, kt_num_gpu_experts=None, kt_max_deferred_experts_per_token=None, dllm_algorithm=None, dllm_algorithm_config=None, dllm_fdfo=True, disaggregation_mode='decode', disaggregation_transfer_backend='mooncake', disaggregation_bootstrap_port=28800, disaggregation_ib_device='mlx5_0,mlx5_1,mlx5_2,mlx5_3', disaggregation_decode_enable_radix_cache=False, disaggregation_decode_enable_offload_kvcache=False, num_reserved_decode_tokens=512, disaggregation_decode_extra_slots=0, disaggregation_decode_polling_interval=1, optimistic_prefill_attempts=0, encoder_only=False, language_only=False, encoder_transfer_backend='zmq_to_tokenizer', encoder_urls=[], encoder_bootstrap_port=8997, encoder_register_urls=[], enable_adaptive_dispatch_to_encoder=False, enable_pdmux=False, pdmux_config_path=None, sm_group_num=8, custom_weight_loader=[], weight_loader_disable_mmap=False, weight_loader_prefetch_checkpoints=False, weight_loader_prefetch_num_threads=4, weight_loader_drop_cache_after_load=False, remote_instance_weight_loader_seed_instance_ip=None, remote_instance_weight_loader_seed_instance_service_port=None, remote_instance_weight_loader_send_weights_group_ports=None, remote_instance_weight_loader_backend='nccl', remote_instance_weight_loader_start_seed_via_transfer_engine=False, engine_info_bootstrap_port=6789, modelexpress_config=None, download_dir=None, model_checksum=None, delete_ckpt_after_loading=False, decrypted_config_file=None, decrypted_draft_config_file=None, checkpoint_engine_wait_weights_before_ready=False, enable_prefill_delayer=False, prefill_delayer_max_delay_passes=30, prefill_delayer_token_usage_low_watermark=None, prefill_delayer_forward_passes_buckets=None, prefill_delayer_wait_seconds_buckets=None, prefill_delayer_queue_min_ratio=None, prefill_delayer_max_delay_ms=None, min_free_slots_delay=None, enable_deterministic_inference=False, rl_on_policy_target=None, kv_canary='none', kv_canary_real_data='none', kv_canary_sweep_interval=0, enable_dynamic_batch_tokenizer=False, dynamic_batch_tokenizer_batch_size=32, dynamic_batch_tokenizer_batch_timeout=0.002, enable_tokenizer_batch_encode=False, disable_tokenizer_batch_decode=False, debug_tensor_dump_output_folder=None, debug_tensor_dump_layers=None, debug_tensor_dump_input_file=None, enable_memory_saver=False, enable_weights_cpu_backup=False, enable_draft_weights_cpu_backup=False, enable_custom_logit_processor=False, enable_return_hidden_states=False, enable_return_routed_experts=False, enable_return_indexer_topk=False, disable_outlines_disk_cache=False, enable_mis=False, weight_cache_mode='off', weight_cache_socket=None, weight_cache_timeout=1800, forward_hooks=None, msprobe_dump_config=None)
+[2026-08-27 04:36:06] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:06] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:06] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:06] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:06] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:06] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:06] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:06] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:24 TP18 EP18] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:24 TP18 EP18] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:24 TP21 EP21] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:24 TP21 EP21] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:24 TP20 EP20] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:24 TP20 EP20] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:25 TP19 EP19] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:25 TP19 EP19] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:25 TP22 EP22] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:25 TP22 EP22] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:25 TP23 EP23] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:25 TP23 EP23] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:26 TP17 EP17] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:26 TP17 EP17] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:26 TP16 EP16] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:26 TP16 EP16] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:27 TP19 EP19] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:27 TP21 EP21] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:27 TP18 EP18] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:27 TP20 EP20] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:27 TP23 EP23] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:27 TP22 EP22] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:27 TP17 EP17] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:27 TP16 EP16] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:29.187384   239 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:29.187532   239 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:29.187558   239 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.7 port: 12001
+I0827 04:36:29.187624   239 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.7:16248
+I0827 04:36:29.187814   239 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:29.191661   235 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:29.191802   235 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:29.191823   235 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.7 port: 12001
+I0827 04:36:29.191901   235 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.7:16408
+I0827 04:36:29.192099   235 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:29.195215   242 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:29.195319   242 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:29.195333   242 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.7 port: 12001
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:29.195333   238 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:29.195374   242 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.7:15148
+I0827 04:36:29.195490   238 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:29.195513   238 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.7 port: 12001
+I0827 04:36:29.195519   242 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:29.195581   238 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.7:16909
+I0827 04:36:29.195815   238 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:29.198207   236 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:29.198346   236 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:29.198370   236 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.7 port: 12001
+I0827 04:36:29.198427   236 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.7:16738
+I0827 04:36:29.198611   236 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:29.200070   242 topology.cpp:127] Device mlx5_0 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:29.201086   237 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:29.201221   237 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:29.201239   237 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.7 port: 12001
+I0827 04:36:29.201300   237 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.7:16522
+I0827 04:36:29.201495   237 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:29.201511   240 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:29.201632   240 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:29.201648   240 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.7 port: 12001
+I0827 04:36:29.201701   240 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.7:16431
+I0827 04:36:29.201871   240 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:29.204236   241 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:29.204355   242 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:29.204380   241 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:29.204393   241 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.7 port: 12001
+I0827 04:36:29.204454   241 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.7:16508
+I0827 04:36:29.204633   241 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:29.204896   238 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:29.207675   237 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:29.208006   240 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:29.208549   242 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:29.212560   242 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:29.214218   242 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:29.214243   242 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:29.214290   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.214713   240 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:29.214757   237 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:29.219102   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:29.219870   239 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:29.220057   235 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:29.220343   242 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:07
+I0827 04:36:29.220369   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.222033   240 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:29.222069   237 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:29.223913   236 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:29.228577   235 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:29.228634   239 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:29.230391   237 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:29.230446   240 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:29.232455   236 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:29.233546   237 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:29.233578   237 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:29.233631   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.234635   240 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:29.234665   240 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:29.234716   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.235774   238 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:29.235827   241 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:29.237140   235 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:29.241117   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:29.241130   236 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:29.242069   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:29.242725   237 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:07
+I0827 04:36:29.242758   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.243661   240 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:07
+I0827 04:36:29.243691   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.245620   235 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:29.247138   238 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:29.247233   241 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:29.248172   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:29.248514   235 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:29.248550   235 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:29.248612   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.248719   236 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:29.249847   242 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:07
+I0827 04:36:29.249886   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.251354   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:29.251672   236 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:29.251719   236 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:29.251781   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.252889   240 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:07
+I0827 04:36:29.252921   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.255584   239 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:29.256263   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:29.257805   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:29.257980   235 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:07
+I0827 04:36:29.258016   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.258512   238 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:29.258567   241 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:29.259447   242 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:07
+I0827 04:36:29.259477   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.259642   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:29.261375   236 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:07
+I0827 04:36:29.261410   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.261518   238 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:29.261557   238 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:29.261615   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.261615   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:29.262753   239 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:29.263202   240 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:07
+I0827 04:36:29.263229   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.265475   239 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:29.265504   239 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:29.265558   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.269449   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:29.271157   236 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:07
+I0827 04:36:29.271186   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.271761   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:29.272920   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:29.273422   237 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:07
+I0827 04:36:29.273456   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.274566   239 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:07
+I0827 04:36:29.274600   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.281176   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:29.282341   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:29.282708   237 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:07
+I0827 04:36:29.282739   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.284137   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:29.284178   239 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:07
+I0827 04:36:29.284219   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.285904   235 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:07
+I0827 04:36:29.285953   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.288185   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:29.289810   242 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:07
+I0827 04:36:29.289940   242 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:29 TP23 EP23] Init torch distributed begin.
+I0827 04:36:29.291602   241 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:29.291882   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:29.291936   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:29.293576   240 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:07
+I0827 04:36:29.293699   240 transfer_engine_impl.cpp:388] installTransport, type=rdma
+I0827 04:36:29.293921   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+[2026-08-27 04:36:29 TP21 EP21] Init torch distributed begin.
+I0827 04:36:29.294121   238 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:07
+I0827 04:36:29.294169   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.294241   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:29.295475   239 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:07
+I0827 04:36:29.295507   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.295553   241 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:29.295591   241 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:29.295648   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.296002   235 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:07
+I0827 04:36:29.296041   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.296232   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:29.298113   236 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:07
+I0827 04:36:29.298154   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.303915   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:29.305596   235 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:07
+I0827 04:36:29.305743   235 transfer_engine_impl.cpp:388] installTransport, type=rdma
+I0827 04:36:29.306217   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+[2026-08-27 04:36:29 TP16 EP16] Init torch distributed begin.
+I0827 04:36:29.307994   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:29.308105   236 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:07
+I0827 04:36:29.308277   236 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:29 TP17 EP17] Init torch distributed begin.
+I0827 04:36:29.310081   237 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:07
+I0827 04:36:29.310261   237 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:29 TP18 EP18] Init torch distributed begin.
+I0827 04:36:29.320421   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:29.321812   239 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:07
+I0827 04:36:29.321919   239 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:29 TP20 EP20] Init torch distributed begin.
+I0827 04:36:29.323829   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:29.323894   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:29.325821   238 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:07
+I0827 04:36:29.325820   241 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:07
+I0827 04:36:29.325871   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.325872   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.360142   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:29.360253   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:29.362092   241 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:07
+I0827 04:36:29.362147   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.362201   238 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:07
+I0827 04:36:29.362243   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.392091   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:29.392130   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:29.394111   238 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:07
+I0827 04:36:29.394127   241 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:07
+I0827 04:36:29.394166   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:29.394286   238 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:29 TP19 EP19] Init torch distributed begin.
+I0827 04:36:29.420166   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:29.421864   241 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:07
+I0827 04:36:29.421993   241 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:29 TP22 EP22] Init torch distributed begin.
+[2026-08-27 04:36:48 TP23 EP23] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP22 EP22] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP21 EP21] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP20 EP20] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP19 EP19] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP18 EP18] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP17 EP17] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP16 EP16] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP22 EP22] Init torch distributed ends. elapsed=18.81 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP23 EP23] Init torch distributed ends. elapsed=18.94 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP21 EP21] Init torch distributed ends. elapsed=18.93 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP20 EP20] Init torch distributed ends. elapsed=18.91 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP19 EP19] Init torch distributed ends. elapsed=18.83 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP18 EP18] Init torch distributed ends. elapsed=18.92 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP17 EP17] Init torch distributed ends. elapsed=18.92 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP16 EP16] Init torch distributed ends. elapsed=18.92 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP21 EP21] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP16 EP16] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP22 EP22] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP20 EP20] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP23 EP23] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP18 EP18] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP19 EP19] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP22 EP22] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP20 EP20] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP21 EP21] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP16 EP16] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP23 EP23] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP19 EP19] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP18 EP18] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP17 EP17] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP17 EP17] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP20 EP20] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP22 EP22] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP18 EP18] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP23 EP23] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP19 EP19] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP21 EP21] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP20 EP20] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP16 EP16] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP18 EP18] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP22 EP22] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP19 EP19] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP21 EP21] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP23 EP23] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP16 EP16] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP17 EP17] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP17 EP17] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:50 TP23 EP23] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP22 EP22] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP21 EP21] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP19 EP19] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP18 EP18] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP20 EP20] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP16 EP16] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP17 EP17] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:38:53 TP16 EP16] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:53 TP16 EP16] Load weight end. elapsed=124.66 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:53 TP17 EP17] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:53 TP17 EP17] Load weight end. elapsed=124.69 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:53 TP23 EP23] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:53 TP23 EP23] Load weight end. elapsed=124.71 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:53 TP21 EP21] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:53 TP21 EP21] Load weight end. elapsed=124.74 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:53 TP18 EP18] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:53 TP18 EP18] Load weight end. elapsed=124.82 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:53 TP20 EP20] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:53 TP20 EP20] Load weight end. elapsed=124.82 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:53 TP22 EP22] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:53 TP22 EP22] Load weight end. elapsed=124.92 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:53 TP19 EP19] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:53 TP19 EP19] Load weight end. elapsed=125.03 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:39:01 TP17 EP17] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP17 EP17] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP23 EP23] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP23 EP23] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP16 EP16] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP16 EP16] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP18 EP18] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP18 EP18] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP22 EP22] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP22 EP22] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP21 EP21] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP21 EP21] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP20 EP20] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP20 EP20] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP19 EP19] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP19 EP19] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:02 TP23 EP23] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP18 EP18] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP21 EP21] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP22 EP22] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP20 EP20] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP17 EP17] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP19 EP19] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP16 EP16] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP21 EP21] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP22 EP22] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP23 EP23] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP17 EP17] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP16 EP16] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP20 EP20] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP18 EP18] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP19 EP19] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP21 EP21] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP23 EP23] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP22 EP22] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP17 EP17] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP23 EP23] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP21 EP21] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP20 EP20] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP16 EP16] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP22 EP22] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP17 EP17] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP20 EP20] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP16 EP16] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP18 EP18] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP19 EP19] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP18 EP18] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP19 EP19] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP23 EP23] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP23 EP23] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:39:02 TP17 EP17] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP17 EP17] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP16 EP16] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP16 EP16] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:39:02 TP21 EP21] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP21 EP21] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP22 EP22] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP22 EP22] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP20 EP20] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP20 EP20] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:39:02 TP18 EP18] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP18 EP18] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP19 EP19] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP19 EP19] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:42:56 TP19 EP19] Capture target decode CUDA graph end. elapsed=233.80 s, mem usage=0.62 GB, avail mem=8.51 GB.
+[2026-08-27 04:42:56 TP19 EP19] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP16 EP16] Capture target decode CUDA graph end. elapsed=233.84 s, mem usage=0.60 GB, avail mem=8.52 GB.
+[2026-08-27 04:42:56 TP16 EP16] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP22 EP22] Capture target decode CUDA graph end. elapsed=233.84 s, mem usage=0.56 GB, avail mem=8.59 GB.
+[2026-08-27 04:42:56 TP22 EP22] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP17 EP17] Capture target decode CUDA graph end. elapsed=233.88 s, mem usage=0.58 GB, avail mem=8.58 GB.
+[2026-08-27 04:42:56 TP17 EP17] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP21 EP21] Capture target decode CUDA graph end. elapsed=233.90 s, mem usage=0.58 GB, avail mem=8.58 GB.
+[2026-08-27 04:42:56 TP21 EP21] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP20 EP20] Capture target decode CUDA graph end. elapsed=233.91 s, mem usage=0.60 GB, avail mem=8.52 GB.
+[2026-08-27 04:42:56 TP20 EP20] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP23 EP23] Capture target decode CUDA graph end. elapsed=234.01 s, mem usage=0.62 GB, avail mem=8.51 GB.
+[2026-08-27 04:42:56 TP23 EP23] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP18 EP18] Capture target decode CUDA graph end. elapsed=234.00 s, mem usage=0.56 GB, avail mem=8.59 GB.
+[2026-08-27 04:42:56 TP18 EP18] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+W0827 04:42:57.619849  2423 memory_location.cpp:73] Failed to get NUMA node, addr: 0x9680d380, len: 23168: Operation not permitted [1]
+W0827 04:42:57.619855  2422 memory_location.cpp:73] Failed to get NUMA node, addr: 0x9696f4c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.619967  2424 memory_location.cpp:73] Failed to get NUMA node, addr: 0x966bf4c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.620213  2425 memory_location.cpp:73] Failed to get NUMA node, addr: 0x82b481c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.620281  2426 memory_location.cpp:73] Failed to get NUMA node, addr: 0x988cf580, len: 185344: Operation not permitted [1]
+W0827 04:42:57.620285  2427 memory_location.cpp:73] Failed to get NUMA node, addr: 0x988fc9c0, len: 185344: Operation not permitted [1]
+W0827 04:42:57.620325  2428 memory_location.cpp:73] Failed to get NUMA node, addr: 0x96988ec0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.620384  2429 memory_location.cpp:73] Failed to get NUMA node, addr: 0x96949540, len: 46336: Operation not permitted [1]
+W0827 04:42:57.620546  2430 memory_location.cpp:73] Failed to get NUMA node, addr: 0x96790780, len: 23168: Operation not permitted [1]
+W0827 04:42:57.620606  2431 memory_location.cpp:73] Failed to get NUMA node, addr: 0x967ccac0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.651996  2570 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6c8ac200, len: 23168: Operation not permitted [1]
+W0827 04:42:57.652108  2571 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6c6f5f80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.652212  2572 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6c8ea380, len: 23168: Operation not permitted [1]
+W0827 04:42:57.652417  2573 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6d2e8ec0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.652568  2574 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6ea4eb80, len: 185344: Operation not permitted [1]
+W0827 04:42:57.652587  2576 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6cb20d80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.652642  2577 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6c987340, len: 46336: Operation not permitted [1]
+W0827 04:42:57.652667  2575 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6ea7bfc0, len: 185344: Operation not permitted [1]
+W0827 04:42:57.652815  2578 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6c6eb200, len: 23168: Operation not permitted [1]
+W0827 04:42:57.652889  2579 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6cb0a2c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.713306  2681 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8a7feb00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.713331  2680 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ad55240, len: 23168: Operation not permitted [1]
+W0827 04:42:57.713303  2679 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ad28280, len: 23168: Operation not permitted [1]
+W0827 04:42:57.713424  2682 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8abc1200, len: 23168: Operation not permitted [1]
+W0827 04:42:57.713523  2683 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ca64980, len: 185344: Operation not permitted [1]
+W0827 04:42:57.713629  2685 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ca91dc0, len: 185344: Operation not permitted [1]
+W0827 04:42:57.713903  2687 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ac54c00, len: 46336: Operation not permitted [1]
+W0827 04:42:57.714316  2688 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ab93e40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.714335  2689 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8abe8700, len: 23168: Operation not permitted [1]
+W0827 04:42:57.714352  2686 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8a7f7880, len: 23168: Operation not permitted [1]
+W0827 04:42:57.717235  2716 memory_location.cpp:73] Failed to get NUMA node, addr: 0x739fc3c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.717247  2718 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ff359c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.717484  2725 memory_location.cpp:73] Failed to get NUMA node, addr: 0x75e8fa00, len: 185344: Operation not permitted [1]
+W0827 04:42:57.717736  2729 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ff46700, len: 46336: Operation not permitted [1]
+W0827 04:42:57.718014  2728 memory_location.cpp:73] Failed to get NUMA node, addr: 0x73b6a280, len: 23168: Operation not permitted [1]
+W0827 04:42:57.718112  2730 memory_location.cpp:73] Failed to get NUMA node, addr: 0x73dda340, len: 23168: Operation not permitted [1]
+W0827 04:42:57.718163  2731 memory_location.cpp:73] Failed to get NUMA node, addr: 0x73dbf000, len: 23168: Operation not permitted [1]
+W0827 04:42:57.718196  2721 memory_location.cpp:73] Failed to get NUMA node, addr: 0x73a8f640, len: 23168: Operation not permitted [1]
+W0827 04:42:57.718261  2723 memory_location.cpp:73] Failed to get NUMA node, addr: 0x73c5ba40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.724411  2727 memory_location.cpp:73] Failed to get NUMA node, addr: 0x75ebce40, len: 185344: Operation not permitted [1]
+W0827 04:42:57.755818  2858 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5e996800, len: 23168: Operation not permitted [1]
+W0827 04:42:57.755931  2859 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ec52140, len: 23168: Operation not permitted [1]
+W0827 04:42:57.756026  2861 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5e767ec0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.756074  2862 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60bbb440, len: 185344: Operation not permitted [1]
+W0827 04:42:57.756379  2865 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4aecca80, len: 46336: Operation not permitted [1]
+W0827 04:42:57.756517  2863 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60be8880, len: 185344: Operation not permitted [1]
+W0827 04:42:57.756536  2864 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5eb0d880, len: 23168: Operation not permitted [1]
+W0827 04:42:57.756650  2866 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ea7fe80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.757373  2860 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5eb4d800, len: 23168: Operation not permitted [1]
+W0827 04:42:57.757380  2867 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ea97dc0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.818434  3070 memory_location.cpp:73] Failed to get NUMA node, addr: 0x965568c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.818459  3075 memory_location.cpp:73] Failed to get NUMA node, addr: 0x96700fc0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.818470  3072 memory_location.cpp:73] Failed to get NUMA node, addr: 0x96541bc0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.820174  3079 memory_location.cpp:73] Failed to get NUMA node, addr: 0x984fb440, len: 185344: Operation not permitted [1]
+W0827 04:42:57.822142  3084 memory_location.cpp:73] Failed to get NUMA node, addr: 0x962df200, len: 23168: Operation not permitted [1]
+W0827 04:42:57.822369  3086 memory_location.cpp:73] Failed to get NUMA node, addr: 0x831c0780, len: 46336: Operation not permitted [1]
+W0827 04:42:57.822384  3110 memory_location.cpp:73] Failed to get NUMA node, addr: 0x965cd3c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.822431  3082 memory_location.cpp:73] Failed to get NUMA node, addr: 0x98528880, len: 185344: Operation not permitted [1]
+W0827 04:42:57.822527  3127 memory_location.cpp:73] Failed to get NUMA node, addr: 0x9642bbc0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.829262  3077 memory_location.cpp:73] Failed to get NUMA node, addr: 0x9640b580, len: 23168: Operation not permitted [1]
+W0827 04:42:57.832078  3193 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ce567c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.832136  3194 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8cbe8d00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.832149  3195 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8cc94640, len: 23168: Operation not permitted [1]
+W0827 04:42:57.832294  3196 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8d79f140, len: 23168: Operation not permitted [1]
+W0827 04:42:57.832419  3199 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ce4bec0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.832562  3201 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8d7e4b40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.832683  3202 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8cbbd940, len: 23168: Operation not permitted [1]
+W0827 04:42:57.833209  3198 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ef88080, len: 185344: Operation not permitted [1]
+W0827 04:42:57.833267  3200 memory_location.cpp:73] Failed to get NUMA node, addr: 0x79c278c0, len: 46336: Operation not permitted [1]
+W0827 04:42:57.837312  3197 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ef5ac40, len: 185344: Operation not permitted [1]
+W0827 04:42:58.193480  3458 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59a23f40, len: 23168: Operation not permitted [1]
+W0827 04:42:58.193507  3459 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59aa70c0, len: 23168: Operation not permitted [1]
+W0827 04:42:58.193545  3460 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5999c100, len: 23168: Operation not permitted [1]
+W0827 04:42:58.193585  3461 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59bc3040, len: 23168: Operation not permitted [1]
+W0827 04:42:58.193631  3462 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5bd1e5c0, len: 185344: Operation not permitted [1]
+W0827 04:42:58.193667  3463 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5bd4ba00, len: 185344: Operation not permitted [1]
+W0827 04:42:58.193720  3464 memory_location.cpp:73] Failed to get NUMA node, addr: 0x599fa280, len: 23168: Operation not permitted [1]
+W0827 04:42:58.193746  3466 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59dd6100, len: 23168: Operation not permitted [1]
+W0827 04:42:58.193764  3467 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59cbd400, len: 23168: Operation not permitted [1]
+W0827 04:42:58.193776  3465 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59db0340, len: 46336: Operation not permitted [1]
+[2026-08-27 04:43:00] Dummy health check server started in background thread at 0.0.0.0:30000
+I0827 04:48:13.990782  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:13.991083  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.005937  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.006215  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.026726  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.034708  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.038136  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.038393  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.052722  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.052948  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.069641  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.089646  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.092875  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.099138  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.099377  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.099617  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.102869  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.103116  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.103413  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.103614  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.103897  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.104102  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.104316  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.107553  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.107847  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.113867  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.125564  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.131559  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.131772  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.140426  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.140640  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.140889  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.141057  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.144480  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.144747  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.160434  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.160689  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.160965  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.177232  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.188812  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.189050  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.189292  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.189566  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.189749  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.197841  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.198038  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.198292  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.198523  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.198736  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.198998  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.210616  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.210834  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.211103  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.214471  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.216403  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.217782  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.227747  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.228065  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.228319  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.228518  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.230293  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.244758  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.244995  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.245216  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.246076  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.247792  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.248039  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.248274  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.251454  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.255178  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.255391  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.256364  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.256808  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.260069  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.260272  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.260452  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.260666  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.260878  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.266151  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.266335  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.266537  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.266744  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.266952  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.267238  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.267465  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.268203  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.270165  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.272876  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.273144  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.275923  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.276090  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.281610  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.281848  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.285796  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.285949  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.287339  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.289616  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.289852  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.290091  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.290292  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.290467  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.290657  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.290853  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.291211  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.295532  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.297179  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.297688  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.299065  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.299283  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.300338  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.309764  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.310025  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.310344  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.310527  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.310770  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.314235  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.317247  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.318601  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.324312  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.338757  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.343169  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.344022  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.344889  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.353394  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.354696  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.354852  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.360512  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.363333  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.363579  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.363832  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.364054  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.364255  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.364542  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.364830  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.368968  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.369174  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.369390  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.373200  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.374004  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.374215  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.374486  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.379066  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.379289  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.383756  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.384938  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.393990  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.394248  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.394496  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.400096  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.400372  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.400589  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.400769  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.400981  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.401257  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.401494  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.401587  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.402990  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.409467  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.409701  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.410995  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.412505  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.423301  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.423527  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.423753  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.425726  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.426009  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.426232  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.429703  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.432900  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.436066  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.437069  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.437340  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.437568  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.437778  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.438095  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.441175  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.442378  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.443518  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.443746  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.443964  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.444185  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.444403  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.453183  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.454483  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.458822  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.460139  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.460391  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.461473  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.468333  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.469964  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.474977  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.475178  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.475373  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.475994  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.476274  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.482146  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.483319  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.483541  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.483824  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.487998  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.491807  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.492008  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.492403  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.492631  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.492843  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.493103  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.493304  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.493522  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.493757  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.493925  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.493986  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.494186  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.494228  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.494501  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.494679  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.494771  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.494979  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.495187  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.495486  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.495695  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.495930  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.496155  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.496392  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.498075  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.499614  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.499698  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.507068  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.507290  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.515178  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.515554  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.515731  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.515975  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.516929  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.517546  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.519863  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.520967  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.522394  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.522583  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.522799  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.523052  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.523262  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.523473  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.523695  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.523783  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.523902  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.524128  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.524420  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.524629  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.525529  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.525698  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.526990  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.535063  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.535307  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.535513  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.536439  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.536978  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.537204  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.537415  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.537595  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.539846  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.540122  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.540378  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.540510  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.540766  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.541023  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.541293  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.544811  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.547500  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.547943  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.548959  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.551965  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.553215  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.563808  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.564100  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.568046  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.569260  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.569701  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.571031  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.573542  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.579483  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.581025  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.587301  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.587486  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.587728  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.588649  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.594245  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.594436  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.594633  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.594759  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.596094  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.596485  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.596745  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.597023  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.597223  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.597502  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.597754  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.598001  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.598238  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.598425  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.598681  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.598914  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.599107  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.599298  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.601323  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.601538  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.602514  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.606913  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.611363  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.611377  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.612835  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.615579  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.616807  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.616865  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.616990  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.617111  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.617369  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.617590  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.617800  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.617998  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.618206  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.621685  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.623104  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.633111  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.633315  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.634286  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.634464  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.634523  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.634792  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.634966  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.635149  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.635445  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.635624  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.635821  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.636071  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.637221  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.643270  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.644847  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.648780  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.648998  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.656845  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.657044  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.657281  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.658142  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.660220  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.661625  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.664076  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.665403  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.667707  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.669119  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.670938  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.677649  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.677932  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.677983  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.678122  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.678347  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.678530  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.678735  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.678929  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.679240  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.679463  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.679695  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.679894  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.680079  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.680286  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.680457  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.680696  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.681843  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.683002  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.683204  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.683274  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.683362  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.683475  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.683499  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.683585  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.683856  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.684543  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.687750  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.689368  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.690239  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.691855  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.693795  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.696607  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.696817  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.698160  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.703644  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.706352  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.706555  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.706766  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.707012  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.707203  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.707451  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.707659  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.707913  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.708099  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.708461  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.708745  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.708987  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.709229  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.713573  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.714908  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.718238  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.718487  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.718716  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.719660  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.720849  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.722200  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.724702  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.725086  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.725322  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.725628  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.725817  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.726325  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.731271  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.736089  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.736286  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.736464  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.737399  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.740564  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.742311  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.743717  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.746318  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.746944  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.747165  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.748356  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.749109  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.749446  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.749665  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.750972  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.755626  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.756073  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.756265  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.757405  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.762998  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.764305  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.765604  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.765775  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.766307  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.766513  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.766744  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.766980  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.767177  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.767390  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.767621  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.767817  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.775632  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.778723  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.778945  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.779145  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.780236  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.783447  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.786221  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.786506  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.789175  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.789419  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.790586  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.790643  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.790793  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.790974  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.791939  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.796165  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.797827  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.802098  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.813362  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.814442  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.814667  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.814668  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.814859  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.820125  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.820905  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.821156  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.822324  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.832154  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.832475  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.832693  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.832880  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.833066  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.833330  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.833523  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.833789  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.834038  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.834266  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.834484  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.834671  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.834861  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.835052  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.835216  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.835410  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.838090  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.848765  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.849062  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.849324  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.849581  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.849845  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.850172  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.858408  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.858647  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.859768  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.860574  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.860812  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.861027  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.861253  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.861526  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.861723  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.861910  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.862110  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.862299  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.862488  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.864478  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.872546  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.872754  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.872980  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.873731  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.873880  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.873962  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.874325  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.875612  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.877731  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.879717  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.880618  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.885710  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.886065  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.886333  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.890219  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.891563  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.895009  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.903152  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.905467  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.913136  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.915786  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.916013  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.918720  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.918928  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.919982  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.920882  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.921152  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.921317  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.921556  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.921797  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.921981  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.922211  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.924968  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.925161  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.926721  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.930300  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.931375  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.932979  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.933200  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.935876  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.941114  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.943547  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.943744  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.943778  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.943972  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.944057  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.944993  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.954442  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.955266  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.955477  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.955673  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.956692  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.974392  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.974649  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.974893  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.975108  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.985150  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.985353  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.985610  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.985852  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.986055  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.986230  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.986440  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.986611  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.991794  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.991982  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.992164  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.992372  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.004820  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.007462  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.012593  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.012835  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.013056  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.013247  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.013500  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.013749  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.016324  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.016539  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.021533  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.024173  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.024427  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.025575  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.067353  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.067580  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.067747  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.067951  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.068106  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.068390  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.068591  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.068789  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.078953  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.079149  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.079314  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.080277  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.090672  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.090881  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.091123  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.092058  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.132000  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.132241  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.132424  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.132637  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.132885  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.133169  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.133404  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.133606  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.133805  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.133991  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.134207  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.134473  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.134696  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.134949  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.135172  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.135368  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.160171  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.160442  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.160633  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.160794  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.161010  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.161175  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.161370  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.161597  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.161856  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.162032  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.171926  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.172152  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.172313  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.173470  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.217530  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.217757  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.217985  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.218160  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.218354  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.218546  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.218766  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.218966  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.229063  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.229290  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.229496  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.230377  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.240563  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.240742  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.240916  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.241938  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.282371  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.282605  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.282783  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.282986  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.283192  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.283366  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.283558  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.283744  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.283931  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.284118  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.284356  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.284595  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.284804  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.285007  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.285272  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.290516  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.310380  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.310582  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.310774  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.310997  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.311203  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.311424  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.311609  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.311781  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.311981  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.312191  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.322089  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.322285  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.322484  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.323439  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.366820  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.367026  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.367223  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.367421  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.367637  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.367813  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.368016  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.368258  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.378153  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.378351  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.378584  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.379503  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.389467  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.389665  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.389904  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.390815  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.430491  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.430742  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.430959  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.431205  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.431394  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.431623  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.431800  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.432049  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.432286  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.432466  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.432688  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.432901  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.433138  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.433328  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.433526  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.436129  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.458094  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.458309  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.458535  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.458752  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.458948  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.459110  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.459302  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.459517  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.459699  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.459882  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.469916  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.470148  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.470368  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.471292  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.448522  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.450094  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.460287  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.460638  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.462020  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.466852  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.473448  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.481580  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.482398  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.482719  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.483039  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.483215  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.483292  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.483968  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.488898  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.493845  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.494148  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.495286  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.501287  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.504213  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.512477  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.512745  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.513012  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.513972  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.297312  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.299115  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.304275  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.314786  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.316474  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.328058  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.328325  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.328570  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.329540  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.338943  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.339195  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.339458  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.340605  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.341210  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.346589  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.357823  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.358044  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.358083  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.358332  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.358369  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.359782  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.363950  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.366775  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.367130  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.368332  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.369814  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.374266  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.375430  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.378304  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.380679  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.381263  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.381505  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.381752  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.382050  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.382229  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.382687  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.386806  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.386886  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.388373  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.392937  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.394059  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.395437  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.395885  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.396198  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.397209  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.397620  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.398900  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.399183  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.399243  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.400318  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.401305  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.403239  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.404074  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.404656  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.409005  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.411810  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.414434  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.415934  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.417624  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.417887  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.418118  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.420249  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.423964  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.424198  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.424408  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.430177  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.433308  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.433637  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.436582  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.439628  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.440941  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.394086  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.394385  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.395802  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.408921  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.409214  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.410487  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.420368  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.420595  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.421932  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.426435  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.573661  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.574067  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.575153  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.109197  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.110759  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.117308  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.127852  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.129487  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.134262  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.139751  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.141630  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.146772  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.148381  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.148725  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.149887  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.154366  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.161194  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.162659  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.168562  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.654505  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.655917  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.669334  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.669617  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.669867  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.670902  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.679297  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.680840  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.687160  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.692883  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.693225  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.693225  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.693482  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.693727  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.694342  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.694828  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.699385  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.700112  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.701340  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.702790  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.708935  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.719955  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.720265  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.720535  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.721411  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.727005  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.728433  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.462263  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.462824  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.463836  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.475934  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.477473  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.481695  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.487771  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.489451  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.495987  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.185277  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.186878  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.193236  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:47.566026  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:47.575708  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:47.584995  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:48.551697  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:48.561645  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:48.570935  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:49.336337  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:49.347584  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:49.348870  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:49.353668  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:49.362033  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:50.265223  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:50.274431  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:50.283521  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.110039  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.122382  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.127803  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.131213  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.163244  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.171993  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.179919  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.997390  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.007871  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.013643  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.018814  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.023959  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.033485  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.799268  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.810883  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.819684  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:53.708249  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:53.717978  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:53.727499  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:53.735821  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:54.567750  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:54.576763  1369 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:55.320358  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:55.447818  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:55.457465  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:57.588092  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:57.606285  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:57.611370  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:57.639269  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:57.640733  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:57.646224  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:57.660796  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:57.732604  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:57.734110  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:58.280643  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:58.297982  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:59.042151  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:59.051748  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:08.244721  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:08.253723  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:09.194365  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:10.135797  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:10.203588  1372 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:10.219120  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:10.975451  1368 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:10.985549  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:11.843453  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:11.845074  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:12.790898  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:13.633699  1371 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:13.643146  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:15.418679  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:17.010130  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:17.183473  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:17.197549  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:17.986510  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:18.003332  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:18.214159  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:18.224337  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:18.241684  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:18.243786  1363 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:18.258334  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:18.350939  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:18.364851  1370 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:18.893000  1367 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:29.860329  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:46.812764  1345 rdma_endpoint.cpp:671] Received RDMA ready ACK.
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/logs/manual_d_node3.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/logs/manual_d_node3.log
new file mode 100644
index 0000000..1c0a7ea
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/logs/manual_d_node3.log
@@ -0,0 +1,1461 @@
+/usr/local/lib/python3.12/dist-packages/torchao/quantization/quant_api.py:1731: SyntaxWarning: invalid escape sequence '\.'
+  """Configuration class for applying different quantization configs to modules or parameters based on their fully qualified names (FQNs).
+/sgl-workspace/sglang/python/sglang/launch_server.py:56: UserWarning: 'python -m sglang.launch_server' is still supported, but 'sglang serve' is the recommended entrypoint.
+  Example: sglang serve --model-path <model> [options]
+  warnings.warn(
+[2026-08-27 04:36:00] KV cache is forced as chunk cache for decode server
+[2026-08-27 04:36:03] Breakable CUDA graph is incompatible with MLA attention; disabling prefill CUDA graph.
+[2026-08-27 04:36:03] Attention backend not specified. Use triton backend by default.
+[2026-08-27 04:36:05] server_args=ServerArgs(model_path='/data/hf_models/Kimi-K3', tokenizer_path='/data/hf_models/Kimi-K3', tokenizer_mode='auto', tokenizer_backend='huggingface', tokenizer_worker_num=1, detokenizer_worker_num=1, skip_tokenizer_init=False, load_format='auto', model_loader_extra_config='{}', trust_remote_code=True, context_length=None, is_embedding=False, enable_multimodal=None, revision=None, model_impl='auto', model_config_parser='auto', json_model_override_args='{}', dtype='auto', quantization=None, quantization_param_path=None, kv_cache_dtype='auto', enable_fp32_lm_head=False, modelopt_quant=None, modelopt_checkpoint_restore_path=None, modelopt_checkpoint_save_path=None, modelopt_export_path=None, quantize_and_serve=False, rl_quant_profile=None, enable_tf32_matmul=False, mem_fraction_static=0.88, max_running_requests=None, max_queued_requests=None, max_total_tokens=None, chunked_prefill_size=8192, enable_dynamic_chunking=False, max_prefill_tokens=16384, prefill_max_requests=None, schedule_policy='fcfs', enable_priority_scheduling=False, disable_priority_preemption=False, default_priority_value=None, abort_on_priority_when_disabled=False, schedule_low_priority_values_first=False, priority_scheduling_preemption_threshold=10, retraction_policy='length', schedule_conservativeness=1.0, page_size=64, swa_full_tokens_ratio=0.8, disable_hybrid_swa_memory=False, radix_eviction_policy='lru', prefill_only_disable_kv_cache=False, disable_radix_cache=True, enable_page_major_kv_layout=False, enable_unified_memory=False, disable_chunked_prefix_cache=False, disable_overlap_schedule=False, num_continuous_decode_steps=1, scheduler_recv_interval=1, enable_mixed_chunk=False, nccl_port=None, dist_timeout=3600, dist_init_addr='174.1.60.5:20000', nnodes=4, node_rank=3, tp_size=32, dcp_size=1, pp_size=1, pp_max_micro_batch_size=None, pp_async_batch_depth=0, dp_size=1, load_balance_method='round_robin', attn_cp_size=1, moe_dp_size=1, dcp_comm_backend='ag_rs', dcp_replicate_q_proj=None, dwdp_size=1, enable_prefill_cp=False, cp_strategy=None, enable_dsa_cache_layer_split=False, enable_dsa_prefill_context_parallel=False, dsa_prefill_cp_mode='round-robin-split', enable_prefill_context_parallel=False, prefill_cp_mode='in-seq-split', enable_cp_decode_attn_tp=False, enable_dp_attention=False, enable_dp_attention_local_control_broadcast=False, enable_dp_lm_head=False, enable_attn_tp_input_scattered=False, disable_attn_tp_gather=False, enable_p2p_check=False, device='cuda', base_gpu_id=0, gpu_id_step=1, random_seed=297832107, watchdog_timeout=300, soft_watchdog_timeout=None, sleep_on_idle=False, use_ray=False, custom_sigquit_handler=None, numa_node=None, gc_threshold=None, host='0.0.0.0', port=30000, fastapi_root_path='', smg_grpc_mode=False, grpc_mode=False, grpc_port=None, sidecar=None, sidecar_args=None, skip_server_warmup=False, warmups=None, enable_http2=False, ssl_keyfile=None, ssl_certfile=None, ssl_ca_certs=None, ssl_keyfile_password=None, enable_ssl_refresh=False, api_key=None, admin_api_key=None, served_model_name='kimi-k3', weight_version='default', chat_template=None, hf_chat_template_name=None, completion_template=None, file_storage_path='sglang_storage', enable_cache_report=False, reasoning_parser=None, default_chat_template_kwargs=None, strip_thinking_cache=False, enable_strict_thinking=False, tool_call_parser=None, tool_server=None, sampling_defaults='model', asr_max_buffer_seconds=60, asr_max_concurrent_sessions=32, preferred_sampling_params=None, allow_auto_truncate=False, stream_interval=1, batch_notify_size=16, stream_response_default_include_usage=False, incremental_streaming_output=False, enable_streaming_session=False, enable_session_radix_cache=False, log_level='info', log_level_http=None, log_requests=False, log_requests_level=2, log_requests_format='text', log_requests_target=None, uvicorn_access_log_exclude_prefixes=[], crash_dump_folder=None, show_time_cost=False, enable_metrics=False, smg_http_sidecar_port=None, enable_mfu_metrics=False, enable_metrics_for_all_schedulers=False, load_snapshot_publish_interval=15, tokenizer_metrics_custom_labels_header='x-custom-labels', tokenizer_metrics_allowed_custom_labels=None, extra_metric_labels=None, bucket_time_to_first_token=None, bucket_inter_token_latency=None, bucket_e2e_request_latency=None, prompt_tokens_buckets=None, generation_tokens_buckets=None, gc_warning_threshold_secs=0.0, decode_log_interval=40, enable_request_time_stats_logging=False, kv_events_config=None, enable_forward_pass_metrics=False, forward_pass_metrics_worker_id='', forward_pass_metrics_ipc_name=None, enable_trace=False, trace_modules='request', otlp_traces_endpoint='localhost:4317', export_metrics_to_file=False, export_metrics_to_file_dir=None, stat_loggers=None, constrained_json_whitespace_pattern=None, constrained_json_disable_any_whitespace=False, attention_backend='triton', decode_attention_backend=None, prefill_attention_backend=None, sampling_backend='flashinfer', grammar_backend='xgrammar', radix_cache_backend=None, mm_attention_backend=None, fp8_gemm_runner_backend='auto', fp4_gemm_runner_backend='auto', bf16_gemm_backend='auto', dsa_prefill_backend=None, dsa_decode_backend=None, dsa_paged_mqa_logits_backend='auto', dsa_topk_backend='sgl-kernel', disable_flashinfer_autotune=False, flashinfer_autotune_skip_ops=None, mamba_backend='triton', cuda_graph_config=CudaGraphConfig(decode=PhaseConfig(backend='full', max_bs=16, bs=[1, 2, 4, 8, 12, 16], tc_compiler='eager', full_prefill_max_req=None), prefill=PhaseConfig(backend='disabled', max_bs=2048, bs=[4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], tc_compiler='eager', full_prefill_max_req=None)), cuda_graph_backend_decode=None, cuda_graph_backend_prefill=None, cuda_graph_max_bs_decode=16, cuda_graph_max_bs_prefill=None, cuda_graph_bs_decode=None, cuda_graph_bs_prefill=None, cuda_graph_tc_compiler=None, disable_prefill_cuda_graph=False, disable_decode_cuda_graph=False, disable_cuda_graph=False, disable_cuda_graph_padding=False, enable_profile_cuda_graph=False, enable_cudagraph_gc=False, debug_cuda_graph=False, enable_layerwise_nvtx_marker=False, enable_nccl_nvls=False, enable_symm_mem=False, triton_attention_reduce_in_fp32=False, triton_attention_num_kv_splits=8, triton_attention_split_tile_size=None, flashinfer_mla_disable_ragged=False, enable_fused_qk_norm_rope=False, enable_precise_embedding_interpolation=False, enable_fused_moe_sum_all_reduce=False, enable_deepseek_v4_fp4_indexer=False, disable_custom_all_reduce=False, enable_mscclpp=False, enable_torch_symm_mem=False, enable_scattered_sconv=False, pre_warm_nccl=False, enable_quant_communications=False, enable_flashinfer_allreduce_fusion=False, enforce_disable_flashinfer_allreduce_fusion=False, flashinfer_allreduce_fusion_backend=None, enable_aiter_allreduce_fusion=False, enable_torch_compile=False, enable_torch_compile_debug_mode=False, torch_compile_max_bs=32, torchao_config='', speculative_algorithm=None, speculative_draft_model_path=None, speculative_draft_model_revision=None, speculative_draft_load_format=None, speculative_num_steps=None, speculative_eagle_topk=None, speculative_num_draft_tokens=None, speculative_dflash_block_size=None, speculative_dspark_block_size=None, speculative_dspark_sps_table_path=None, speculative_dspark_confidence_sts_path=None, speculative_dspark_align_verify_tokens_to_graph_tier=False, speculative_accept_threshold_single=1.0, speculative_accept_threshold_acc=1.0, speculative_use_rejection_sampling=False, speculative_token_map=None, speculative_attention_mode='prefill', speculative_draft_attention_backend=None, speculative_draft_window_size=None, speculative_moe_runner_backend='marlin', speculative_moe_a2a_backend=None, speculative_draft_model_quantization=None, speculative_skip_dp_mlp_sync=False, enable_multi_layer_eagle=False, speculative_adaptive=False, speculative_adaptive_config=None, decoupled_spec_bind_endpoint=None, decoupled_spec_connect_endpoints=None, decoupled_spec_rank=None, decoupled_spec_role='null', spec_trace_dir=None, speculative_ngram_min_bfs_breadth=1, speculative_ngram_max_bfs_breadth=10, speculative_ngram_match_type='BFS', speculative_ngram_max_trie_depth=18, speculative_ngram_capacity=10000000, speculative_ngram_external_corpus_path=None, speculative_ngram_external_sam_budget=0, speculative_ngram_external_corpus_max_tokens=10000000, ep_size=32, moe_a2a_backend='none', moe_runner_backend='marlin', flashinfer_mxfp4_moe_precision='default', deepep_mode='auto', fuseep_mode=2, deepep_dispatcher_output_dtype='auto', ep_num_redundant_experts=0, ep_dispatch_algorithm=None, init_expert_location='trivial', enable_eplb=False, eplb_algorithm='auto', eplb_rebalance_num_iterations=1000, eplb_rebalance_layers_per_chunk=None, eplb_min_rebalancing_utilization_threshold=1.0, expert_distribution_recorder_mode=None, expert_distribution_recorder_buffer_size=1000, enable_expert_distribution_metrics=False, deepep_config=None, moe_dense_tp_size=None, elastic_ep_backend=None, enable_elastic_expert_backup=False, mooncake_ib_device=None, enable_waterfill=False, ep_join_mode=None, ep_join_rank_offset=0, elastic_ep_initial_size=None, max_ep_size=None, elastic_ep_scale_timeout=600, elastic_ep_rejoin=False, disable_flashinfer_cutlass_moe_fp4_allgather=False, disable_shared_experts_fusion=False, enforce_shared_experts_fusion=False, max_mamba_cache_size=None, mamba_ssm_dtype=None, mamba_max_states_per_path=-1, enable_mamba_cache_stochastic_rounding=False, mamba_cache_philox_rounds=0, mamba_full_memory_ratio=0.21, mamba_radix_cache_strategy='extra_buffer_lazy', uses_mamba_radix_cache=False, mamba_track_interval=256, enable_int8_mamba_checkpoint=False, int8_mamba_ckpt_size=None, linear_attn_backend='triton', linear_attn_decode_backend=None, linear_attn_prefill_backend=None, linear_attn_verify_backend=None, enable_linear_replayssm=False, linear_replayssm_cache_len=16, enable_linear_replayssm_spec=False, enable_hierarchical_cache=False, hicache_ratio=2.0, hicache_size=0, hicache_write_policy='write_through', hicache_io_backend='kernel', hicache_mem_layout='page_first', hicache_storage_backend=None, hicache_storage_prefetch_policy='timeout', hicache_storage_backend_extra_config=None, enable_hisparse=False, hisparse_config=None, enable_broadcast_mm_inputs_process=False, enable_prefix_mm_cache=False, mm_enable_dp_encoder=False, mm_process_config={}, mm_processor_worker_num=0, mm_io_worker_num=0, limit_mm_data_per_request=None, enable_mm_global_cache=False, disable_fast_image_processor=False, mm_feature_transport='cpu', keep_mm_feature_on_device=False, enable_lora=None, enable_lora_overlap_loading=None, max_lora_rank=None, lora_target_modules=None, lora_paths=None, max_loaded_loras=None, max_loras_per_batch=8, lora_eviction_policy='lru', lora_backend='csgmv', max_lora_chunk_size=16, experts_shared_outer_loras=None, lora_use_virtual_experts=False, lora_strict_loading=False, lora_drain_wait_threshold=0.0, enable_two_batch_overlap=False, enable_single_batch_overlap=False, tbo_token_distribution_threshold=0.48, cpu_offload_gb=0, offload_group_size=-1, offload_num_in_group=1, offload_prefetch_step=1, offload_mode='cpu', enable_lmcache=False, lmcache_config_file=None, enable_flexkv=False, flexkv_config_file=None, kt_weight_path=None, kt_method='AMXINT4', kt_cpuinfer=None, kt_threadpool_count=2, kt_num_gpu_experts=None, kt_max_deferred_experts_per_token=None, dllm_algorithm=None, dllm_algorithm_config=None, dllm_fdfo=True, disaggregation_mode='decode', disaggregation_transfer_backend='mooncake', disaggregation_bootstrap_port=28800, disaggregation_ib_device='mlx5_0,mlx5_1,mlx5_2,mlx5_3', disaggregation_decode_enable_radix_cache=False, disaggregation_decode_enable_offload_kvcache=False, num_reserved_decode_tokens=512, disaggregation_decode_extra_slots=0, disaggregation_decode_polling_interval=1, optimistic_prefill_attempts=0, encoder_only=False, language_only=False, encoder_transfer_backend='zmq_to_tokenizer', encoder_urls=[], encoder_bootstrap_port=8997, encoder_register_urls=[], enable_adaptive_dispatch_to_encoder=False, enable_pdmux=False, pdmux_config_path=None, sm_group_num=8, custom_weight_loader=[], weight_loader_disable_mmap=False, weight_loader_prefetch_checkpoints=False, weight_loader_prefetch_num_threads=4, weight_loader_drop_cache_after_load=False, remote_instance_weight_loader_seed_instance_ip=None, remote_instance_weight_loader_seed_instance_service_port=None, remote_instance_weight_loader_send_weights_group_ports=None, remote_instance_weight_loader_backend='nccl', remote_instance_weight_loader_start_seed_via_transfer_engine=False, engine_info_bootstrap_port=6789, modelexpress_config=None, download_dir=None, model_checksum=None, delete_ckpt_after_loading=False, decrypted_config_file=None, decrypted_draft_config_file=None, checkpoint_engine_wait_weights_before_ready=False, enable_prefill_delayer=False, prefill_delayer_max_delay_passes=30, prefill_delayer_token_usage_low_watermark=None, prefill_delayer_forward_passes_buckets=None, prefill_delayer_wait_seconds_buckets=None, prefill_delayer_queue_min_ratio=None, prefill_delayer_max_delay_ms=None, min_free_slots_delay=None, enable_deterministic_inference=False, rl_on_policy_target=None, kv_canary='none', kv_canary_real_data='none', kv_canary_sweep_interval=0, enable_dynamic_batch_tokenizer=False, dynamic_batch_tokenizer_batch_size=32, dynamic_batch_tokenizer_batch_timeout=0.002, enable_tokenizer_batch_encode=False, disable_tokenizer_batch_decode=False, debug_tensor_dump_output_folder=None, debug_tensor_dump_layers=None, debug_tensor_dump_input_file=None, enable_memory_saver=False, enable_weights_cpu_backup=False, enable_draft_weights_cpu_backup=False, enable_custom_logit_processor=False, enable_return_hidden_states=False, enable_return_routed_experts=False, enable_return_indexer_topk=False, disable_outlines_disk_cache=False, enable_mis=False, weight_cache_mode='off', weight_cache_socket=None, weight_cache_timeout=1800, forward_hooks=None, msprobe_dump_config=None)
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:36:05] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:16 TP30 EP30] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:16 TP30 EP30] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:19 TP30 EP30] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:19.321749   241 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:19.321810   241 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:19.321820   241 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.8 port: 12001
+I0827 04:36:19.321856   241 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.8:16569
+I0827 04:36:19.321940   241 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:19.343153   241 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:19.347820   241 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:19.373265   241 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:19.380316   241 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:19.383023   241 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:19.383050   241 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:19.383101   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:19.390141   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:19.391583   241 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:08
+I0827 04:36:19.391615   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:19.417495   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:19.418977   241 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:08
+I0827 04:36:19.419003   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:19.445492   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:19.446861   241 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:08
+I0827 04:36:19.446884   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:19.473483   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:19.474817   241 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:08
+I0827 04:36:19.474922   241 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:19 TP30 EP30] Init torch distributed begin.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:20 TP27 EP27] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:20 TP27 EP27] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:21 TP25 EP25] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:21 TP25 EP25] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:21 TP26 EP26] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:21 TP26 EP26] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:21 TP28 EP28] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:21 TP28 EP28] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:36:21 TP24 EP24] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:21 TP24 EP24] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:21 TP29 EP29] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:21 TP29 EP29] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:21 TP31 EP31] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:36:21 TP31 EP31] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:36:22 TP27 EP27] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:22.734625   238 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:22.734687   238 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:22.734699   238 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.8 port: 12001
+I0827 04:36:22.734735   238 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.8:16566
+I0827 04:36:22.734848   238 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:22.760079   238 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:22.768187   238 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:22.777189   238 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:22.786154   238 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:22.789806   238 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:22.789830   238 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:22.789880   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:22.797986   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:22.799417   238 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:08
+I0827 04:36:22.799446   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:22.825555   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:22.826912   238 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:08
+I0827 04:36:22.826946   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:22.853536   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:22.854885   238 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:08
+I0827 04:36:22.854909   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:22.881553   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:22.882910   238 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:08
+I0827 04:36:22.883030   238 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:22 TP27 EP27] Init torch distributed begin.
+[2026-08-27 04:36:22 TP25 EP25] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:22 TP28 EP28] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:22 TP26 EP26] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:23 TP29 EP29] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:36:23 TP24 EP24] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:23.476856   236 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:23.476951   236 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:23.476967   236 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.8 port: 12001
+I0827 04:36:23.477013   236 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.8:15425
+I0827 04:36:23.477150   236 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:23.477173   239 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:23.477272   239 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:23.477288   239 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.8 port: 12001
+I0827 04:36:23.477344   239 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.8:15710
+I0827 04:36:23.477483   239 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:23.483197   236 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:23.483803   239 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:23.488940   236 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:23.489802   239 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:23.494866   236 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:23.495663   239 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:23.500973   236 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:23.501806   239 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:23.503927   236 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:23.503952   236 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:23.503998   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.504750   239 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:23.504770   239 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:23.504812   239 rdma_context.cpp:205] Using SIEVE endpoint store
+[2026-08-27 04:36:23 TP31 EP31] Multimodal data loading enabled with 16 worker threads (auto).
+I0827 04:36:23.510515   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:23.511322   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:23.511830   236 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:08
+I0827 04:36:23.511855   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.512616   239 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:08
+I0827 04:36:23.512640   239 rdma_context.cpp:205] Using SIEVE endpoint store
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:23.522899   237 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:23.523005   237 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:23.523020   237 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.8 port: 12001
+I0827 04:36:23.523065   237 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.8:16748
+I0827 04:36:23.523202   237 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:23.529469   237 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:23.535485   237 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:23.537460   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:23.537528   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:23.538903   236 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:08
+I0827 04:36:23.538913   239 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:08
+I0827 04:36:23.538928   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.538945   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.541347   237 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:23.546439   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:23.547672   237 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:23.547796   239 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:08
+I0827 04:36:23.547820   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.550041   237 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:23.550061   237 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:23.550102   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.555119   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:23.556490   239 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:08
+I0827 04:36:23.556598   239 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:23 TP28 EP28] Init torch distributed begin.
+I0827 04:36:23.557152   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:23.558562   237 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:08
+I0827 04:36:23.558589   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.565423   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:23.566766   236 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:08
+I0827 04:36:23.566788   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.589463   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:23.590811   237 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:08
+I0827 04:36:23.590835   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.593444   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:23.594800   236 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:08
+I0827 04:36:23.594898   236 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:23 TP25 EP25] Init torch distributed begin.
+I0827 04:36:23.598161   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:23.599563   237 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:08
+I0827 04:36:23.599589   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.625564   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:23.626922   237 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:08
+I0827 04:36:23.627038   237 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:23 TP26 EP26] Init torch distributed begin.
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:23.899124   240 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:23.899245   240 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:23.899261   240 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.8 port: 12001
+I0827 04:36:23.899313   240 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.8:15127
+I0827 04:36:23.899492   240 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:23.907753   235 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:23.907863   235 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:23.907878   235 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.8 port: 12001
+I0827 04:36:23.907927   235 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.8:15330
+I0827 04:36:23.908092   235 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:23.914011   235 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:23.919618   235 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:23.925108   235 topology.cpp:127] Device mlx5_2 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:36:23.925450   242 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:36:23.925558   242 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:36:23.925576   242 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.8 port: 12001
+I0827 04:36:23.925631   242 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.8:16235
+I0827 04:36:23.925693   240 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:23.925761   242 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:36:23.930419   242 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:36:23.933247   240 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:23.934639   242 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:36:23.939190   242 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:23.940605   240 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:36:23.943866   242 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:23.945531   242 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:23.945554   242 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:23.945595   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.947860   240 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:23.949263   235 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:36:23.951108   240 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:23.951148   240 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:23.951216   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.952417   235 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:36:23.952445   235 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:36:23.952502   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.958555   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:23.959836   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:23.960163   240 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:08
+I0827 04:36:23.960197   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.961452   235 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:08
+I0827 04:36:23.961483   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.969228   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:23.969322   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:36:23.970860   235 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:08
+I0827 04:36:23.970890   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.971050   242 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:08
+I0827 04:36:23.971084   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.979120   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:23.980706   242 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:08
+I0827 04:36:23.980736   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:23.989722   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:36:23.991313   240 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:08
+I0827 04:36:23.991346   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:24.001633   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:24.003211   235 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:08
+I0827 04:36:24.003237   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:24.005753   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:24.007426   242 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:08
+I0827 04:36:24.007457   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:24.021715   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:36:24.023293   240 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:08
+I0827 04:36:24.023322   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:36:24.034039   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:24.034071   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:24.035696   235 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:08
+I0827 04:36:24.035816   235 transfer_engine_impl.cpp:388] installTransport, type=rdma
+I0827 04:36:24.035794   242 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:08
+I0827 04:36:24.035987   242 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:24 TP24 EP24] Init torch distributed begin.
+[2026-08-27 04:36:24 TP31 EP31] Init torch distributed begin.
+I0827 04:36:24.049829   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:36:24.051388   240 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:08
+I0827 04:36:24.051532   240 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:36:24 TP29 EP29] Init torch distributed begin.
+[2026-08-27 04:36:48 TP31 EP31] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP30 EP30] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP28 EP28] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP29 EP29] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP27 EP27] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP26 EP26] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP25 EP25] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP24 EP24] CustomAllreduce is disabled because this process group spans across nodes.
+[2026-08-27 04:36:48 TP31 EP31] Init torch distributed ends. elapsed=24.20 s, mem usage=0.40 GB
+[2026-08-27 04:36:48 TP30 EP30] Init torch distributed ends. elapsed=28.76 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP29 EP29] Init torch distributed ends. elapsed=24.18 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP28 EP28] Init torch distributed ends. elapsed=24.68 s, mem usage=0.40 GB
+[2026-08-27 04:36:48 TP27 EP27] Init torch distributed ends. elapsed=25.35 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP26 EP26] Init torch distributed ends. elapsed=24.61 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP25 EP25] Init torch distributed ends. elapsed=24.64 s, mem usage=0.41 GB
+[2026-08-27 04:36:48 TP24 EP24] Init torch distributed ends. elapsed=24.20 s, mem usage=0.44 GB
+[2026-08-27 04:36:48 TP28 EP28] Load weight begin. avail mem=82.19 GB
+[2026-08-27 04:36:48 TP27 EP27] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP30 EP30] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP24 EP24] Load weight begin. avail mem=82.15 GB
+[2026-08-27 04:36:48 TP29 EP29] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP26 EP26] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP25 EP25] Load weight begin. avail mem=82.18 GB
+[2026-08-27 04:36:48 TP28 EP28] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP30 EP30] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP27 EP27] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP29 EP29] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP24 EP24] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP26 EP26] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP25 EP25] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP31 EP31] Load weight begin. avail mem=82.19 GB
+[2026-08-27 04:36:48 TP31 EP31] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:36:48 TP27 EP27] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP28 EP28] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP24 EP24] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP27 EP27] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP30 EP30] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP29 EP29] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP26 EP26] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP25 EP25] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP28 EP28] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP24 EP24] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP29 EP29] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP31 EP31] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:36:48 TP30 EP30] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP26 EP26] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP25 EP25] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:48 TP31 EP31] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=marlin, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:36:50 TP29 EP29] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP28 EP28] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP30 EP30] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP26 EP26] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP25 EP25] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP31 EP31] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP27 EP27] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:36:50 TP24 EP24] multimem all-gather disabled because the TP group spans across nodes.
+[2026-08-27 04:38:54 TP26 EP26] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:54 TP27 EP27] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:54 TP24 EP24] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:54 TP26 EP26] Load weight end. elapsed=125.61 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:54 TP29 EP29] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:54 TP27 EP27] Load weight end. elapsed=125.61 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:54 TP24 EP24] Load weight end. elapsed=125.61 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.58 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:54 TP29 EP29] Load weight end. elapsed=125.61 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:54 TP25 EP25] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:54 TP25 EP25] Load weight end. elapsed=125.64 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:54 TP28 EP28] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:54 TP28 EP28] Load weight end. elapsed=125.69 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:54 TP30 EP30] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:54 TP30 EP30] Load weight end. elapsed=125.71 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:38:54 TP31 EP31] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:38:54 TP31 EP31] Load weight end. elapsed=125.77 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=23.61 GB, mem usage=58.57 GB.
+[2026-08-27 04:39:01 TP26 EP26] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP26 EP26] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP30 EP30] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP30 EP30] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP28 EP28] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP28 EP28] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP27 EP27] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP27 EP27] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP25 EP25] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP25 EP25] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP31 EP31] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP31 EP31] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP24 EP24] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP24 EP24] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:01 TP29 EP29] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:39:01 TP29 EP29] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:39:02 TP29 EP29] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP28 EP28] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP26 EP26] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP27 EP27] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP30 EP30] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP25 EP25] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP31 EP31] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP24 EP24] max_running_requests is capped to 181 by the mamba state cache (max_mamba_cache_size=181, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:39:02 TP29 EP29] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP30 EP30] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP28 EP28] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP26 EP26] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP27 EP27] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP25 EP25] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP24 EP24] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP31 EP31] Mamba Cache is allocated. max_mamba_cache_size: 181, conv_state size: 0.08GB, ssm_state size: 2.30GB 
+[2026-08-27 04:39:02 TP29 EP29] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP29 EP29] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP28 EP28] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP28 EP28] Memory pool end. avail mem=9.18 GB
+[2026-08-27 04:39:02 TP27 EP27] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP24 EP24] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP26 EP26] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP27 EP27] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP31 EP31] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP24 EP24] Memory pool end. avail mem=9.14 GB
+[2026-08-27 04:39:02 TP26 EP26] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP31 EP31] Memory pool end. avail mem=9.18 GB
+[2026-08-27 04:39:02 TP30 EP30] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP25 EP25] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 440256, KV size: 11.34 GB
+[2026-08-27 04:39:02 TP30 EP30] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP25 EP25] Memory pool end. avail mem=9.17 GB
+[2026-08-27 04:39:02 TP29 EP29] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP29 EP29] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP24 EP24] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP24 EP24] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:39:02 TP28 EP28] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP28 EP28] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.17 GB
+[2026-08-27 04:39:02 TP25 EP25] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP31 EP31] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP25 EP25] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP31 EP31] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.17 GB
+[2026-08-27 04:39:02 TP26 EP26] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP26 EP26] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:39:02 TP27 EP27] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP27 EP27] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.13 GB
+[2026-08-27 04:39:02 TP30 EP30] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:39:02 TP30 EP30] Capture target decode CUDA graph begin. backend=full, num_tokens_per_req=1, bs=[1, 2, 4, 8, 12, 16], avail mem=9.16 GB
+[2026-08-27 04:42:56 TP30 EP30] Capture target decode CUDA graph end. elapsed=233.69 s, mem usage=0.56 GB, avail mem=8.59 GB.
+[2026-08-27 04:42:56 TP30 EP30] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP27 EP27] Capture target decode CUDA graph end. elapsed=233.73 s, mem usage=0.62 GB, avail mem=8.51 GB.
+[2026-08-27 04:42:56 TP27 EP27] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP31 EP31] Capture target decode CUDA graph end. elapsed=233.74 s, mem usage=0.62 GB, avail mem=8.55 GB.
+[2026-08-27 04:42:56 TP31 EP31] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP26 EP26] Capture target decode CUDA graph end. elapsed=233.76 s, mem usage=0.56 GB, avail mem=8.59 GB.
+[2026-08-27 04:42:56 TP26 EP26] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP29 EP29] Capture target decode CUDA graph end. elapsed=233.82 s, mem usage=0.58 GB, avail mem=8.58 GB.
+[2026-08-27 04:42:56 TP29 EP29] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP28 EP28] Capture target decode CUDA graph end. elapsed=233.79 s, mem usage=0.60 GB, avail mem=8.56 GB.
+[2026-08-27 04:42:56 TP28 EP28] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP24 EP24] Capture target decode CUDA graph end. elapsed=233.88 s, mem usage=0.60 GB, avail mem=8.52 GB.
+[2026-08-27 04:42:56 TP24 EP24] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:42:56 TP25 EP25] Capture target decode CUDA graph end. elapsed=233.89 s, mem usage=0.58 GB, avail mem=8.58 GB.
+[2026-08-27 04:42:56 TP25 EP25] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+W0827 04:42:57.132246  2431 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5913c7c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.132292  2432 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58be65c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.132460  2435 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5af7a600, len: 185344: Operation not permitted [1]
+W0827 04:42:57.132488  2433 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58ed5900, len: 23168: Operation not permitted [1]
+W0827 04:42:57.132501  2436 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5afa7a40, len: 185344: Operation not permitted [1]
+W0827 04:42:57.132512  2437 memory_location.cpp:73] Failed to get NUMA node, addr: 0x590583c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.132552  2434 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58c15080, len: 23168: Operation not permitted [1]
+W0827 04:42:57.132607  2439 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59046600, len: 23168: Operation not permitted [1]
+W0827 04:42:57.132640  2438 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5afd4e80, len: 46336: Operation not permitted [1]
+W0827 04:42:57.132716  2440 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58e5fc80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.276726  2580 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a339340, len: 23168: Operation not permitted [1]
+W0827 04:42:57.276731  2579 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59705880, len: 23168: Operation not permitted [1]
+W0827 04:42:57.276762  2581 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a32b080, len: 23168: Operation not permitted [1]
+W0827 04:42:57.276839  2582 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a331fc0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.276887  2583 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5bb81840, len: 185344: Operation not permitted [1]
+W0827 04:42:57.276939  2584 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5bbaec80, len: 185344: Operation not permitted [1]
+W0827 04:42:57.276983  2585 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59b7b6c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.277042  2587 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59953f00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.277055  2586 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59517380, len: 46336: Operation not permitted [1]
+W0827 04:42:57.277086  2588 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59b87640, len: 23168: Operation not permitted [1]
+W0827 04:42:57.448414  2727 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a4bd3c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.448436  2728 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ac9eb00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.448488  2729 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a2fe240, len: 23168: Operation not permitted [1]
+W0827 04:42:57.448798  2735 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a6135c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.448812  2734 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59ed25c0, len: 46336: Operation not permitted [1]
+W0827 04:42:57.448840  2733 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a3c3080, len: 23168: Operation not permitted [1]
+W0827 04:42:57.448849  2736 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a181e80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.451195  2730 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5acaf240, len: 23168: Operation not permitted [1]
+W0827 04:42:57.451217  2731 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5c61b840, len: 185344: Operation not permitted [1]
+W0827 04:42:57.455173  2732 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5c648c80, len: 185344: Operation not permitted [1]
+W0827 04:42:57.460536  2816 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a391f40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.460651  2817 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ae89700, len: 23168: Operation not permitted [1]
+W0827 04:42:57.460843  2819 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ae0fa00, len: 23168: Operation not permitted [1]
+W0827 04:42:57.460901  2826 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a4e2c80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.460979  2830 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ae7a300, len: 23168: Operation not permitted [1]
+W0827 04:42:57.461052  2821 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5c6d6880, len: 185344: Operation not permitted [1]
+W0827 04:42:57.461104  2820 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a22ed40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.462260  2823 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5c703cc0, len: 185344: Operation not permitted [1]
+W0827 04:42:57.463069  2827 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4678c140, len: 46336: Operation not permitted [1]
+W0827 04:42:57.463085  2829 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a645e80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.466511  2955 memory_location.cpp:73] Failed to get NUMA node, addr: 0x597592c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.466544  2958 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59908b40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.466676  2960 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5a2f9c80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.466849  2953 memory_location.cpp:73] Failed to get NUMA node, addr: 0x599f9a80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.467304  2964 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5b9e4d40, len: 185344: Operation not permitted [1]
+W0827 04:42:57.470024  2966 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5993fa80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.471112  2970 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59898900, len: 23168: Operation not permitted [1]
+W0827 04:42:57.471313  2962 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5b9b7900, len: 185344: Operation not permitted [1]
+W0827 04:42:57.475107  2968 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5991bac0, len: 46336: Operation not permitted [1]
+W0827 04:42:57.475121  2972 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59b5ea40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.493731  3089 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59564740, len: 23168: Operation not permitted [1]
+W0827 04:42:57.493901  3097 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5b355380, len: 185344: Operation not permitted [1]
+W0827 04:42:57.494069  3104 memory_location.cpp:73] Failed to get NUMA node, addr: 0x45dbac00, len: 46336: Operation not permitted [1]
+W0827 04:42:57.494072  3105 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5902f500, len: 23168: Operation not permitted [1]
+W0827 04:42:57.494150  3094 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59498600, len: 23168: Operation not permitted [1]
+W0827 04:42:57.494194  3099 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5b3827c0, len: 185344: Operation not permitted [1]
+W0827 04:42:57.499953  3107 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59486740, len: 23168: Operation not permitted [1]
+W0827 04:42:57.502707  3102 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59016c80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.504236  3091 memory_location.cpp:73] Failed to get NUMA node, addr: 0x594d2200, len: 23168: Operation not permitted [1]
+W0827 04:42:57.504359  3095 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59354e40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.514688  3200 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5885e940, len: 23168: Operation not permitted [1]
+W0827 04:42:57.514716  3202 memory_location.cpp:73] Failed to get NUMA node, addr: 0x587d55c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.514720  3201 memory_location.cpp:73] Failed to get NUMA node, addr: 0x586bd5c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.514801  3204 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5aa56c00, len: 185344: Operation not permitted [1]
+W0827 04:42:57.514810  3205 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5aa84040, len: 185344: Operation not permitted [1]
+W0827 04:42:57.514727  3203 memory_location.cpp:73] Failed to get NUMA node, addr: 0x44d09c40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.514963  3206 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58692200, len: 23168: Operation not permitted [1]
+W0827 04:42:57.515017  3209 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58865100, len: 23168: Operation not permitted [1]
+W0827 04:42:57.515022  3207 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58ae8bc0, len: 46336: Operation not permitted [1]
+W0827 04:42:57.515030  3208 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58683a40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.548924  3433 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5895be80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.548955  3434 memory_location.cpp:73] Failed to get NUMA node, addr: 0x593aa600, len: 23168: Operation not permitted [1]
+W0827 04:42:57.549010  3435 memory_location.cpp:73] Failed to get NUMA node, addr: 0x59399e80, len: 23168: Operation not permitted [1]
+W0827 04:42:57.549088  3436 memory_location.cpp:73] Failed to get NUMA node, addr: 0x58a4fa40, len: 23168: Operation not permitted [1]
+W0827 04:42:57.549110  3437 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ab36800, len: 185344: Operation not permitted [1]
+W0827 04:42:57.549167  3438 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ab63c40, len: 185344: Operation not permitted [1]
+W0827 04:42:57.551182  3439 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5936aac0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.551239  3441 memory_location.cpp:73] Failed to get NUMA node, addr: 0x455ad780, len: 23168: Operation not permitted [1]
+W0827 04:42:57.551265  3442 memory_location.cpp:73] Failed to get NUMA node, addr: 0x594169c0, len: 23168: Operation not permitted [1]
+W0827 04:42:57.551280  3440 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5853c980, len: 46336: Operation not permitted [1]
+[2026-08-27 04:42:59] Dummy health check server started in background thread at 0.0.0.0:30000
+I0827 04:48:13.997741  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:13.998107  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.014123  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.014460  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.045612  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.045893  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.063987  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.064270  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.072747  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.073004  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.082927  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.089962  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.090194  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.112046  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.112280  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.112514  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.115789  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.119459  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.119691  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.119886  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.120070  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.120432  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.120639  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.120857  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.127898  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.128136  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.153139  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.153350  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.170001  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.170248  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.170490  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.170712  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.171002  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.171211  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.171445  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.185374  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.185580  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.185820  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.186059  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.186275  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.218560  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.218878  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.219130  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.219380  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.239995  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.240248  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.240523  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.240849  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.241052  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.241318  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.241586  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.241847  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.242057  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.242271  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.242539  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.256910  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.258225  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.261878  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.262174  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.262421  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.262710  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.263056  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.263356  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.263574  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.265990  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.267298  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.281507  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.281759  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.281965  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.282989  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.289211  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.290711  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.291896  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.292208  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.292472  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.292713  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.292904  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.293118  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.293318  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.301455  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.301728  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.301918  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.302150  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.302408  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.302699  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.302891  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.303146  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.303368  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.303570  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.305574  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.306890  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.314224  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.315574  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.319671  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.319937  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.320156  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.326327  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.326542  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.330421  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.330546  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.330736  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.330998  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.331259  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.331496  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.331717  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.331938  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.332140  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.332340  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.332551  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.333663  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.338346  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.339445  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.339735  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.340694  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.345638  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.360560  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.361940  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.369402  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.369591  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.370901  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.377556  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.384063  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.384287  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.384516  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.385501  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.397218  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.397437  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.397625  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.398703  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.398864  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.399148  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.407058  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.407372  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.407644  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.407857  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.408042  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.408264  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.408543  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.408807  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.411617  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.411850  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.412068  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.414391  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.415768  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.423707  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.425092  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.425122  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.430542  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.430770  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.431028  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.431236  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.431470  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.434294  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.434525  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.435901  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.436133  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.437299  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.446563  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.446866  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.451089  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.453542  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.454408  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.455781  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.467486  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.467700  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.467938  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.468218  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.468482  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.468770  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.469058  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.472146  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.473528  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.480324  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.480634  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.481218  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.483843  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.484179  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.484373  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.485363  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.498565  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.498824  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.499142  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.502869  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.508553  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.511684  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.511938  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.513093  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.523473  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.523744  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.523972  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.524215  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.529819  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.531201  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.539283  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.540769  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.542526  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.542757  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.543026  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.543277  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.543532  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.543817  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.544056  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.544275  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.544507  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.544736  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.544970  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.545243  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.545476  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.545689  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.546013  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.551106  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.551363  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.551533  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.551725  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.551961  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.552152  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.552357  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.552618  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.555388  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.555505  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.558743  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.564610  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.564857  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.569089  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.570464  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.581254  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.581478  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.586069  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.586848  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.587337  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.589694  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.589882  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.590091  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.590329  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.590632  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.590873  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.591192  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.591454  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.591703  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.591930  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.592170  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.592396  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.592653  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.595007  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.596319  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.597950  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.599296  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.609437  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.609653  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.609938  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.610687  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.621923  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.622153  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.622318  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.623299  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.648794  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.649142  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.664134  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.664382  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.669473  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.669734  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.672606  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.672832  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.673055  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.673255  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.673466  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.673666  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.676374  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.676550  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.681711  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.689419  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.692066  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.692279  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.694999  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.695231  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.695472  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.698184  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.709776  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.717428  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.720180  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.720382  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.720589  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.720793  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.731559  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.731812  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.731972  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.742295  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.750133  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.750379  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.750625  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.750792  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.778556  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.778743  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.778970  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.779207  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.789633  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.789860  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.790076  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.795363  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.795607  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.795882  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.796195  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.796408  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.796597  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.796849  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.797055  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.797358  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.797609  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.800431  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.808163  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.808368  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.808629  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.808866  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.811561  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.811780  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.828982  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.830327  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.832255  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.835093  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.835287  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.835476  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.836141  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.840837  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.841059  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.841271  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.841498  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.841699  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.841900  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.842151  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.842352  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.842628  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.842826  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.843127  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.843389  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.856551  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.856762  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.856982  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.857259  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.857627  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.858045  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.863099  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.863319  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.882516  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.882750  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.890803  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.891052  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.891294  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.892642  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.892895  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.893175  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.893406  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.893621  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.894102  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.894142  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.896976  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.897236  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.897420  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.900352  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.902928  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.903223  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.903735  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.904199  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.909134  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.909372  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.913172  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.920277  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.920501  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.920786  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.920931  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.921003  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.921264  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.921265  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.921586  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.943581  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.943809  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.944059  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.944257  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.944458  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.944726  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.944954  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.945147  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.959767  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.960093  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.960309  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.960552  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.960808  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.961053  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.961282  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.961503  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.961782  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.962018  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.962278  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.962548  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.962779  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.963052  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.969125  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.969360  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.969631  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.969812  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.969980  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.970170  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.970402  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.970584  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.970847  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.976580  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.977794  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.981958  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.984225  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.984476  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.998291  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.998525  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.998750  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:14.999001  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.004917  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.005141  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.021842  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.023211  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.028899  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.028923  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.029189  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.029403  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.029655  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.029892  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.030141  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.030356  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.030586  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.030768  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.051952  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.052182  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.052386  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.052623  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.052874  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.053148  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.053403  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.054157  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.056090  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.056403  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.056617  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.056838  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.060328  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.068521  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.071317  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.071501  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.071699  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.077070  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.080605  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.080899  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.081156  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.081415  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.081660  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.081939  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.082155  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.082402  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.082623  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.082731  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.082814  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.083032  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.085912  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.086201  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.096980  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.098359  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.099546  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.102448  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.102720  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.102989  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.103223  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.103464  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.104277  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.108438  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.108711  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.108911  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.109165  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.109467  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.109733  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.110004  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.110239  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.110486  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.110714  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.110724  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.114183  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.122400  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.122675  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.122920  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.123195  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.123414  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.123673  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.123876  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.126001  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.126210  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.126441  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.126644  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.126863  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.127283  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.131070  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.131325  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.131569  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.131783  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.132396  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.134912  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.135169  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.135370  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.136338  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.140527  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.191358  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.191627  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.191875  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.192153  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.192342  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.197578  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.202698  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.202907  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.208132  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.208374  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.211109  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.211313  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.211566  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.214330  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.214540  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.214802  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.215395  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.216657  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.220089  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.220728  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.225381  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.227144  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.228049  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.228245  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.231040  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.231236  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.234068  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.240044  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.240254  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.240460  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.241421  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.244509  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.250010  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.252700  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.252844  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.252897  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.253087  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.253091  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.253288  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.253509  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.253962  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.263445  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.263675  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.269212  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.270474  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.273865  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.276260  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.279080  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.279291  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.279481  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.279681  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.282605  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.295027  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.295212  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.295395  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.298802  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.299988  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.300223  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.300441  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.300705  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.304690  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.307312  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.307541  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.308730  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.310832  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.313612  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.318708  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.318970  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.319183  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.319437  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.319680  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.319883  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.320147  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.320412  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.325534  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.325740  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.325922  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.326180  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.328954  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.329154  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.329488  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.341765  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.341984  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.347205  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.347421  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.347626  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.347848  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.348078  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.348287  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.348484  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.348680  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.350585  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.350865  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.356657  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.356881  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.357982  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.388767  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.389058  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.389303  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.399434  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.399714  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.409890  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.410177  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.410370  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.410568  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.415786  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.416013  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.416213  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.416402  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.421667  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.421880  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.434330  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.434568  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.444756  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.444969  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.445178  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.445397  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.445652  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.450814  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.451030  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.451257  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.451485  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.471215  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.471483  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.476500  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.476737  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.476964  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.477159  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.477352  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.477542  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.477756  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.477946  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.498270  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.498482  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.518587  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.518834  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.519048  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.519256  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.519551  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.519752  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.520037  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.520202  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.522962  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.523137  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.523340  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.523535  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.523741  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.523979  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.524163  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.524384  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.536993  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.542232  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.542461  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.542663  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.542858  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.543066  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.543337  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.543718  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.551376  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.551574  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.552415  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.552659  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.552764  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.552887  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.553094  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.595072  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.595356  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.595624  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.595870  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.596087  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.596268  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.596544  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.596803  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.597059  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.597254  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.614900  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.615161  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.615470  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.615666  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.615878  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.616151  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.616405  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.653784  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.654006  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.654214  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.654405  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.654635  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.654825  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.655045  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.655264  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.655457  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.655639  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.655900  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.656116  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.656322  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.656612  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.656821  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.704331  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.704610  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.704798  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.705024  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.705224  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.705487  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.705684  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.705888  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.706089  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.706321  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.706543  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.706796  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.707023  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.707245  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.707461  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.707670  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.707892  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.708086  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.708282  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.728194  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.728391  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.728615  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.728813  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.729079  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.729300  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.729538  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.729822  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.737394  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.737599  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:48:15.738767  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.451416  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.452901  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.467027  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.467362  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.468367  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.468600  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.468681  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.469802  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.473881  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.484158  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.484486  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.484820  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.485107  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.485355  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.485599  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.487003  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.490974  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.493090  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.493911  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.495500  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.501397  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.502763  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.503086  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.503175  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.503513  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.504180  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.504545  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.508555  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.515570  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.516867  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.518400  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.518653  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.518872  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.519848  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.521056  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.527827  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.528055  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.529295  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.544860  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.545089  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.545290  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.545504  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 04:57:14.546707  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.295120  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.296660  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.302729  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.316711  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.318199  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.323024  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.333889  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.334213  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.334910  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.335601  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.336284  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.340574  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.340840  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.342151  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.347932  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.348773  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.349017  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.350518  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.357188  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.358728  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.360765  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.362203  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.363308  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.366835  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.371485  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.371713  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.373103  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.377363  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.377615  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.377725  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.378958  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.379127  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.382095  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.382319  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.383715  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.383874  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.384758  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.388401  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.392554  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.394655  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.396057  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.398231  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.398520  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.399797  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.400022  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.402158  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.402491  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.402809  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.403522  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.404407  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.409164  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.413456  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.414511  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.416069  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.420145  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.421628  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.421854  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.422119  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.423108  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.427431  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.440521  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.440709  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.440948  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.441156  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.441833  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.447186  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:21.448485  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.392441  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.393929  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.404150  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.404464  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:04:41.405719  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.603718  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.605327  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.611670  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:05:39.613114  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.138854  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.139194  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.140391  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.148223  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.149956  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.155593  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:07:33.157107  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.663831  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.665465  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.672602  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.675256  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.675565  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.676985  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.683609  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.684000  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.685106  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.687062  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.688648  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.689873  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.693089  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.697635  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.697898  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.699429  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.700985  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.703440  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.703691  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.704947  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.707213  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.710541  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.712060  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.718225  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.719748  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.721971  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.723291  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.725860  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:08:18.727556  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.461587  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.463192  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.469041  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.485112  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.485399  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.486702  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.491591  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.499243  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.500805  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.507115  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.510190  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.510407  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.511715  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:14:42.517388  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.183045  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.184537  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:18:33.190621  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:47.592581  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:47.604241  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:48.578392  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:48.587992  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:48.597157  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:48.605240  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:49.361904  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:49.370812  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:49.376947  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:49.384886  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.134733  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.145203  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.152640  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.159878  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.185983  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.195700  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.204260  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:51.212427  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.024849  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.026432  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.036523  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.047621  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.048969  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.057524  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.825331  1430 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.834313  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.843958  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:52.852244  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:54.591287  1334 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:54.605118  1378 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:55.477183  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:57.617564  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:57.677883  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:58.316752  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:58.327472  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:21:59.071226  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:06.744064  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:07.467173  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:09.220778  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:09.244503  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:10.161363  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:10.162961  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:10.231134  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:15.442806  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:18.016840  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:18.924312  1421 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:28.097031  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:33.444605  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:35.220340  1364 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:51.309719  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:51.309983  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:51.311292  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:51.317121  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:52.153962  1365 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:59.389873  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:22:59.512159  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:23:07.393066  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:23:08.411403  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:23:09.137790  1300 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:23:09.151702  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
+I0827 05:23:10.920717  1423 rdma_endpoint.cpp:671] Received RDMA ready ACK.
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/logs/manual_mc_master.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/logs/manual_mc_master.log
new file mode 100644
index 0000000..a39916d
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/logs/manual_mc_master.log
@@ -0,0 +1,391 @@
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+I0827 04:28:16.835302    99 master.cpp:1366] Master service started on port 50051, max_threads=16, enable_metric_reporting=1, metrics_port=9003, default_kv_lease_ttl=10000, default_kv_soft_pin_ttl=1800000, allow_evict_soft_pinned_objects=1, eviction_ratio=0.05, eviction_high_watermark_ratio=0.9, enable_ha=0, enable_offload=0, enable_kv_events=0, kv_events_bind_endpoint=, kv_events_backend_id=, offload_on_evict=0, offload_force_evict=0, offloading_queue_limit=50000, offload_cap_ratio=0.5, ha_backend_type=etcd, ha_backend_connstring=, etcd_endpoints=, client_ttl=10, rpc_thread_num=16, rpc_port=50051, rpc_address=0.0.0.0, rpc_interface=, rpc_conn_timeout_seconds=0, rpc_enable_tcp_no_delay=1, rpc protocol=tcp, cluster_id=mooncake_cluster, root_fs_dir=, global_file_segment_size=9223372036854775807, memory_allocator=offset, enable_http_metadata_server=0, http_metadata_server_port=8080, http_metadata_server_host=0.0.0.0, enable_metadata_cleanup_on_timeout=0, put_start_discard_timeout_sec=30, put_start_release_timeout_sec=600, max_total_finished_tasks=10000, max_total_pending_tasks=10000, max_total_processing_tasks=10000, pending_task_timeout_sec=300, processing_task_timeout_sec=300, enable_snapshot=0, enable_snapshot_restore=0, snapshot_interval_seconds=600, snapshot_backup_dir=, snapshot_object_store_type=, snapshot_catalog_store_type=, snapshot_retention_count=2, max_retry_attempts=10, enable_cxl=0, cxl_path=/dev/dax0.0, cxl_size=8589934592
+I0827 04:28:16.866606   108 master_service.cpp:1520] Task cleanup thread started
+I0827 04:28:16.867491    99 master_admin_service.cpp:311] Master admin server started on port 9003
+I0827 04:28:16.869802   111 master_admin_service.cpp:302] Master Admin Metrics: role=standby, state=starting, service_ready=false, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:28:26.870203   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:28:36.870568   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:28:46.870873   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:28:56.871232   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:29:06.871615   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:29:16.871969   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:29:26.872316   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:29:36.872687   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:29:46.873068   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:29:56.873467   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:30:06.873914   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:30:16.874214   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:30:26.874567   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:30:36.874953   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:30:46.875332   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:30:56.875741   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:31:06.876119   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:31:16.876497   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:31:26.876866   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:31:36.877233   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:31:46.877605   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:31:56.877948   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:32:06.878437   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:32:16.879210   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:32:26.879711   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:32:36.880080   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:32:46.880565   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:32:56.881004   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:33:06.881398   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:33:16.881786   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:33:26.882170   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:33:36.882550   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:33:46.882939   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:33:56.883315   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:34:06.883746   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:34:16.884127   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:34:26.884568   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:34:36.884953   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:34:46.885334   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:34:56.885706   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:35:06.886085   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:35:16.886464   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:35:26.886835   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:35:36.887218   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:35:46.887596   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:35:56.887964   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:36:06.888356   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:36:16.888742   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:36:26.889113   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:36:36.889495   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:36:46.889882   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:36:56.890261   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:37:06.890656   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:37:16.891040   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:37:26.891430   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:37:36.891836   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:37:46.892215   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:37:56.892597   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:38:06.892982   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:38:16.893360   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:38:26.893747   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:38:36.894131   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:38:46.894515   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:38:56.894920   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:39:06.895303   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:39:16.895701   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:39:26.896085   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:39:36.896471   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:39:46.896852   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:39:56.897240   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:40:06.897629   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:40:16.898005   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:40:26.898384   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:40:36.898761   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:40:46.899144   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:40:56.899530   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:41:06.899937   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:41:16.900316   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:41:26.900695   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:41:36.901089   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:41:46.901479   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:41:56.901856   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:42:06.902242   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:42:16.902666   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:42:26.903064   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:42:36.903537   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:42:46.903987   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:42:56.904397   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:43:06.904863   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:43:16.905256   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:43:26.905628   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:43:36.905993   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:43:46.906375   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:43:56.906759   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:44:06.907138   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:44:16.907518   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:44:26.907922   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:44:36.908319   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:44:46.908928   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:44:56.909319   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:45:06.909730   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:45:16.910125   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:45:26.910491   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:45:36.910866   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:45:46.911247   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:45:56.911623   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:46:06.911993   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:46:16.912426   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:46:26.912810   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:46:36.913184   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:46:46.913556   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:46:56.913941   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:47:06.914324   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:47:16.914712   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:47:26.915107   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:47:36.915493   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:47:46.915882   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:47:56.916265   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:48:06.916646   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:48:16.917042   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:48:26.917429   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:48:36.917809   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:48:46.918201   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:48:56.918605   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:49:06.918992   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:49:16.919385   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:49:26.919793   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:49:36.920181   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:49:46.920554   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:49:56.920938   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:50:06.921316   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:50:16.921710   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:50:26.922091   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:50:36.922484   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:50:46.922891   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:50:56.923311   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:51:06.923827   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:51:16.924309   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:51:26.924793   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:51:36.925184   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:51:46.925541   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:51:56.925873   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:52:06.926222   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:52:16.926549   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:52:26.926885   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:52:36.927227   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:52:46.927629   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:52:56.927992   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:53:06.928339   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:53:16.928666   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:53:26.928995   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:53:36.929322   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:53:46.929646   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:53:56.930056   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:54:06.930403   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:54:16.930779   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:54:26.931145   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:54:36.931521   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:54:46.931916   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:54:56.932343   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:55:06.932718   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:55:16.933064   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:55:26.933395   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:55:36.933718   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:55:46.934052   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:55:56.934386   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:56:06.934712   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:56:16.935036   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:56:26.935360   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:56:36.935707   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:56:46.936054   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:56:56.936385   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:57:06.936734   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:57:16.937136   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:57:26.937498   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:57:36.937844   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:57:46.938345   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:57:56.938614   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:58:06.938869   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:58:16.939126   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:58:26.939358   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:58:36.939590   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:58:46.939940   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:58:56.940317   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:59:06.940711   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:59:16.941143   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:59:26.941520   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:59:36.941883   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:59:46.942258   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 04:59:56.942601   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:00:06.942938   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:00:16.943261   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:00:26.943506   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:00:36.943737   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:00:46.943951   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:00:56.944247   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:01:06.944607   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:01:16.944896   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:01:26.945246   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:01:36.945606   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:01:46.945940   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:01:56.946230   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:02:06.946504   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:02:16.946818   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:02:26.947096   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:02:36.947449   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:02:46.947767   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:02:56.948025   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:03:06.948321   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:03:16.948577   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:03:26.948939   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:03:36.949285   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:03:46.949551   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:03:56.949826   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:04:06.950160   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:04:16.950445   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:04:26.950799   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:04:36.951154   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:04:46.951506   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:04:56.951869   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:05:06.952240   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:05:16.952581   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:05:26.952893   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:05:36.953207   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:05:46.953542   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:05:56.953862   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:06:06.954190   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:06:16.954527   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:06:26.954862   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:06:36.955159   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:06:46.955430   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:06:56.955727   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:07:06.956014   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:07:16.956292   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:07:26.956629   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:07:36.956969   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:07:46.957286   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:07:56.957538   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:08:06.957857   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:08:16.958175   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:08:26.958505   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:08:36.958873   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:08:46.959213   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:08:56.959580   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:09:06.959961   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:09:16.960343   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:09:26.960726   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:09:36.961073   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:09:46.961421   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:09:56.961745   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:10:06.962085   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:10:16.962410   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:10:26.962738   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:10:36.963074   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:10:46.963395   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:10:56.963728   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:11:06.964046   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:11:16.964359   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:11:26.964668   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:11:36.964977   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:11:46.965345   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:11:56.965719   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:12:06.966044   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:12:16.966377   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:12:26.966706   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:12:36.967024   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:12:46.967339   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:12:56.967662   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:13:06.967989   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:13:16.968293   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:13:26.968617   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:13:36.968954   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:13:46.969303   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:13:56.969642   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:14:06.969978   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:14:16.970424   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:14:26.970767   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:14:36.971103   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:14:46.971450   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:14:56.971846   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:15:06.972218   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:15:16.972595   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:15:26.972944   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:15:36.973273   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:15:46.973601   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:15:56.973927   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:16:06.974232   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:16:16.974531   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:16:26.974860   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:16:36.975173   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:16:46.975457   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:16:56.975767   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:17:06.976097   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:17:16.976425   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:17:26.976751   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:17:36.977074   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:17:46.977340   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:17:56.977588   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:18:06.977923   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:18:16.978257   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:18:26.978587   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:18:36.978917   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:18:46.979271   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:18:56.979615   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:19:06.979946   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:19:16.980322   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:19:26.980680   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:19:36.981024   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:19:46.981348   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:19:56.981689   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:20:06.982029   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:20:16.982355   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:20:26.982690   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:20:36.983021   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:20:46.983355   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:20:56.983701   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:21:06.984002   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:21:16.984310   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:21:26.984650   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:21:36.985025   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:21:46.985387   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:21:56.985735   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:22:06.986088   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:22:16.986429   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:22:26.986770   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:22:36.987112   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:22:46.987457   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:22:56.987808   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:23:06.988150   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:23:16.988442   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:23:26.988807   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:23:36.989144   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:23:46.989451   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:23:56.989902   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:24:06.990252   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:24:16.990571   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:24:26.990865   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:24:36.991190   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:24:46.991520   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:24:56.991888   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:25:06.992215   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:25:16.992558   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:25:26.993088   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:25:36.993425   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:25:46.993746   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:25:56.994050   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:26:06.994376   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:26:16.994699   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:26:26.995038   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:26:36.995332   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:26:46.995611   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:26:56.995985   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:27:06.996469   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:27:16.996830   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:27:26.997224   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:27:36.997588   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:27:46.997939   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:27:56.998252   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:28:06.998566   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:28:16.998903   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:28:26.999221   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:28:36.999548   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:28:46.999909   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:28:57.000319   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:29:07.000710   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:29:17.001116   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:29:27.001483   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:29:37.001886   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:29:47.002224   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:29:57.002625   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:30:07.002960   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:30:17.003293   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:30:27.003625   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:30:37.003929   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:30:47.004259   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:30:57.004590   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:31:07.004930   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:31:17.005257   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:31:27.005599   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:31:37.005923   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:31:47.006310   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:31:57.006702   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:32:07.007108   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:32:17.007465   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:32:27.007848   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
+I0827 05:32:37.008191   111 master_admin_service.cpp:302] Master Admin Metrics: role=leader, state=serving, service_ready=true, master={Mem Storage: 0 B / 0 B | NVMe-oF SSD: 0 B / 0 B | SSD Storage: 0 B / 0 B | Keys: 0 (soft-pinned: 0) | Clients: 0 | Requests (Success/Total per sec): PutStart=0.00/0.00, PutEnd=0.00/0.00, PutRevoke=0.00/0.00, Get=0.00/0.00, Exist=0.00/0.00, Del=0.00/0.00, DelAll=0.00/0.00, Ping=0.00/0.00, CopyStart=0.00/0.00, CopyEnd=0.00/0.00, CopyRevoke=0.00/0.00, MoveStart=0.00/0.00, MoveEnd=0.00/0.00, MoveRevoke=0.00/0.00, EvictDiskReplica=0.00/0.00 | Batch Requests (per sec, Req=Success/PartialSuccess/Total, Item=Success/Total): PutStart:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutEnd:(Req=0.00/0.00/0.00, Item=0.00/0.00), PutRevoke:(Req=0.00/0.00/0.00, Item=0.00/0.00), Get:(Req=0.00/0.00/0.00, Item=0.00/0.00), ExistKey:(Req=0.00/0.00/0.00, Item=0.00/0.00), QueryIp:(Req=0.00/0.00/0.00, Item=0.00/0.00), Clear:(Req=0.00/0.00/0.00, Item=0.00/0.00), CreateMoveTask:(Req=0.00/0.00), CreateCopyTask:(Req=0.00/0.00), QueryTask:(Req=0.00/0.00), FetchTasks:(Req=0.00/0.00), MarkTaskToComplete:(Req=0.00/0.00) | Eviction: Success/Attempts=0/0, AllocFail=0, keys=0, size=0 B | Mem Eviction: Success/Attempts=0/0, keys=0, size=0 B | NoF Eviction: Success/Attempts=0/0, keys=0, size=0 B | Discard: Released/Total=0/0, StagingSize=0 B | Promotion: in_flight=0, admitted=0, completed=0, failed=0, cancelled=0, expired=0, bytes=0 B, rejected(freq/wm/cap)=0/0/0 | Snapshots: Success=0, Fail=0}, ha={HA Metrics Summary: last_seq=0, applied_seq=0, lag=0, pending=0, mutation_queue=0, batch_commits=0, sync_commits=0, skipped=0, checksum_fail=0, etcd_fail=0, watch_disconn=0, state=0}
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/logs/manual_p_node0.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/logs/manual_p_node0.log
new file mode 100644
index 0000000..0f3c481
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/logs/manual_p_node0.log
@@ -0,0 +1,1338 @@
+/usr/local/lib/python3.12/dist-packages/torchao/quantization/quant_api.py:1731: SyntaxWarning: invalid escape sequence '\.'
+  """Configuration class for applying different quantization configs to modules or parameters based on their fully qualified names (FQNs).
+/sgl-workspace/sglang/python/sglang/launch_server.py:56: UserWarning: 'python -m sglang.launch_server' is still supported, but 'sglang serve' is the recommended entrypoint.
+  Example: sglang serve --model-path <model> [options]
+  warnings.warn(
+[2026-08-27 04:28:54] Breakable CUDA graph is incompatible with MLA attention; disabling prefill CUDA graph.
+[2026-08-27 04:28:55] Attention backend not specified. Use triton backend by default.
+[2026-08-27 04:28:55] Pipeline parallelism is incompatible with overlap schedule.
+[2026-08-27 04:28:58] server_args=ServerArgs(model_path='/data/hf_models/Kimi-K3', tokenizer_path='/data/hf_models/Kimi-K3', tokenizer_mode='auto', tokenizer_backend='huggingface', tokenizer_worker_num=1, detokenizer_worker_num=1, skip_tokenizer_init=False, load_format='auto', model_loader_extra_config='{}', trust_remote_code=True, context_length=None, is_embedding=False, enable_multimodal=None, revision=None, model_impl='auto', model_config_parser='auto', json_model_override_args='{}', dtype='auto', quantization=None, quantization_param_path=None, kv_cache_dtype='auto', enable_fp32_lm_head=False, modelopt_quant=None, modelopt_checkpoint_restore_path=None, modelopt_checkpoint_save_path=None, modelopt_export_path=None, quantize_and_serve=False, rl_quant_profile=None, enable_tf32_matmul=False, mem_fraction_static=0.88, max_running_requests=None, max_queued_requests=None, max_total_tokens=None, chunked_prefill_size=8192, enable_dynamic_chunking=False, max_prefill_tokens=16384, prefill_max_requests=None, schedule_policy='fcfs', enable_priority_scheduling=False, disable_priority_preemption=False, default_priority_value=None, abort_on_priority_when_disabled=False, schedule_low_priority_values_first=False, priority_scheduling_preemption_threshold=10, retraction_policy='length', schedule_conservativeness=1.0, page_size=64, swa_full_tokens_ratio=0.8, disable_hybrid_swa_memory=False, radix_eviction_policy='lru', prefill_only_disable_kv_cache=False, disable_radix_cache=True, enable_page_major_kv_layout=False, enable_unified_memory=False, disable_chunked_prefix_cache=False, disable_overlap_schedule=True, num_continuous_decode_steps=1, scheduler_recv_interval=1, enable_mixed_chunk=False, nccl_port=None, dist_timeout=3600, dist_init_addr='174.1.60.1:20000', nnodes=4, node_rank=0, tp_size=4, dcp_size=1, pp_size=8, pp_max_micro_batch_size=None, pp_async_batch_depth=0, dp_size=1, load_balance_method='follow_bootstrap_room', attn_cp_size=1, moe_dp_size=1, dcp_comm_backend='ag_rs', dcp_replicate_q_proj=None, dwdp_size=1, enable_prefill_cp=False, cp_strategy=None, enable_dsa_cache_layer_split=False, enable_dsa_prefill_context_parallel=False, dsa_prefill_cp_mode='round-robin-split', enable_prefill_context_parallel=False, prefill_cp_mode='in-seq-split', enable_cp_decode_attn_tp=False, enable_dp_attention=False, enable_dp_attention_local_control_broadcast=False, enable_dp_lm_head=False, enable_attn_tp_input_scattered=False, disable_attn_tp_gather=False, enable_p2p_check=False, device='cuda', base_gpu_id=0, gpu_id_step=1, random_seed=984607766, watchdog_timeout=300, soft_watchdog_timeout=None, sleep_on_idle=False, use_ray=False, custom_sigquit_handler=None, numa_node=None, gc_threshold=None, host='0.0.0.0', port=30000, fastapi_root_path='', smg_grpc_mode=False, grpc_mode=False, grpc_port=None, sidecar=None, sidecar_args=None, skip_server_warmup=False, warmups=None, enable_http2=False, ssl_keyfile=None, ssl_certfile=None, ssl_ca_certs=None, ssl_keyfile_password=None, enable_ssl_refresh=False, api_key=None, admin_api_key=None, served_model_name='kimi-k3', weight_version='default', chat_template=None, hf_chat_template_name=None, completion_template=None, file_storage_path='sglang_storage', enable_cache_report=False, reasoning_parser=None, default_chat_template_kwargs=None, strip_thinking_cache=False, enable_strict_thinking=False, tool_call_parser=None, tool_server=None, sampling_defaults='model', asr_max_buffer_seconds=60, asr_max_concurrent_sessions=32, preferred_sampling_params=None, allow_auto_truncate=False, stream_interval=1, batch_notify_size=16, stream_response_default_include_usage=False, incremental_streaming_output=False, enable_streaming_session=False, enable_session_radix_cache=False, log_level='info', log_level_http=None, log_requests=False, log_requests_level=2, log_requests_format='text', log_requests_target=None, uvicorn_access_log_exclude_prefixes=[], crash_dump_folder=None, show_time_cost=False, enable_metrics=False, smg_http_sidecar_port=None, enable_mfu_metrics=False, enable_metrics_for_all_schedulers=False, load_snapshot_publish_interval=15, tokenizer_metrics_custom_labels_header='x-custom-labels', tokenizer_metrics_allowed_custom_labels=None, extra_metric_labels=None, bucket_time_to_first_token=None, bucket_inter_token_latency=None, bucket_e2e_request_latency=None, prompt_tokens_buckets=None, generation_tokens_buckets=None, gc_warning_threshold_secs=0.0, decode_log_interval=40, enable_request_time_stats_logging=False, kv_events_config=None, enable_forward_pass_metrics=False, forward_pass_metrics_worker_id='', forward_pass_metrics_ipc_name=None, enable_trace=False, trace_modules='request', otlp_traces_endpoint='localhost:4317', export_metrics_to_file=False, export_metrics_to_file_dir=None, stat_loggers=None, constrained_json_whitespace_pattern=None, constrained_json_disable_any_whitespace=False, attention_backend='triton', decode_attention_backend=None, prefill_attention_backend=None, sampling_backend='flashinfer', grammar_backend='xgrammar', radix_cache_backend=None, mm_attention_backend=None, fp8_gemm_runner_backend='auto', fp4_gemm_runner_backend='auto', bf16_gemm_backend='auto', dsa_prefill_backend=None, dsa_decode_backend=None, dsa_paged_mqa_logits_backend='auto', dsa_topk_backend='sgl-kernel', disable_flashinfer_autotune=False, flashinfer_autotune_skip_ops=None, mamba_backend='triton', cuda_graph_config=CudaGraphConfig(decode=PhaseConfig(backend='disabled', max_bs=16, bs=[1, 2, 4, 8, 12, 16], tc_compiler='eager', full_prefill_max_req=None), prefill=PhaseConfig(backend='disabled', max_bs=2048, bs=[4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], tc_compiler='eager', full_prefill_max_req=None)), cuda_graph_backend_decode=None, cuda_graph_backend_prefill=None, cuda_graph_max_bs_decode=16, cuda_graph_max_bs_prefill=None, cuda_graph_bs_decode=None, cuda_graph_bs_prefill=None, cuda_graph_tc_compiler=None, disable_prefill_cuda_graph=False, disable_decode_cuda_graph=False, disable_cuda_graph=False, disable_cuda_graph_padding=False, enable_profile_cuda_graph=False, enable_cudagraph_gc=False, debug_cuda_graph=False, enable_layerwise_nvtx_marker=False, enable_nccl_nvls=False, enable_symm_mem=False, triton_attention_reduce_in_fp32=False, triton_attention_num_kv_splits=8, triton_attention_split_tile_size=None, flashinfer_mla_disable_ragged=False, enable_fused_qk_norm_rope=False, enable_precise_embedding_interpolation=False, enable_fused_moe_sum_all_reduce=False, enable_deepseek_v4_fp4_indexer=False, disable_custom_all_reduce=False, enable_mscclpp=False, enable_torch_symm_mem=False, enable_scattered_sconv=False, pre_warm_nccl=False, enable_quant_communications=False, enable_flashinfer_allreduce_fusion=False, enforce_disable_flashinfer_allreduce_fusion=False, flashinfer_allreduce_fusion_backend=None, enable_aiter_allreduce_fusion=False, enable_torch_compile=False, enable_torch_compile_debug_mode=False, torch_compile_max_bs=32, torchao_config='', speculative_algorithm=None, speculative_draft_model_path=None, speculative_draft_model_revision=None, speculative_draft_load_format=None, speculative_num_steps=None, speculative_eagle_topk=None, speculative_num_draft_tokens=None, speculative_dflash_block_size=None, speculative_dspark_block_size=None, speculative_dspark_sps_table_path=None, speculative_dspark_confidence_sts_path=None, speculative_dspark_align_verify_tokens_to_graph_tier=False, speculative_accept_threshold_single=1.0, speculative_accept_threshold_acc=1.0, speculative_use_rejection_sampling=False, speculative_token_map=None, speculative_attention_mode='prefill', speculative_draft_attention_backend=None, speculative_draft_window_size=None, speculative_moe_runner_backend='flashinfer_mxfp4', speculative_moe_a2a_backend=None, speculative_draft_model_quantization=None, speculative_skip_dp_mlp_sync=False, enable_multi_layer_eagle=False, speculative_adaptive=False, speculative_adaptive_config=None, decoupled_spec_bind_endpoint=None, decoupled_spec_connect_endpoints=None, decoupled_spec_rank=None, decoupled_spec_role='null', spec_trace_dir=None, speculative_ngram_min_bfs_breadth=1, speculative_ngram_max_bfs_breadth=10, speculative_ngram_match_type='BFS', speculative_ngram_max_trie_depth=18, speculative_ngram_capacity=10000000, speculative_ngram_external_corpus_path=None, speculative_ngram_external_sam_budget=0, speculative_ngram_external_corpus_max_tokens=10000000, ep_size=4, moe_a2a_backend='none', moe_runner_backend='flashinfer_mxfp4', flashinfer_mxfp4_moe_precision='default', deepep_mode='auto', fuseep_mode=2, deepep_dispatcher_output_dtype='auto', ep_num_redundant_experts=0, ep_dispatch_algorithm=None, init_expert_location='trivial', enable_eplb=False, eplb_algorithm='auto', eplb_rebalance_num_iterations=1000, eplb_rebalance_layers_per_chunk=None, eplb_min_rebalancing_utilization_threshold=1.0, expert_distribution_recorder_mode=None, expert_distribution_recorder_buffer_size=1000, enable_expert_distribution_metrics=False, deepep_config=None, moe_dense_tp_size=None, elastic_ep_backend=None, enable_elastic_expert_backup=False, mooncake_ib_device=None, enable_waterfill=False, ep_join_mode=None, ep_join_rank_offset=0, elastic_ep_initial_size=None, max_ep_size=None, elastic_ep_scale_timeout=600, elastic_ep_rejoin=False, disable_flashinfer_cutlass_moe_fp4_allgather=False, disable_shared_experts_fusion=False, enforce_shared_experts_fusion=False, max_mamba_cache_size=None, mamba_ssm_dtype=None, mamba_max_states_per_path=-1, enable_mamba_cache_stochastic_rounding=False, mamba_cache_philox_rounds=0, mamba_full_memory_ratio=0.36, mamba_radix_cache_strategy='extra_buffer_lazy', uses_mamba_radix_cache=False, mamba_track_interval=256, enable_int8_mamba_checkpoint=False, int8_mamba_ckpt_size=None, linear_attn_backend='triton', linear_attn_decode_backend=None, linear_attn_prefill_backend=None, linear_attn_verify_backend=None, enable_linear_replayssm=False, linear_replayssm_cache_len=16, enable_linear_replayssm_spec=False, enable_hierarchical_cache=False, hicache_ratio=2.0, hicache_size=0, hicache_write_policy='write_through', hicache_io_backend='kernel', hicache_mem_layout='page_first', hicache_storage_backend=None, hicache_storage_prefetch_policy='timeout', hicache_storage_backend_extra_config=None, enable_hisparse=False, hisparse_config=None, enable_broadcast_mm_inputs_process=False, enable_prefix_mm_cache=False, mm_enable_dp_encoder=False, mm_process_config={}, mm_processor_worker_num=0, mm_io_worker_num=0, limit_mm_data_per_request=None, enable_mm_global_cache=False, disable_fast_image_processor=False, mm_feature_transport='cpu', keep_mm_feature_on_device=False, enable_lora=None, enable_lora_overlap_loading=None, max_lora_rank=None, lora_target_modules=None, lora_paths=None, max_loaded_loras=None, max_loras_per_batch=8, lora_eviction_policy='lru', lora_backend='csgmv', max_lora_chunk_size=16, experts_shared_outer_loras=None, lora_use_virtual_experts=False, lora_strict_loading=False, lora_drain_wait_threshold=0.0, enable_two_batch_overlap=False, enable_single_batch_overlap=False, tbo_token_distribution_threshold=0.48, cpu_offload_gb=0, offload_group_size=-1, offload_num_in_group=1, offload_prefetch_step=1, offload_mode='cpu', enable_lmcache=False, lmcache_config_file=None, enable_flexkv=False, flexkv_config_file=None, kt_weight_path=None, kt_method='AMXINT4', kt_cpuinfer=None, kt_threadpool_count=2, kt_num_gpu_experts=None, kt_max_deferred_experts_per_token=None, dllm_algorithm=None, dllm_algorithm_config=None, dllm_fdfo=True, disaggregation_mode='prefill', disaggregation_transfer_backend='mooncake', disaggregation_bootstrap_port=28800, disaggregation_ib_device='mlx5_0,mlx5_1,mlx5_2,mlx5_3', disaggregation_decode_enable_radix_cache=False, disaggregation_decode_enable_offload_kvcache=False, num_reserved_decode_tokens=512, disaggregation_decode_extra_slots=None, disaggregation_decode_polling_interval=1, optimistic_prefill_attempts=0, encoder_only=False, language_only=False, encoder_transfer_backend='zmq_to_tokenizer', encoder_urls=[], encoder_bootstrap_port=8997, encoder_register_urls=[], enable_adaptive_dispatch_to_encoder=False, enable_pdmux=False, pdmux_config_path=None, sm_group_num=8, custom_weight_loader=[], weight_loader_disable_mmap=False, weight_loader_prefetch_checkpoints=False, weight_loader_prefetch_num_threads=4, weight_loader_drop_cache_after_load=False, remote_instance_weight_loader_seed_instance_ip=None, remote_instance_weight_loader_seed_instance_service_port=None, remote_instance_weight_loader_send_weights_group_ports=None, remote_instance_weight_loader_backend='nccl', remote_instance_weight_loader_start_seed_via_transfer_engine=False, engine_info_bootstrap_port=6789, modelexpress_config=None, download_dir=None, model_checksum=None, delete_ckpt_after_loading=False, decrypted_config_file=None, decrypted_draft_config_file=None, checkpoint_engine_wait_weights_before_ready=False, enable_prefill_delayer=False, prefill_delayer_max_delay_passes=30, prefill_delayer_token_usage_low_watermark=None, prefill_delayer_forward_passes_buckets=None, prefill_delayer_wait_seconds_buckets=None, prefill_delayer_queue_min_ratio=None, prefill_delayer_max_delay_ms=None, min_free_slots_delay=None, enable_deterministic_inference=False, rl_on_policy_target=None, kv_canary='none', kv_canary_real_data='none', kv_canary_sweep_interval=0, enable_dynamic_batch_tokenizer=False, dynamic_batch_tokenizer_batch_size=32, dynamic_batch_tokenizer_batch_timeout=0.002, enable_tokenizer_batch_encode=False, disable_tokenizer_batch_decode=False, debug_tensor_dump_output_folder=None, debug_tensor_dump_layers=None, debug_tensor_dump_input_file=None, enable_memory_saver=False, enable_weights_cpu_backup=False, enable_draft_weights_cpu_backup=False, enable_custom_logit_processor=False, enable_return_hidden_states=False, enable_return_routed_experts=False, enable_return_indexer_topk=False, disable_outlines_disk_cache=False, enable_mis=False, weight_cache_mode='off', weight_cache_socket=None, weight_cache_timeout=1800, forward_hooks=None, msprobe_dump_config=None)
+[2026-08-27 04:28:58] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:58] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:58] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:58] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:58] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:58] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:58] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:58] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:01] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:01] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:01] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:03] Multimodal processor concurrency enabled with 2 isolated worker threads (auto).
+[2026-08-27 04:29:03] CommonKVBootstrapServer started successfully on 0.0.0.0:28800
+[2026-08-27 04:29:03] No HuggingFace chat template found
+[2026-08-27 04:29:03] No chat template found, defaulting to 'string' content format
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:15 PP0 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:15 PP0 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:16] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:16] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:16 PP1 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:16 PP1 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+/sgl-workspace/sglang/python/sglang/srt/utils/hf_transformers/tokenizer.py:423: UserWarning: Using a slow tokenizer. This might cause a significant slowdown. Consider using a fast tokenizer instead.
+  warnings.warn(
+[2026-08-27 04:29:16] Applying special tokens cache patch for Kimi tokenizer: <class 'transformers_modules.Kimi_hyphen_K3.caf40b2773f9029e.tokenization_kimi.TikTokenTokenizer'>
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:16 PP1 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:16 PP1 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:16 PP0 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:16 PP0 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:16 PP1 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:16 PP1 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:17 PP0 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:17 PP0 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:17 PP1 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:17 PP1 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:17 PP0 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:17 PP0 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:17 PP0 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:17.587481   237 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:17.587555   237 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:17.587566   237 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.1 port: 12001
+I0827 04:29:17.587602   237 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.1:16830
+I0827 04:29:17.587723   237 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:17.613691   237 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:17.620473   237 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:17.627439   237 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:17.653980   237 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:17.656663   237 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:17.656692   237 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:17.656746   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:17.663833   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:17.665359   237 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:01
+I0827 04:29:17.665387   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:17.690173   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:17.691654   237 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:01
+I0827 04:29:17.691689   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:17.718215   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:17.719650   237 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:01
+I0827 04:29:17.719686   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:17.746297   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:17.747779   237 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:01
+I0827 04:29:17.747891   237 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:17 PP0 TP1 EP1] Init torch distributed begin.
+[2026-08-27 04:29:17 PP1 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:18.219898   241 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:18.219964   241 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:18.219973   241 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.1 port: 12001
+I0827 04:29:18.220007   241 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.1:15425
+I0827 04:29:18.220109   241 transfer_engine_impl.cpp:258] Auto-discovering topology...
+[2026-08-27 04:29:18 PP1 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+I0827 04:29:18.244532   241 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:18.252967   241 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:18.261787   241 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:18.289944   241 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:18.293610   241 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:18.293633   241 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:18.293681   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.301677   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:18.303126   241 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:01
+I0827 04:29:18.303155   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.330181   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:18.331590   241 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:01
+I0827 04:29:18.331614   241 rdma_context.cpp:205] Using SIEVE endpoint store
+[2026-08-27 04:29:18 PP0 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+I0827 04:29:18.358213   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:18.359581   241 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:01
+I0827 04:29:18.359604   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.386204   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:18.387562   241 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:01
+I0827 04:29:18.387681   241 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:18 PP1 TP1 EP1] Init torch distributed begin.
+[2026-08-27 04:29:18 PP1 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:18.588883   242 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:18.588970   242 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:18.588984   242 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.1 port: 12001
+I0827 04:29:18.589022   242 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.1:15650
+I0827 04:29:18.589143   242 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:18.594419   242 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:18.599439   242 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:18.604553   242 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:18.609825   242 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:18.612036   242 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:18.612056   242 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:18.612097   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.618021   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:18.619308   242 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:01
+I0827 04:29:18.619333   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.646080   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:18.647465   242 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:01
+I0827 04:29:18.647495   242 rdma_context.cpp:205] Using SIEVE endpoint store
+[2026-08-27 04:29:18 PP0 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+I0827 04:29:18.674199   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:18.675598   242 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:01
+I0827 04:29:18.675621   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.682866   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:18.684276   242 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:01
+I0827 04:29:18.684391   242 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:18 PP1 TP2 EP2] Init torch distributed begin.
+[2026-08-27 04:29:18 PP1 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:18.695660   238 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:18.695755   238 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:18.695767   238 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.1 port: 12001
+I0827 04:29:18.695806   238 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.1:15123
+I0827 04:29:18.695909   238 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:18.701093   238 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:18.706161   238 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:18.711323   238 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:18.737910   238 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:18.740556   238 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:18.740602   238 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:18.740651   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.747717   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:18.749193   238 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:01
+I0827 04:29:18.749220   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.774145   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:18.775534   238 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:01
+I0827 04:29:18.775558   238 rdma_context.cpp:205] Using SIEVE endpoint store
+[2026-08-27 04:29:18 PP0 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+I0827 04:29:18.802196   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:18.803623   238 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:01
+I0827 04:29:18.803649   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:18.810976   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:18.812341   238 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:01
+I0827 04:29:18.812443   238 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:18 PP0 TP2 EP2] Init torch distributed begin.
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:19.084213   243 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:19.084332   243 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:19.084348   243 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.1 port: 12001
+I0827 04:29:19.084400   243 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.1:15411
+I0827 04:29:19.084602   243 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:19.114387   243 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:19.121773   243 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:19.146405   243 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:19.153759   243 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:19.156515   243 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:19.156548   243 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:19.156603   243 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.163982   243 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:19.165694   243 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:01
+I0827 04:29:19.165726   243 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.173307   243 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:19.174904   243 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:01
+I0827 04:29:19.174932   243 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.206454   243 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:19.208142   243 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:01
+I0827 04:29:19.208168   243 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.215909   243 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:19.217583   243 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:01
+I0827 04:29:19.217706   243 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:19 PP1 TP3 EP3] Init torch distributed begin.
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:19.239588   236 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:19.239699   236 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:19.239715   236 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.1 port: 12001
+I0827 04:29:19.239763   236 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.1:15605
+I0827 04:29:19.239933   236 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:19.240546   240 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:19.240644   240 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:19.240658   240 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.1 port: 12001
+I0827 04:29:19.240702   240 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.1:15487
+I0827 04:29:19.240862   240 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:19.242301   239 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:19.242408   239 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:19.242422   239 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.1 port: 12001
+I0827 04:29:19.242470   239 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.1:15552
+I0827 04:29:19.242636   239 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:19.246201   240 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:19.248451   239 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:19.251204   240 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:19.254226   239 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:19.256254   240 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:19.259852   239 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:19.261372   240 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:19.263411   240 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:19.263432   240 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:19.263473   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.265838   239 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:19.268059   239 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:19.268082   239 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:19.268126   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.269106   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:19.270298   236 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:19.270401   240 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:01
+I0827 04:29:19.270423   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.274480   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:19.275991   239 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:01
+I0827 04:29:19.276019   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.276877   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:19.278193   240 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:01
+I0827 04:29:19.278220   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.297820   236 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:19.302485   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:19.302623   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:19.303958   239 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:01
+I0827 04:29:19.303983   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.304121   240 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:01
+I0827 04:29:19.304147   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.311621   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:19.313341   240 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:01
+I0827 04:29:19.313478   240 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:19 PP1 TP0 EP0] Init torch distributed begin.
+I0827 04:29:19.322126   236 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:19.330281   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:19.331759   239 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:01
+I0827 04:29:19.331784   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.339274   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:19.340766   239 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:01
+I0827 04:29:19.340884   239 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:19 PP0 TP3 EP3] Init torch distributed begin.
+I0827 04:29:19.345958   236 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:19.348632   236 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:19.348659   236 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:19.348712   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.355962   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:19.417273   236 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:01
+I0827 04:29:19.417309   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.424894   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:19.426437   236 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:01
+I0827 04:29:19.426491   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.434231   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:19.435868   236 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:01
+I0827 04:29:19.435896   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:19.443578   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:19.445148   236 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:01
+I0827 04:29:19.445264   236 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:19 PP0 TP0 EP0] Init torch distributed begin.
+[2026-08-27 04:29:26 PP0 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:26 PP1 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:27 PP1 TP2 EP2] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP1 TP0 EP0] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP1 TP3 EP3] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP1 TP1 EP1] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP0 TP3 EP3] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP0 TP0 EP0] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP0 TP1 EP1] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP0 TP2 EP2] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP0 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:27 PP0 TP3 EP3] sglang is using nccl==2.28.9
+[2026-08-27 04:29:27 PP0 TP1 EP1] sglang is using nccl==2.28.9
+[2026-08-27 04:29:27 PP0 TP2 EP2] sglang is using nccl==2.28.9
+[2026-08-27 04:29:27 PP0 TP0 EP0] Init torch distributed ends. elapsed=8.22 s, mem usage=0.46 GB
+[2026-08-27 04:29:27 PP1 TP3 EP3] Init torch distributed ends. elapsed=8.45 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP1 TP1 EP1] Init torch distributed ends. elapsed=9.28 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP1 TP2 EP2] Init torch distributed ends. elapsed=8.98 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP1 TP0 EP0] Init torch distributed ends. elapsed=8.35 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP0 TP2 EP2] Init torch distributed ends. elapsed=8.85 s, mem usage=0.46 GB
+[2026-08-27 04:29:27 PP0 TP3 EP3] Init torch distributed ends. elapsed=8.32 s, mem usage=0.46 GB
+[2026-08-27 04:29:27 PP0 TP1 EP1] Init torch distributed ends. elapsed=9.92 s, mem usage=0.46 GB
+[2026-08-27 04:29:27 PP1 TP0 EP0] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP1 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP0 TP0 EP0] Load weight begin. avail mem=82.12 GB
+[2026-08-27 04:29:27 PP1 TP3 EP3] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP1 TP2 EP2] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP0 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP1 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP1 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP1 TP1 EP1] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP0 TP2 EP2] Load weight begin. avail mem=82.12 GB
+[2026-08-27 04:29:27 PP1 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP0 TP3 EP3] Load weight begin. avail mem=82.12 GB
+[2026-08-27 04:29:27 PP0 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP0 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP1 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:27 PP1 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:27 PP0 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:27 PP1 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:27 PP1 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:27 PP1 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:27 PP0 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:27 PP1 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:27 PP1 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:27 PP1 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:27 PP0 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:27 PP0 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:27 PP0 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:27 PP0 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:27 PP0 TP1 EP1] Load weight begin. avail mem=82.12 GB
+[2026-08-27 04:29:27 PP0 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP0 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP0 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+
Multi-thread loading shards:   0% Completed | 0/96 [00:00<?, ?it/s]
Multi-thread loading shards:   1% Completed | 1/96 [00:00<01:01,  1.55it/s]
Multi-thread loading shards:   2% Completed | 2/96 [00:09<08:49,  5.63s/it]
Multi-thread loading shards:   3% Completed | 3/96 [00:12<06:35,  4.25s/it]
Multi-thread loading shards:   4% Completed | 4/96 [00:14<05:26,  3.55s/it]
Multi-thread loading shards:   5% Completed | 5/96 [00:17<04:43,  3.11s/it]
Multi-thread loading shards:   6% Completed | 6/96 [00:19<04:17,  2.86s/it]
Multi-thread loading shards:   7% Completed | 7/96 [00:21<04:00,  2.71s/it]
Multi-thread loading shards:   8% Completed | 8/96 [00:24<03:49,  2.61s/it]
Multi-thread loading shards:   9% Completed | 9/96 [00:26<03:44,  2.58s/it]
Multi-thread loading shards:  10% Completed | 10/96 [00:29<03:40,  2.56s/it]
Multi-thread loading shards:  11% Completed | 11/96 [00:31<03:36,  2.54s/it]
Multi-thread loading shards:  12% Completed | 12/96 [00:32<02:43,  1.95s/it]
Multi-thread loading shards:  14% Completed | 13/96 [00:32<02:05,  1.52s/it]
Multi-thread loading shards:  15% Completed | 14/96 [00:33<01:33,  1.14s/it]
Multi-thread loading shards:  16% Completed | 15/96 [00:33<01:21,  1.01s/it]
Multi-thread loading shards:  17% Completed | 16/96 [00:34<01:19,  1.00it/s]
Multi-thread loading shards:  18% Completed | 17/96 [00:35<01:02,  1.26it/s]
Multi-thread loading shards:  19% Completed | 18/96 [00:35<00:51,  1.53it/s]
Multi-thread loading shards:  20% Completed | 19/96 [00:35<00:42,  1.80it/s]
Multi-thread loading shards:  21% Completed | 20/96 [00:36<00:37,  2.01it/s]
Multi-thread loading shards:  22% Completed | 21/96 [00:36<00:33,  2.23it/s]
Multi-thread loading shards:  23% Completed | 22/96 [00:36<00:30,  2.43it/s]
Multi-thread loading shards:  24% Completed | 23/96 [00:37<00:27,  2.68it/s]
Multi-thread loading shards:  25% Completed | 24/96 [00:37<00:25,  2.80it/s]
Multi-thread loading shards:  26% Completed | 25/96 [00:37<00:24,  2.89it/s]
Multi-thread loading shards:  27% Completed | 26/96 [00:38<00:23,  2.99it/s]
Multi-thread loading shards:  28% Completed | 27/96 [00:38<00:22,  3.01it/s]
Multi-thread loading shards:  29% Completed | 28/96 [00:38<00:22,  2.99it/s]
Multi-thread loading shards:  30% Completed | 29/96 [00:39<00:22,  2.98it/s]
Multi-thread loading shards:  31% Completed | 30/96 [00:39<00:22,  2.98it/s]
Multi-thread loading shards:  32% Completed | 31/96 [00:39<00:21,  3.01it/s]
Multi-thread loading shards:  33% Completed | 32/96 [00:40<00:21,  3.00it/s]
Multi-thread loading shards:  34% Completed | 33/96 [00:41<00:35,  1.79it/s]
Multi-thread loading shards:  35% Completed | 34/96 [00:41<00:30,  2.02it/s]
Multi-thread loading shards:  36% Completed | 35/96 [00:41<00:27,  2.23it/s]
Multi-thread loading shards:  38% Completed | 36/96 [00:42<00:24,  2.41it/s]
Multi-thread loading shards:  39% Completed | 37/96 [00:42<00:22,  2.57it/s]
Multi-thread loading shards:  40% Completed | 38/96 [00:42<00:21,  2.64it/s]
Multi-thread loading shards:  41% Completed | 39/96 [00:43<00:20,  2.73it/s]
Multi-thread loading shards:  42% Completed | 40/96 [00:43<00:20,  2.80it/s]
Multi-thread loading shards:  43% Completed | 41/96 [00:43<00:19,  2.79it/s]
Multi-thread loading shards:  44% Completed | 42/96 [00:44<00:19,  2.80it/s]
Multi-thread loading shards:  45% Completed | 43/96 [00:44<00:22,  2.39it/s]
Multi-thread loading shards:  46% Completed | 44/96 [00:45<00:20,  2.51it/s]
Multi-thread loading shards:  47% Completed | 45/96 [00:45<00:19,  2.64it/s]
Multi-thread loading shards:  48% Completed | 46/96 [00:45<00:18,  2.67it/s]
Multi-thread loading shards:  49% Completed | 47/96 [00:46<00:18,  2.61it/s]
Multi-thread loading shards:  50% Completed | 48/96 [00:46<00:17,  2.72it/s]
Multi-thread loading shards:  51% Completed | 49/96 [00:47<00:16,  2.84it/s]
Multi-thread loading shards:  52% Completed | 50/96 [00:47<00:16,  2.81it/s]
Multi-thread loading shards:  53% Completed | 51/96 [00:47<00:15,  2.93it/s]
Multi-thread loading shards:  54% Completed | 52/96 [00:48<00:21,  2.05it/s]
Multi-thread loading shards:  55% Completed | 53/96 [00:49<00:33,  1.30it/s]
Multi-thread loading shards:  56% Completed | 54/96 [00:50<00:27,  1.55it/s]
Multi-thread loading shards:  57% Completed | 55/96 [00:50<00:22,  1.82it/s]
Multi-thread loading shards:  58% Completed | 56/96 [00:50<00:19,  2.06it/s]
Multi-thread loading shards:  59% Completed | 57/96 [00:51<00:17,  2.25it/s]
Multi-thread loading shards:  60% Completed | 58/96 [00:51<00:15,  2.40it/s]
Multi-thread loading shards:  61% Completed | 59/96 [00:51<00:14,  2.56it/s]
Multi-thread loading shards:  62% Completed | 60/96 [00:52<00:13,  2.67it/s]
Multi-thread loading shards:  64% Completed | 61/96 [00:52<00:12,  2.77it/s]
Multi-thread loading shards:  65% Completed | 62/96 [00:52<00:11,  2.84it/s]
Multi-thread loading shards:  66% Completed | 63/96 [00:53<00:11,  2.89it/s]
Multi-thread loading shards:  67% Completed | 64/96 [00:53<00:11,  2.87it/s]
Multi-thread loading shards:  68% Completed | 65/96 [00:54<00:10,  2.89it/s]
Multi-thread loading shards:  69% Completed | 66/96 [00:54<00:10,  2.91it/s]
Multi-thread loading shards:  70% Completed | 67/96 [00:54<00:10,  2.90it/s]
Multi-thread loading shards:  71% Completed | 68/96 [00:55<00:09,  2.90it/s]
Multi-thread loading shards:  72% Completed | 69/96 [00:55<00:09,  2.89it/s]
Multi-thread loading shards:  73% Completed | 70/96 [00:56<00:14,  1.85it/s]
Multi-thread loading shards:  74% Completed | 71/96 [00:56<00:11,  2.10it/s]
Multi-thread loading shards:  75% Completed | 72/96 [00:57<00:10,  2.23it/s]
Multi-thread loading shards:  76% Completed | 73/96 [00:57<00:09,  2.39it/s]
Multi-thread loading shards:  77% Completed | 74/96 [00:57<00:08,  2.48it/s]
Multi-thread loading shards:  78% Completed | 75/96 [00:58<00:09,  2.15it/s]
Multi-thread loading shards:  79% Completed | 76/96 [00:59<00:10,  1.89it/s]
Multi-thread loading shards:  80% Completed | 77/96 [00:59<00:09,  2.05it/s]
Multi-thread loading shards:  81% Completed | 78/96 [00:59<00:08,  2.24it/s]
Multi-thread loading shards:  82% Completed | 79/96 [01:00<00:07,  2.39it/s]
Multi-thread loading shards:  83% Completed | 80/96 [01:00<00:06,  2.51it/s]
Multi-thread loading shards:  84% Completed | 81/96 [01:00<00:05,  2.65it/s]
Multi-thread loading shards:  85% Completed | 82/96 [01:01<00:04,  2.84it/s]
Multi-thread loading shards:  86% Completed | 83/96 [01:02<00:06,  1.99it/s]
Multi-thread loading shards:  88% Completed | 84/96 [01:02<00:06,  1.99it/s]
Multi-thread loading shards:  90% Completed | 86/96 [01:02<00:03,  3.27it/s]
Multi-thread loading shards:  92% Completed | 88/96 [01:02<00:01,  4.72it/s]
Multi-thread loading shards:  94% Completed | 90/96 [01:02<00:00,  6.49it/s]
Multi-thread loading shards:  96% Completed | 92/96 [01:03<00:00,  8.37it/s]
Multi-thread loading shards:  98% Completed | 94/96 [01:03<00:00,  8.91it/s]
Multi-thread loading shards: 100% Completed | 96/96 [01:03<00:00,  9.35it/s]
Multi-thread loading shards: 100% Completed | 96/96 [01:03<00:00,  1.51it/s]
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP0 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:33 PP0 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:33 PP0 TP2 EP2] Load weight end. elapsed=65.63 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=39.03 GB, mem usage=43.09 GB.
+[2026-08-27 04:30:33 PP0 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:33 PP0 TP0 EP0] Load weight end. elapsed=65.81 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=39.03 GB, mem usage=43.09 GB.
+[2026-08-27 04:30:35 PP0 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:35 PP0 TP3 EP3] Load weight end. elapsed=67.37 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=39.03 GB, mem usage=43.09 GB.
+[2026-08-27 04:30:35 PP0 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:35 PP0 TP1 EP1] Load weight end. elapsed=67.46 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=39.03 GB, mem usage=43.09 GB.
+[2026-08-27 04:30:35 PP0 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:35 PP0 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:35 PP0 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:35 PP0 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:35 PP0 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:35 PP0 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:35 PP0 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:35 PP0 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:36 PP1 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:36 PP1 TP3 EP3] Load weight end. elapsed=68.32 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=36.12 GB, mem usage=45.92 GB.
+[2026-08-27 04:30:36 PP1 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP1 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP1 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP1 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP1 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP1 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP1 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP1 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:37 PP1 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:37 PP1 TP1 EP1] Load weight end. elapsed=69.55 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=36.12 GB, mem usage=45.92 GB.
+[2026-08-27 04:30:37 PP1 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:37 PP1 TP0 EP0] Load weight end. elapsed=69.66 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=36.12 GB, mem usage=45.92 GB.
+[2026-08-27 04:30:39 PP1 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:39 PP1 TP2 EP2] Load weight end. elapsed=71.33 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=36.12 GB, mem usage=45.92 GB.
+[2026-08-27 04:30:39 PP1 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:39 PP1 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:39 PP1 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:39 PP1 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:39 PP1 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:39 PP1 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:39 PP1 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:39 PP1 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:53 PP0 TP0 EP0] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:53 PP1 TP3 EP3] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:53 PP1 TP2 EP2] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:53 PP1 TP1 EP1] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:53 PP1 TP0 EP0] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:53 PP0 TP2 EP2] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:53 PP0 TP3 EP3] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:53 PP0 TP1 EP1] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:53 PP0 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:53 PP1 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:53 PP1 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:53 PP1 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:53 PP0 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:53 PP1 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:53 PP0 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:53 PP0 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:53 PP0 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 8.16 GB
+[2026-08-27 04:30:53 PP0 TP0 EP0] Memory pool end. avail mem=30.14 GB
+[2026-08-27 04:30:53 PP0 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 8.16 GB
+[2026-08-27 04:30:53 PP0 TP2 EP2] Memory pool end. avail mem=30.14 GB
+[2026-08-27 04:30:53 PP1 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:53 PP1 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:53 PP1 TP0 EP0] Memory pool end. avail mem=23.24 GB
+[2026-08-27 04:30:53 PP1 TP2 EP2] Memory pool end. avail mem=23.24 GB
+[2026-08-27 04:30:53 PP0 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 8.16 GB
+[2026-08-27 04:30:53 PP1 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:53 PP0 TP3 EP3] Memory pool end. avail mem=30.14 GB
+[2026-08-27 04:30:53 PP1 TP1 EP1] Memory pool end. avail mem=23.24 GB
+[2026-08-27 04:30:53 PP1 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:53 PP0 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 8.16 GB
+[2026-08-27 04:30:53 PP1 TP3 EP3] Memory pool end. avail mem=23.24 GB
+[2026-08-27 04:30:53 PP0 TP1 EP1] Memory pool end. avail mem=30.14 GB
+[2026-08-27 04:30:54 PP0 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:30:54 PP0 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:30:54 PP0 TP0 EP0] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp0_pp0_dp0.json
+[2026-08-27 04:30:54 PP1 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:30:54 PP1 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:30:54 PP0 TP2 EP2] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp2_pp0_dp0.json
+[2026-08-27 04:30:54 PP1 TP2 EP2] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp2_pp1_dp0.json
+[2026-08-27 04:30:54 PP1 TP0 EP0] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp0_pp1_dp0.json
+[2026-08-27 04:30:54 PP1 TP1 EP1] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp1_pp1_dp0.json
+[2026-08-27 04:30:54 PP1 TP3 EP3] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp3_pp1_dp0.json
+[2026-08-27 04:30:54 PP0 TP1 EP1] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp1_pp0_dp0.json
+[2026-08-27 04:30:54 PP0 TP3 EP3] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp3_pp0_dp0.json
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.46profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.32profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.51profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  6.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.09profile/s]
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.07profile/s]
+[TensorRT-LLM][INFO] Set logger level to INFO
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.48profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.11profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.29profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.37profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.16profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.82profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.18profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.58profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:01<00:02,  1.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.66profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.03profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.49profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:01<00:01,  2.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.25profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:01,  2.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:02<00:00,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:02<00:00,  3.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.20profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.01profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  9.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  9.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  9.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 12.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 12.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 11.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 11.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.78profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.16profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  3.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.75profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.50profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.92profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.68profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  3.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.97profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.87profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 10.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.51profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.74profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 11.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.30profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.70profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.72profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.63profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.53profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.29profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.58profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  8.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 10.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.71profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.62profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.43profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.18profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.29profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.37profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.53profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.37profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.35profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.00profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.71profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.60profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.78profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.79profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  8.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.86profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  8.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.16profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.78profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  3.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.15profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.94profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.93profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.25profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.23profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.97profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.79profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.96profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.64profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.12profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.75profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.12profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.25profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.91profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.05profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.66profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.90profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  9.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.73profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.55profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.49profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.13profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.74profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.17profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  3.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.72profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.57profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.59profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.64profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.83profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 13.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.49profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 12.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  9.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  7.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.32profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.93profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.29profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.72profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.75profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.46profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.51profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  3.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.90profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.16profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.72profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.63profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.83profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.70profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.59profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.55profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  6.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.93profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.33profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.75profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.30profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  6.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  7.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.79profile/s]
+[2026-08-27 04:31:21 PP0 TP1 EP1] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.53profile/s][2026-08-27 04:31:21 PP0 TP0 EP0] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.97profile/s]
+[2026-08-27 04:31:21 PP0 TP3 EP3] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.43profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.06profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s][2026-08-27 04:31:21 PP0 TP2 EP2] FlashInfer autotune completed.
+[2026-08-27 04:31:21 PP0 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:21 PP0 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:21 PP0 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:21 PP0 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:21 PP0 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:21 PP0 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:21 PP0 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:21 PP0 TP0 EP0] max_total_num_tokens=3801408, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=54, context_len=1048576, available_gpu_mem=27.35 GB
+[2026-08-27 04:31:21 PP0 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.86profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.58profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:01<00:07,  1.18s/profile]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  2.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.27profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.55profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  3.96profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.86profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  6.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  7.01profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.23profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.05profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  6.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  7.06profile/s]
+[2026-08-27 04:31:26 PP1 TP1 EP1] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.40profile/s]
+[2026-08-27 04:31:26 PP1 TP2 EP2] FlashInfer autotune completed.
+[2026-08-27 04:31:26 PP1 TP0 EP0] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.81profile/s]
+[2026-08-27 04:31:26 PP1 TP3 EP3] FlashInfer autotune completed.
+[2026-08-27 04:31:26 PP1 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:26 PP1 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:26 PP1 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:26 PP1 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:26 PP1 TP0 EP0] max_total_num_tokens=3801408, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=54, context_len=1048576, available_gpu_mem=20.45 GB
+[2026-08-27 04:31:26 PP1 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:26 PP1 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:26 PP1 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:26 PP1 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+W0827 04:31:30.805795  2607 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fe48940, len: 6912: Operation not permitted [1]
+W0827 04:31:30.805797  2608 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4f684400, len: 6912: Operation not permitted [1]
+W0827 04:31:30.805945  2609 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4f962700, len: 6912: Operation not permitted [1]
+W0827 04:31:30.805979  2610 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4f9e8a00, len: 6912: Operation not permitted [1]
+W0827 04:31:30.806130  2612 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fe5f640, len: 55296: Operation not permitted [1]
+W0827 04:31:30.806133  2611 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4f83f000, len: 55296: Operation not permitted [1]
+W0827 04:31:30.806264  2613 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fe328c0, len: 6912: Operation not permitted [1]
+W0827 04:31:30.806334  2614 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fa75cc0, len: 13824: Operation not permitted [1]
+W0827 04:31:30.806449  2615 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fa87f80, len: 6912: Operation not permitted [1]
+W0827 04:31:30.806493  2616 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fa29a80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.203712  2635 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50c65e80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.203859  2636 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50c6bb40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.203975  2637 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50d1ffc0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.204025  2638 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50a0c300, len: 6912: Operation not permitted [1]
+W0827 04:31:31.204075  2640 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50de1940, len: 55296: Operation not permitted [1]
+W0827 04:31:31.204082  2639 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50dc93c0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.204438  2641 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50c87940, len: 6912: Operation not permitted [1]
+W0827 04:31:31.204561  2642 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50cffb00, len: 13824: Operation not permitted [1]
+W0827 04:31:31.204641  2644 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50dc7380, len: 6912: Operation not permitted [1]
+W0827 04:31:31.204722  2643 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50c243c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.323861  2661 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50b13c40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.323863  2662 memory_location.cpp:73] Failed to get NUMA node, addr: 0x511191c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.323912  2663 memory_location.cpp:73] Failed to get NUMA node, addr: 0x511b37c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.324069  2664 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50b2da80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.324110  2666 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5115c980, len: 55296: Operation not permitted [1]
+W0827 04:31:31.324155  2665 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5116ac40, len: 55296: Operation not permitted [1]
+W0827 04:31:31.324479  2667 memory_location.cpp:73] Failed to get NUMA node, addr: 0x511a7e80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.324587  2668 memory_location.cpp:73] Failed to get NUMA node, addr: 0x511b8cc0, len: 13824: Operation not permitted [1]
+W0827 04:31:31.324621  2669 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50fc6800, len: 6912: Operation not permitted [1]
+W0827 04:31:31.324682  2670 memory_location.cpp:73] Failed to get NUMA node, addr: 0x511bc780, len: 6912: Operation not permitted [1]
+W0827 04:31:31.543082  2689 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4902e440, len: 6912: Operation not permitted [1]
+W0827 04:31:31.543203  2690 memory_location.cpp:73] Failed to get NUMA node, addr: 0x49484200, len: 6912: Operation not permitted [1]
+W0827 04:31:31.543314  2691 memory_location.cpp:73] Failed to get NUMA node, addr: 0x49119600, len: 6912: Operation not permitted [1]
+W0827 04:31:31.543375  2692 memory_location.cpp:73] Failed to get NUMA node, addr: 0x493dc780, len: 6912: Operation not permitted [1]
+W0827 04:31:31.543467  2693 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51ac5fc0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.543601  2694 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51a9e780, len: 55296: Operation not permitted [1]
+W0827 04:31:31.543751  2695 memory_location.cpp:73] Failed to get NUMA node, addr: 0x49149a40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.543912  2696 memory_location.cpp:73] Failed to get NUMA node, addr: 0x518c0840, len: 13824: Operation not permitted [1]
+W0827 04:31:31.543947  2697 memory_location.cpp:73] Failed to get NUMA node, addr: 0x49303540, len: 6912: Operation not permitted [1]
+W0827 04:31:31.543990  2698 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51951a80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681525  2717 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50376e40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681525  2718 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5058ee40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681525  2716 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5008b980, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681548  2715 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fe0e840, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681727  2719 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5040eb80, len: 55296: Operation not permitted [1]
+W0827 04:31:31.681779  2724 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50209ac0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681811  2720 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5041c3c0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.681828  2722 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5048cc40, len: 13824: Operation not permitted [1]
+W0827 04:31:31.681844  2723 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fde7a00, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681851  2721 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50386580, len: 6912: Operation not permitted [1]
+W0827 04:31:31.730942  2745 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fce4240, len: 6912: Operation not permitted [1]
+W0827 04:31:31.730918  2744 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fed7d40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.730950  2743 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fdcaf80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.731007  2748 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5074c400, len: 55296: Operation not permitted [1]
+W0827 04:31:31.731073  2747 memory_location.cpp:73] Failed to get NUMA node, addr: 0x507bd9c0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.731086  2746 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4f931480, len: 6912: Operation not permitted [1]
+W0827 04:31:31.731294  2749 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd958c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.731361  2751 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd31d80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.731361  2750 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4ffabe00, len: 13824: Operation not permitted [1]
+W0827 04:31:31.731566  2752 memory_location.cpp:73] Failed to get NUMA node, addr: 0x508527c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.808167  2778 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd3ce40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.808190  2777 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd20740, len: 6912: Operation not permitted [1]
+W0827 04:31:31.808192  2776 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4ff86c80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.808637  2782 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fe88780, len: 6912: Operation not permitted [1]
+W0827 04:31:31.808724  2780 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fe48c80, len: 55296: Operation not permitted [1]
+W0827 04:31:31.808760  2781 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fe564c0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.809056  2785 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4ffa7e40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.809049  2779 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4feaa740, len: 6912: Operation not permitted [1]
+W0827 04:31:31.809075  2784 memory_location.cpp:73] Failed to get NUMA node, addr: 0x502fabc0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.809239  2783 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4ffa1f80, len: 13824: Operation not permitted [1]
+W0827 04:31:31.851475  2807 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4ff6fcc0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.851596  2808 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fc0a280, len: 6912: Operation not permitted [1]
+W0827 04:31:31.851615  2809 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd7d700, len: 6912: Operation not permitted [1]
+W0827 04:31:31.851732  2810 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd761c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.851951  2811 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fde6cc0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.851970  2812 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4feeb000, len: 55296: Operation not permitted [1]
+W0827 04:31:31.851962  2813 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fbe2680, len: 6912: Operation not permitted [1]
+W0827 04:31:31.851987  2816 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fbf07c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.852048  2815 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4ffa9940, len: 6912: Operation not permitted [1]
+W0827 04:31:31.852046  2814 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd235c0, len: 13824: Operation not permitted [1]
+[2026-08-27 04:31:32] INFO:     Started server process [1]
+[2026-08-27 04:31:32] INFO:     Waiting for application startup.
+[2026-08-27 04:31:32] Calling super().encode with {'add_special_tokens': False}
+[2026-08-27 04:31:32] Calling super().encode with {'add_special_tokens': False}
+[2026-08-27 04:31:32] INFO:     Application startup complete.
+[2026-08-27 04:31:32] INFO:     Uvicorn running on http://0.0.0.0:30000 (Press CTRL+C to quit)
+[2026-08-27 04:31:33] INFO:     127.0.0.1:60348 - "GET /model_info HTTP/1.1" 200 OK
+[2026-08-27 04:31:33] Start of pd disaggregation warmup ...
+[2026-08-27 04:31:37] INFO:     174.1.60.1:33088 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:38] INFO:     174.1.60.1:33092 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:39] INFO:     174.1.60.1:33106 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:40] INFO:     174.1.60.1:33110 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:41] INFO:     174.1.60.1:42028 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:42] INFO:     174.1.60.1:42030 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:43] INFO:     174.1.60.1:42046 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:44] INFO:     174.1.60.1:42050 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:45] INFO:     174.1.60.1:42060 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:46] INFO:     174.1.60.1:42072 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:47] INFO:     174.1.60.1:42086 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:48] INFO:     174.1.60.1:42092 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:49] INFO:     174.1.60.1:42098 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:50] INFO:     174.1.60.1:42100 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:51] INFO:     174.1.60.1:46852 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:52] INFO:     174.1.60.1:46856 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:53] INFO:     174.1.60.1:46872 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:54] INFO:     174.1.60.1:46876 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:55] INFO:     174.1.60.1:46886 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:56] INFO:     174.1.60.1:46890 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:57] INFO:     174.1.60.1:46898 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:58] INFO:     174.1.60.1:46912 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:31:59] INFO:     174.1.60.1:46928 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:00] INFO:     174.1.60.1:46936 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:01] INFO:     174.1.60.1:57672 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:02] INFO:     174.1.60.1:57676 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:03] INFO:     174.1.60.1:57686 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:04] INFO:     174.1.60.1:57688 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:05] INFO:     174.1.60.1:57696 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:06] INFO:     174.1.60.1:57708 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:12] INFO:     174.1.60.1:39714 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:13] INFO:     174.1.60.1:39728 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:14] INFO:     174.1.60.1:39736 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:15] INFO:     174.1.60.1:39746 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:16] INFO:     174.1.60.1:39760 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:17] INFO:     174.1.60.1:39772 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:18] INFO:     174.1.60.1:39782 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:19] INFO:     174.1.60.1:39786 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:20] INFO:     174.1.60.1:39800 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:21] INFO:     174.1.60.1:41036 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:22] INFO:     174.1.60.1:41052 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:23] INFO:     174.1.60.1:41058 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:24] INFO:     174.1.60.1:41068 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:25] INFO:     174.1.60.1:41074 - "GET /health HTTP/1.1" 503 Service Unavailable
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank1]:[W827 04:32:26.019157362 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank0]:[W827 04:32:26.019963648 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank3]:[W827 04:32:26.020028119 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank2]:[W827 04:32:26.020057305 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank6]:[W827 04:32:26.045976535 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank7]:[W827 04:32:26.046176399 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank5]:[W827 04:32:26.048033943 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank4]:[W827 04:32:26.048331224 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[2026-08-27 04:32:26] INFO:     174.1.60.1:41078 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:27] INFO:     174.1.60.1:41090 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:28] INFO:     174.1.60.1:41102 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:29] INFO:     174.1.60.1:41104 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:30] INFO:     174.1.60.1:41120 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:31] INFO:     174.1.60.1:57002 - "GET /health HTTP/1.1" 503 Service Unavailable
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank6]:[W827 04:32:31.709756552 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank7]:[W827 04:32:31.709839299 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank5]:[W827 04:32:31.709991082 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank4]:[W827 04:32:31.710123265 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[2026-08-27 04:32:32] INFO:     174.1.60.1:57016 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:33] INFO:     174.1.60.1:57030 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:34] INFO:     174.1.60.1:57046 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:35] INFO:     174.1.60.1:57056 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:36] INFO:     174.1.60.1:57066 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:37] INFO:     174.1.60.1:57070 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:38] INFO:     174.1.60.1:57072 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:39] INFO:     174.1.60.1:57082 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:40] INFO:     174.1.60.1:57098 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:41] INFO:     174.1.60.1:56606 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:50] INFO:     174.1.60.1:56620 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:51] INFO:     174.1.60.1:45658 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:52] INFO:     174.1.60.1:45672 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:53] INFO:     174.1.60.1:45674 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:54] INFO:     174.1.60.1:45690 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:55] INFO:     174.1.60.1:45704 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:56] INFO:     174.1.60.1:45708 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:57] INFO:     174.1.60.1:45714 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:32:58] INFO:     174.1.60.1:45722 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:00] INFO:     174.1.60.1:45736 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:01] INFO:     174.1.60.1:45746 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:02] INFO:     174.1.60.1:33310 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:03] INFO:     174.1.60.1:33320 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:04] INFO:     174.1.60.1:33324 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:05] INFO:     174.1.60.1:33328 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:06] INFO:     174.1.60.1:33336 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:07] INFO:     174.1.60.1:33350 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:08] INFO:     174.1.60.1:33354 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:09] INFO:     174.1.60.1:33356 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:10] INFO:     174.1.60.1:33368 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:11] INFO:     174.1.60.1:39318 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:12] INFO:     174.1.60.1:39328 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:13] INFO:     174.1.60.1:39336 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:14] INFO:     174.1.60.1:39344 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:15] INFO:     174.1.60.1:39346 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:16] INFO:     174.1.60.1:39352 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:17] INFO:     174.1.60.1:39364 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:18] INFO:     174.1.60.1:39378 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:19] INFO:     174.1.60.1:39390 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:20] INFO:     174.1.60.1:39402 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:25] INFO:     174.1.60.1:37792 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:26] INFO:     174.1.60.1:37804 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:27] INFO:     174.1.60.1:37812 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:28] INFO:     174.1.60.1:37826 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:29] INFO:     174.1.60.1:37830 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:30] INFO:     174.1.60.1:37844 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:31] INFO:     174.1.60.1:44208 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:32] INFO:     174.1.60.1:44220 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:33] INFO:     174.1.60.1:44226 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:34] INFO:     174.1.60.1:44228 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:35] INFO:     174.1.60.1:44238 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:37] INFO:     174.1.60.1:44248 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:38] INFO:     174.1.60.1:44258 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:39] INFO:     174.1.60.1:44272 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:40] INFO:     174.1.60.1:44276 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:41] INFO:     174.1.60.1:44194 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:42] INFO:     174.1.60.1:44198 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:43] INFO:     174.1.60.1:44212 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:44] INFO:     174.1.60.1:44216 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:45] INFO:     174.1.60.1:44222 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:46] INFO:     174.1.60.1:44236 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:47] INFO:     174.1.60.1:44246 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:48] INFO:     174.1.60.1:44252 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:49] INFO:     174.1.60.1:44264 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:50] INFO:     174.1.60.1:44278 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:51] INFO:     174.1.60.1:35728 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:52] INFO:     174.1.60.1:35736 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:53] INFO:     174.1.60.1:35752 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:54] INFO:     174.1.60.1:35754 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:33:55] INFO:     174.1.60.1:35758 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:04] INFO:     174.1.60.1:44514 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:05] INFO:     174.1.60.1:44518 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:06] INFO:     174.1.60.1:44520 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:07] INFO:     174.1.60.1:44526 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:08] INFO:     174.1.60.1:44530 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:09] INFO:     174.1.60.1:44546 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:10] INFO:     174.1.60.1:44550 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:11] INFO:     174.1.60.1:45098 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:12] INFO:     127.0.0.1:59722 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:12] INFO:     174.1.60.1:45106 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:13] INFO:     174.1.60.1:45122 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:14] INFO:     174.1.60.1:45124 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:15] INFO:     174.1.60.1:45136 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:16] INFO:     174.1.60.1:45146 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:17] INFO:     174.1.60.1:45152 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:19] INFO:     174.1.60.1:45158 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:20] INFO:     174.1.60.1:45170 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:21] INFO:     174.1.60.1:45182 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:22] INFO:     174.1.60.1:39946 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:23] INFO:     174.1.60.1:39958 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:24] INFO:     174.1.60.1:39962 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:25] INFO:     174.1.60.1:39978 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:26] INFO:     174.1.60.1:39990 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:27] INFO:     174.1.60.1:39994 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:28] INFO:     174.1.60.1:39996 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:29] INFO:     174.1.60.1:40002 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:30] INFO:     174.1.60.1:40012 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:31] INFO:     174.1.60.1:46978 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:32] INFO:     174.1.60.1:46988 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:33] INFO:     174.1.60.1:46996 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:34] INFO:     174.1.60.1:47004 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:39] INFO:     174.1.60.1:47020 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:40] INFO:     174.1.60.1:47022 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:41] INFO:     174.1.60.1:59158 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:42] INFO:     174.1.60.1:59170 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:43] INFO:     174.1.60.1:59172 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:44] INFO:     174.1.60.1:59180 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:45] INFO:     174.1.60.1:59184 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:46] INFO:     174.1.60.1:59186 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:47] INFO:     174.1.60.1:59200 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:48] INFO:     174.1.60.1:59202 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:49] INFO:     174.1.60.1:59214 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:50] INFO:     174.1.60.1:59220 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:52] INFO:     174.1.60.1:40566 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:53] INFO:     174.1.60.1:40576 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:54] INFO:     174.1.60.1:40588 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:55] INFO:     174.1.60.1:40596 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:56] INFO:     174.1.60.1:40598 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:57] INFO:     174.1.60.1:40604 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:58] INFO:     174.1.60.1:40608 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:34:59] INFO:     174.1.60.1:40610 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:00] INFO:     174.1.60.1:40622 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:01] INFO:     174.1.60.1:41102 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:02] INFO:     174.1.60.1:41108 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:03] INFO:     174.1.60.1:41122 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:04] INFO:     174.1.60.1:41124 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:05] INFO:     174.1.60.1:41134 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:06] INFO:     174.1.60.1:41140 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:07] INFO:     174.1.60.1:41142 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:08] INFO:     174.1.60.1:41158 - "GET /health HTTP/1.1" 503 Service Unavailable
+[2026-08-27 04:35:09] INFO:     174.1.60.1:41160 - "GET /health HTTP/1.1" 503 Service Unavailable
+[rank3]:[W827 04:35:10.359210278 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank2]:[W827 04:35:10.359272236 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank0]:[W827 04:35:10.360047094 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank1]:[W827 04:35:10.370610409 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[2026-08-27 04:35:10 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.28
+[2026-08-27 04:35:10 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.29
+[2026-08-27 04:35:10] INFO:     127.0.0.1:60364 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 04:35:10] Disaggregation warmup requests completed for all 1 DP ranks
+[2026-08-27 04:35:10] End of disaggregation warmup
+[2026-08-27 04:35:10] The server is fired up and ready to roll!
+[2026-08-27 04:35:29 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.39
+[2026-08-27 04:35:29 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.39
+[2026-08-27 04:35:31 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25.83
+[2026-08-27 04:35:31 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25.84
+[2026-08-27 04:35:32] INFO:     174.1.60.1:49902 - "GET /health HTTP/1.1" 200 OK
+[2026-08-27 04:43:22 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.14
+[2026-08-27 04:43:22 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.14
+[2026-08-27 04:43:23] INFO:     174.1.60.1:33240 - "GET /health HTTP/1.1" 200 OK
+/sgl-workspace/sglang/python/sglang/srt/entrypoints/http_server.py:450: FastAPIDeprecationWarning: ORJSONResponse is deprecated, FastAPI now serializes data directly to JSON bytes via Pydantic when a return type or response model is set, which is faster and doesn't need a custom response class. Read more in the FastAPI docs: https://fastapi.tiangolo.com/advanced/custom-response/#orjson-or-response-model and https://fastapi.tiangolo.com/tutorial/response-model/
+  return await original_handler(ORJSONRequest(request.scope, request.receive))
+[2026-08-27 04:45:43] INFO:     174.1.60.5:35536 - "GET /v1/models HTTP/1.1" 200 OK
+[2026-08-27 04:45:43] INFO:     174.1.60.5:35542 - "GET /v1/models HTTP/1.1" 200 OK
+[2026-08-27 04:48:11] INFO:     174.1.60.5:47258 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 04:48:13 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.22
+[2026-08-27 04:48:13 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.22
+[2026-08-27 04:48:13 PP0 TP1 EP1] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:13 PP0 TP2 EP2] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:13 PP0 TP0 EP0] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:13 PP0 TP3 EP3] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP1 TP1 EP1] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP1 TP2 EP2] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP1 TP0 EP0] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP1 TP3 EP3] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:16] INFO:     174.1.60.5:50348 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 04:48:16] INFO:     174.1.60.5:50364 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 04:54:46] INFO:     174.1.60.5:51618 - "GET /v1/models HTTP/1.1" 200 OK
+[2026-08-27 04:54:46] INFO:     174.1.60.5:51626 - "GET /v1/models HTTP/1.1" 200 OK
+[2026-08-27 04:57:13] INFO:     174.1.60.5:48050 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 04:57:14 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.12
+[2026-08-27 04:57:14 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.12
+[2026-08-27 04:57:14] INFO:     174.1.60.5:48060 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 04:57:14] INFO:     174.1.60.5:48074 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 04:57:40] INFO:     174.1.60.5:51792 - "GET /v1/models HTTP/1.1" 200 OK
+[2026-08-27 04:57:40] INFO:     174.1.60.5:51798 - "GET /v1/models HTTP/1.1" 200 OK
+[2026-08-27 05:00:07] INFO:     174.1.60.5:48432 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:04:21 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1088, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2.55
+[2026-08-27 05:04:21 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1088, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2.55
+[2026-08-27 05:04:22] INFO:     174.1.60.5:36022 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:04:22] INFO:     174.1.60.5:36028 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:04:41 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.19
+[2026-08-27 05:04:41 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.19
+[2026-08-27 05:04:41] INFO:     174.1.60.5:33310 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:04:49] INFO:     174.1.60.5:39866 - "GET /v1/models HTTP/1.1" 200 OK
+[2026-08-27 05:04:49] INFO:     174.1.60.5:39874 - "GET /v1/models HTTP/1.1" 200 OK
+[2026-08-27 05:05:38] INFO:     174.1.60.5:57632 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:05:39 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1.10
+[2026-08-27 05:05:39 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1.10
+[2026-08-27 05:07:16] INFO:     174.1.60.5:42880 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:07:30 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 73.88
+[2026-08-27 05:07:32 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 72.71
+[2026-08-27 05:07:32 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3187.73
+[2026-08-27 05:07:33 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9987.21
+[2026-08-27 05:07:33 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 5011.85
+[2026-08-27 05:07:33 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7713.95
+[2026-08-27 05:07:33] INFO:     174.1.60.5:49324 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:07:38 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8673, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1473.19
+[2026-08-27 05:07:39 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8673, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1235.01
+[2026-08-27 05:07:40 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 481, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4257.97
+[2026-08-27 05:07:40 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 481, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9114.58
+[2026-08-27 05:07:40 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4643.44
+[2026-08-27 05:07:40 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7350.46
+[2026-08-27 05:07:40] INFO:     174.1.60.5:49326 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:07:46 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8656, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1470.23
+[2026-08-27 05:07:47 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8656, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1236.74
+[2026-08-27 05:07:48 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 464, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4291.82
+[2026-08-27 05:07:48 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 464, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9091.63
+[2026-08-27 05:07:48 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4126.73
+[2026-08-27 05:07:48 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6652.98
+[2026-08-27 05:07:48] INFO:     174.1.60.5:38928 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:07:53 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8663, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1477.97
+[2026-08-27 05:07:54 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8663, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1240.34
+[2026-08-27 05:07:55 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 471, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4270.28
+[2026-08-27 05:07:55 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 471, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9111.26
+[2026-08-27 05:07:55 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4676.01
+[2026-08-27 05:07:55 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7851.58
+[2026-08-27 05:07:56] INFO:     174.1.60.5:34214 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:08:01 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8659, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1474.57
+[2026-08-27 05:08:02 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8659, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1237.36
+[2026-08-27 05:08:03 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 467, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4273.37
+[2026-08-27 05:08:03 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 467, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9096.53
+[2026-08-27 05:08:03 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4082.13
+[2026-08-27 05:08:03 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6360.56
+[2026-08-27 05:08:03] INFO:     174.1.60.5:38204 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:08:08 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8692, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1476.50
+[2026-08-27 05:08:10 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8692, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1239.29
+[2026-08-27 05:08:10 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 500, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4267.60
+[2026-08-27 05:08:10 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 500, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8989.39
+[2026-08-27 05:08:11 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4419.86
+[2026-08-27 05:08:11 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8656.56
+[2026-08-27 05:08:11] INFO:     174.1.60.5:47402 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:08:16 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8557, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1477.68
+[2026-08-27 05:08:17 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8557, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1239.64
+[2026-08-27 05:08:18 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 365, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4273.32
+[2026-08-27 05:08:18 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 365, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9138.56
+[2026-08-27 05:08:18 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4206.66
+[2026-08-27 05:08:18 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8364.41
+[2026-08-27 05:08:18] INFO:     174.1.60.5:47404 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:08:24 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8683, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1475.28
+[2026-08-27 05:08:25 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8683, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1237.96
+[2026-08-27 05:08:26 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 491, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4263.38
+[2026-08-27 05:08:26 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 491, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9074.22
+[2026-08-27 05:08:26 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4537.22
+[2026-08-27 05:08:26 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7174.96
+[2026-08-27 05:08:26] INFO:     174.1.60.5:48174 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:08:26] INFO:     174.1.60.5:48178 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:09:17] INFO:     174.1.60.5:41772 - "GET /v1/models HTTP/1.1" 200 OK
+[2026-08-27 05:09:17] INFO:     174.1.60.5:41776 - "GET /v1/models HTTP/1.1" 200 OK
+[2026-08-27 05:11:44] INFO:     174.1.60.5:45272 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:11:45 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.32
+[2026-08-27 05:11:45 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.32
+E0827 05:11:45.751729  1489 worker_pool.cpp:1217] CQ completion timeout diagnostic: context=mlx5_3, outstanding=2, outstanding_age_ms=0, poll_gap_ms=199552, last_poll_interval_ms=0, max_poll_interval_ms=539649, submitted=34398, processed=34396
+[2026-08-27 05:11:46] INFO:     174.1.60.5:45274 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:11:46] INFO:     174.1.60.5:45276 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:12:14] INFO:     174.1.60.5:34478 - "GET /v1/models HTTP/1.1" 200 OK
+[2026-08-27 05:12:14] INFO:     174.1.60.5:34486 - "GET /v1/models HTTP/1.1" 200 OK
+[2026-08-27 05:14:42 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1088, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6.16
+[2026-08-27 05:14:42 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1088, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6.16
+[2026-08-27 05:14:42] INFO:     174.1.60.5:46142 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:14:43] INFO:     174.1.60.5:46156 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:14:43] INFO:     174.1.60.5:46162 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:15:12] INFO:     174.1.60.5:60686 - "GET /v1/models HTTP/1.1" 200 OK
+[2026-08-27 05:15:12] INFO:     174.1.60.5:60696 - "GET /v1/models HTTP/1.1" 200 OK
+[2026-08-27 05:17:40] INFO:     174.1.60.5:49298 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:17:45 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 44.69
+[2026-08-27 05:17:46 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 44.43
+[2026-08-27 05:17:47 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4291.89
+[2026-08-27 05:17:47 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9142.33
+[2026-08-27 05:17:47 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4792.42
+[2026-08-27 05:17:47 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8482.06
+[2026-08-27 05:17:47] INFO:     174.1.60.5:45358 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:17:53 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8673, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1475.80
+[2026-08-27 05:17:54 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8673, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1239.62
+[2026-08-27 05:17:55 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 481, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4273.03
+[2026-08-27 05:17:55 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 481, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9047.83
+[2026-08-27 05:17:55 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4673.57
+[2026-08-27 05:17:55 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8029.35
+[2026-08-27 05:17:55] INFO:     174.1.60.5:37922 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:18:00 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8656, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1479.36
+[2026-08-27 05:18:01 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8656, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1239.75
+[2026-08-27 05:18:02 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 464, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4268.49
+[2026-08-27 05:18:02 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 464, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9151.27
+[2026-08-27 05:18:02 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4773.15
+[2026-08-27 05:18:02 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8723.97
+[2026-08-27 05:18:02] INFO:     174.1.60.5:59842 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:18:08 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8663, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1480.27
+[2026-08-27 05:18:09 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8663, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1240.59
+[2026-08-27 05:18:10 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 471, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4269.06
+[2026-08-27 05:18:10 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 471, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9108.64
+[2026-08-27 05:18:10 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4695.76
+[2026-08-27 05:18:10 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8458.08
+[2026-08-27 05:18:10] INFO:     174.1.60.5:59848 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:18:15 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8659, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1476.05
+[2026-08-27 05:18:17 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8659, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1238.88
+[2026-08-27 05:18:17 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 467, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4269.01
+[2026-08-27 05:18:17 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 467, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9060.08
+[2026-08-27 05:18:17 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4685.85
+[2026-08-27 05:18:17 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8293.92
+[2026-08-27 05:18:18] INFO:     174.1.60.5:46946 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:18:23 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8692, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1478.44
+[2026-08-27 05:18:24 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8692, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1238.63
+[2026-08-27 05:18:25 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 500, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4261.28
+[2026-08-27 05:18:25 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 500, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9140.66
+[2026-08-27 05:18:25 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4667.18
+[2026-08-27 05:18:25 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8159.39
+[2026-08-27 05:18:25] INFO:     174.1.60.5:42086 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:18:31 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8557, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1480.63
+[2026-08-27 05:18:32 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8557, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1240.79
+[2026-08-27 05:18:32 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 365, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4272.82
+[2026-08-27 05:18:33 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 365, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9138.60
+[2026-08-27 05:18:33 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4248.66
+[2026-08-27 05:18:33 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8746.98
+[2026-08-27 05:18:33] INFO:     174.1.60.5:36136 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:18:38 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8683, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1480.90
+[2026-08-27 05:18:39 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8683, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1240.92
+[2026-08-27 05:18:40 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 491, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4261.25
+[2026-08-27 05:18:40 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 491, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9102.59
+[2026-08-27 05:18:40 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4532.61
+[2026-08-27 05:18:40 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8049.42
+[2026-08-27 05:18:40] INFO:     174.1.60.5:36144 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:18:40] INFO:     174.1.60.5:36150 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:19:09] INFO:     174.1.60.5:57946 - "GET /v1/models HTTP/1.1" 200 OK
+[2026-08-27 05:19:09] INFO:     174.1.60.5:57958 - "GET /v1/models HTTP/1.1" 200 OK
+[2026-08-27 05:21:37] INFO:     174.1.60.5:50774 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:37] INFO:     174.1.60.5:50788 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:37] INFO:     174.1.60.5:50802 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:37] INFO:     174.1.60.5:50814 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:37] INFO:     174.1.60.5:50826 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:37] INFO:     174.1.60.5:50840 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:37] INFO:     174.1.60.5:50844 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:37] INFO:     174.1.60.5:50846 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:43 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 2, #pending-token: 8652, #bootstrap-req: 2, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 44.89
+[2026-08-27 05:21:44 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 2, #pending-token: 8652, #bootstrap-req: 2, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 44.68
+[2026-08-27 05:21:44 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 3, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 5768.82
+[2026-08-27 05:21:45 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 3, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 5827.47
+[2026-08-27 05:21:45 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 3, #pending-token: 9185, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9671.83
+[2026-08-27 05:21:45 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 3, #pending-token: 9185, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 19750.66
+[2026-08-27 05:21:46 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 2, #pending-token: 993, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6916.66
+[2026-08-27 05:21:46 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 2, #pending-token: 993, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8672.27
+[2026-08-27 05:21:47 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 2, #pending-token: 9680, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10637.56
+[2026-08-27 05:21:47 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 2, #pending-token: 9680, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10354.39
+[2026-08-27 05:21:48 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1488, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8222.75
+[2026-08-27 05:21:48 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1488, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8344.22
+[2026-08-27 05:21:49 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 10199, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10480.00
+[2026-08-27 05:21:49 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 10199, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10387.32
+[2026-08-27 05:21:50 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 18858, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8578.91
+[2026-08-27 05:21:50 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 18858, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8578.44
+[2026-08-27 05:21:50 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 27591, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9988.39
+[2026-08-27 05:21:51 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 27591, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9972.71
+[2026-08-27 05:21:51 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 53023, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8837.11
+[2026-08-27 05:21:51] INFO:     174.1.60.5:43264 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:52 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 53023, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9171.43
+[2026-08-27 05:21:52 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 44876, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9896.54
+[2026-08-27 05:21:52 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 44876, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10032.97
+[2026-08-27 05:21:53 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 36684, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9174.07
+[2026-08-27 05:21:53] INFO:     174.1.60.5:43266 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:53 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 36684, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9172.94
+[2026-08-27 05:21:54 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 28504, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9586.45
+[2026-08-27 05:21:54 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 28504, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9826.60
+[2026-08-27 05:21:55 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 20312, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9357.29
+[2026-08-27 05:21:55] INFO:     174.1.60.5:43280 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:55 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 20312, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9123.76
+[2026-08-27 05:21:56 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 12139, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9447.89
+[2026-08-27 05:21:56 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 12139, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9658.04
+[2026-08-27 05:21:57 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 3947, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9443.97
+[2026-08-27 05:21:57] INFO:     174.1.60.5:43282 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:57 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 3947, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9638.68
+[2026-08-27 05:21:57 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 3968, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7366.63
+[2026-08-27 05:21:57 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 3968, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8902.23
+[2026-08-27 05:21:57] INFO:     174.1.60.5:43298 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:57] INFO:     174.1.60.5:43310 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:21:59] INFO:     174.1.60.5:43312 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:00] INFO:     174.1.60.5:43328 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:02 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 0, #pending-token: 8608, #bootstrap-req: 4, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1515.73
+[2026-08-27 05:22:03 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 0, #pending-token: 8608, #bootstrap-req: 4, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1303.59
+[2026-08-27 05:22:04 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 416, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4709.52
+[2026-08-27 05:22:04 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 416, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8416.60
+[2026-08-27 05:22:05 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 9149, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10991.56
+[2026-08-27 05:22:05 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 9149, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9867.38
+[2026-08-27 05:22:06 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 957, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7879.22
+[2026-08-27 05:22:06 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 957, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7915.28
+[2026-08-27 05:22:07 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 9638, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10712.86
+[2026-08-27 05:22:07 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 9638, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11079.28
+[2026-08-27 05:22:08 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1446, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8251.19
+[2026-08-27 05:22:08 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1446, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8053.96
+[2026-08-27 05:22:09 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 10033, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10494.65
+[2026-08-27 05:22:09 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 10033, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11035.63
+[2026-08-27 05:22:09 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1841, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8664.83
+[2026-08-27 05:22:10 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1841, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8259.51
+[2026-08-27 05:22:10 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1856, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6733.51
+[2026-08-27 05:22:10 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1856, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2578.77
+[2026-08-27 05:22:11 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8540, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7967.20
+[2026-08-27 05:22:11] INFO:     174.1.60.5:53576 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:11 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8540, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9138.35
+[2026-08-27 05:22:12 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 17066, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6059.77
+[2026-08-27 05:22:12 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 17066, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8568.79
+[2026-08-27 05:22:13 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8910, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10873.13
+[2026-08-27 05:22:13] INFO:     174.1.60.5:53584 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:13 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8910, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9603.70
+[2026-08-27 05:22:14 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 17466, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7794.12
+[2026-08-27 05:22:14 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 17466, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8622.72
+[2026-08-27 05:22:15 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9324, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10734.99
+[2026-08-27 05:22:15] INFO:     174.1.60.5:53590 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:15 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9324, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9808.25
+[2026-08-27 05:22:16 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 17751, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8115.61
+[2026-08-27 05:22:16 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 17751, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8814.22
+[2026-08-27 05:22:16 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9579, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10679.21
+[2026-08-27 05:22:17] INFO:     174.1.60.5:53596 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:17 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9579, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9729.78
+[2026-08-27 05:22:17 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1387, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8283.85
+[2026-08-27 05:22:18 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1387, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9429.57
+[2026-08-27 05:22:18 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1408, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6343.40
+[2026-08-27 05:22:18 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1408, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8479.29
+[2026-08-27 05:22:18] INFO:     174.1.60.5:53608 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:18] INFO:     174.1.60.5:53620 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:20] INFO:     174.1.60.5:53636 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:21] INFO:     174.1.60.5:44034 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:23 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8547, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1526.94
+[2026-08-27 05:22:24 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8547, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1315.46
+[2026-08-27 05:22:25 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 355, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4760.70
+[2026-08-27 05:22:25 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 355, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8428.33
+[2026-08-27 05:22:26 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8932, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10829.34
+[2026-08-27 05:22:26 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8932, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9556.73
+[2026-08-27 05:22:27 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 740, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7781.14
+[2026-08-27 05:22:27 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 740, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7932.50
+[2026-08-27 05:22:27 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9259, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10727.88
+[2026-08-27 05:22:28 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9259, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10866.67
+[2026-08-27 05:22:28 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1067, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8113.36
+[2026-08-27 05:22:29 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1067, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8138.49
+[2026-08-27 05:22:29 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9745, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10655.24
+[2026-08-27 05:22:29 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9745, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10900.49
+[2026-08-27 05:22:30 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1553, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8338.75
+[2026-08-27 05:22:30 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1553, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8227.97
+[2026-08-27 05:22:30 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1600, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6665.39
+[2026-08-27 05:22:31 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1600, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2199.04
+[2026-08-27 05:22:31 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8602, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7710.31
+[2026-08-27 05:22:31] INFO:     174.1.60.5:37214 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:32 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8602, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9261.97
+[2026-08-27 05:22:33 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 410, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6121.64
+[2026-08-27 05:22:33 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 410, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8576.46
+[2026-08-27 05:22:34 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8696, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10967.17
+[2026-08-27 05:22:34] INFO:     174.1.60.5:37230 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:34 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8696, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9567.88
+[2026-08-27 05:22:35 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 504, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7845.65
+[2026-08-27 05:22:35 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 504, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8659.79
+[2026-08-27 05:22:35 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9164, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10983.83
+[2026-08-27 05:22:35] INFO:     174.1.60.5:37246 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:36 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9164, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9874.30
+[2026-08-27 05:22:36 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 972, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7934.46
+[2026-08-27 05:22:36 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 972, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8762.34
+[2026-08-27 05:22:37 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9322, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10758.34
+[2026-08-27 05:22:37] INFO:     174.1.60.5:37262 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:37 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9322, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9763.35
+[2026-08-27 05:22:38 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1130, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8224.94
+[2026-08-27 05:22:38 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1130, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9463.33
+[2026-08-27 05:22:38 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1152, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6381.46
+[2026-08-27 05:22:38 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1152, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8785.04
+[2026-08-27 05:22:38] INFO:     174.1.60.5:37274 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:39] INFO:     174.1.60.5:37280 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:40] INFO:     174.1.60.5:37284 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:42] INFO:     174.1.60.5:57706 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:44 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8602, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1531.42
+[2026-08-27 05:22:44 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8602, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1318.33
+[2026-08-27 05:22:45 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 410, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4775.67
+[2026-08-27 05:22:45 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 410, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8534.85
+[2026-08-27 05:22:46 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 9105, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10895.13
+[2026-08-27 05:22:46 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 9105, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9621.85
+[2026-08-27 05:22:47 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 913, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7859.49
+[2026-08-27 05:22:47 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 913, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8012.84
+[2026-08-27 05:22:48 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9375, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10691.27
+[2026-08-27 05:22:48 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9375, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10846.99
+[2026-08-27 05:22:49 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1183, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8180.68
+[2026-08-27 05:22:49 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1183, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8142.25
+[2026-08-27 05:22:50 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9868, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10633.74
+[2026-08-27 05:22:50 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9868, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10953.99
+[2026-08-27 05:22:51 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1676, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8427.87
+[2026-08-27 05:22:51 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1676, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8243.07
+[2026-08-27 05:22:51 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6603.73
+[2026-08-27 05:22:52 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2380.37
+[2026-08-27 05:22:52 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8467, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7941.05
+[2026-08-27 05:22:52] INFO:     174.1.60.5:59718 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:52 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8467, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9276.71
+[2026-08-27 05:22:53 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 275, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6098.49
+[2026-08-27 05:22:53 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 275, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8603.25
+[2026-08-27 05:22:54 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8988, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11035.64
+[2026-08-27 05:22:54] INFO:     174.1.60.5:59730 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:54 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8988, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9672.42
+[2026-08-27 05:22:55 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 796, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7825.33
+[2026-08-27 05:22:55 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 796, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8635.70
+[2026-08-27 05:22:56 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9363, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10703.37
+[2026-08-27 05:22:56] INFO:     174.1.60.5:59738 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:56 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9363, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9819.19
+[2026-08-27 05:22:57 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 1171, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8175.11
+[2026-08-27 05:22:57 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 1171, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8823.65
+[2026-08-27 05:22:58 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9880, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10599.05
+[2026-08-27 05:22:58] INFO:     174.1.60.5:59744 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:22:58 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9880, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9922.77
+[2026-08-27 05:22:59 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1688, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8423.21
+[2026-08-27 05:22:59 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1688, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9407.16
+[2026-08-27 05:22:59 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6506.01
+[2026-08-27 05:22:59 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8452.04
+[2026-08-27 05:22:59] INFO:     174.1.60.5:59746 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:23:00] INFO:     174.1.60.5:59748 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:23:01] INFO:     174.1.60.5:59088 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:23:03] INFO:     174.1.60.5:59094 - "POST /generate HTTP/1.1" 200 OK
+[2026-08-27 05:23:04 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8601, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1533.96
+[2026-08-27 05:23:05 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8601, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1320.52
+[2026-08-27 05:23:06 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 409, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4797.37
+[2026-08-27 05:23:06 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 409, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8455.98
+[2026-08-27 05:23:07 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 9104, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10736.15
+[2026-08-27 05:23:07 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 9104, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9859.62
+[2026-08-27 05:23:08 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 912, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8011.36
+[2026-08-27 05:23:08 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 912, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8059.42
+[2026-08-27 05:23:08 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9485, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10534.48
+[2026-08-27 05:23:09 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9485, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10717.32
+[2026-08-27 05:23:09 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1293, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8282.08
+[2026-08-27 05:23:10 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1293, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8188.84
+[2026-08-27 05:23:10 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9900, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10556.34
+[2026-08-27 05:23:10 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9900, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10839.28
+[2026-08-27 05:23:11 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1708, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8457.93
+[2026-08-27 05:23:11 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1708, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8302.17
+[2026-08-27 05:23:11 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6651.64
+[2026-08-27 05:23:12 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2387.61
+[2026-08-27 05:23:12 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8525, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7923.97
+[2026-08-27 05:23:13 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8525, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9210.87
+[2026-08-27 05:23:14 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 333, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6091.62
+[2026-08-27 05:23:14 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 333, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8517.86
+[2026-08-27 05:23:15 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8909, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10878.98
+[2026-08-27 05:23:15 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8909, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9736.23
+[2026-08-27 05:23:16 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 717, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7902.78
+[2026-08-27 05:23:16 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 717, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8689.24
+[2026-08-27 05:23:16 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9452, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10764.37
+[2026-08-27 05:23:17 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9452, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9778.82
+[2026-08-27 05:23:17 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 1260, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8087.43
+[2026-08-27 05:23:18 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 1260, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8848.19
+[2026-08-27 05:23:18 PP0 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9830, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10642.74
+[2026-08-27 05:23:18 PP1 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9830, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9957.02
+[2026-08-27 05:23:19 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1638, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8506.01
+[2026-08-27 05:23:19 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1638, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9493.47
+[2026-08-27 05:23:19 PP0 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1664, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6758.94
+[2026-08-27 05:23:19 PP1 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1664, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9006.24
+[2026-08-27 05:23:20] INFO:     174.1.60.5:56558 - "GET /server_info HTTP/1.1" 200 OK
+[2026-08-27 05:23:20] INFO:     174.1.60.5:56572 - "GET /server_info HTTP/1.1" 200 OK
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/logs/manual_p_node1.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/logs/manual_p_node1.log
new file mode 100644
index 0000000..63052e9
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/logs/manual_p_node1.log
@@ -0,0 +1,1043 @@
+/usr/local/lib/python3.12/dist-packages/torchao/quantization/quant_api.py:1731: SyntaxWarning: invalid escape sequence '\.'
+  """Configuration class for applying different quantization configs to modules or parameters based on their fully qualified names (FQNs).
+/sgl-workspace/sglang/python/sglang/launch_server.py:56: UserWarning: 'python -m sglang.launch_server' is still supported, but 'sglang serve' is the recommended entrypoint.
+  Example: sglang serve --model-path <model> [options]
+  warnings.warn(
+[2026-08-27 04:28:43] Breakable CUDA graph is incompatible with MLA attention; disabling prefill CUDA graph.
+[2026-08-27 04:28:43] Attention backend not specified. Use triton backend by default.
+[2026-08-27 04:28:43] Pipeline parallelism is incompatible with overlap schedule.
+[2026-08-27 04:28:46] server_args=ServerArgs(model_path='/data/hf_models/Kimi-K3', tokenizer_path='/data/hf_models/Kimi-K3', tokenizer_mode='auto', tokenizer_backend='huggingface', tokenizer_worker_num=1, detokenizer_worker_num=1, skip_tokenizer_init=False, load_format='auto', model_loader_extra_config='{}', trust_remote_code=True, context_length=None, is_embedding=False, enable_multimodal=None, revision=None, model_impl='auto', model_config_parser='auto', json_model_override_args='{}', dtype='auto', quantization=None, quantization_param_path=None, kv_cache_dtype='auto', enable_fp32_lm_head=False, modelopt_quant=None, modelopt_checkpoint_restore_path=None, modelopt_checkpoint_save_path=None, modelopt_export_path=None, quantize_and_serve=False, rl_quant_profile=None, enable_tf32_matmul=False, mem_fraction_static=0.88, max_running_requests=None, max_queued_requests=None, max_total_tokens=None, chunked_prefill_size=8192, enable_dynamic_chunking=False, max_prefill_tokens=16384, prefill_max_requests=None, schedule_policy='fcfs', enable_priority_scheduling=False, disable_priority_preemption=False, default_priority_value=None, abort_on_priority_when_disabled=False, schedule_low_priority_values_first=False, priority_scheduling_preemption_threshold=10, retraction_policy='length', schedule_conservativeness=1.0, page_size=64, swa_full_tokens_ratio=0.8, disable_hybrid_swa_memory=False, radix_eviction_policy='lru', prefill_only_disable_kv_cache=False, disable_radix_cache=True, enable_page_major_kv_layout=False, enable_unified_memory=False, disable_chunked_prefix_cache=False, disable_overlap_schedule=True, num_continuous_decode_steps=1, scheduler_recv_interval=1, enable_mixed_chunk=False, nccl_port=None, dist_timeout=3600, dist_init_addr='174.1.60.1:20000', nnodes=4, node_rank=1, tp_size=4, dcp_size=1, pp_size=8, pp_max_micro_batch_size=None, pp_async_batch_depth=0, dp_size=1, load_balance_method='follow_bootstrap_room', attn_cp_size=1, moe_dp_size=1, dcp_comm_backend='ag_rs', dcp_replicate_q_proj=None, dwdp_size=1, enable_prefill_cp=False, cp_strategy=None, enable_dsa_cache_layer_split=False, enable_dsa_prefill_context_parallel=False, dsa_prefill_cp_mode='round-robin-split', enable_prefill_context_parallel=False, prefill_cp_mode='in-seq-split', enable_cp_decode_attn_tp=False, enable_dp_attention=False, enable_dp_attention_local_control_broadcast=False, enable_dp_lm_head=False, enable_attn_tp_input_scattered=False, disable_attn_tp_gather=False, enable_p2p_check=False, device='cuda', base_gpu_id=0, gpu_id_step=1, random_seed=577755302, watchdog_timeout=300, soft_watchdog_timeout=None, sleep_on_idle=False, use_ray=False, custom_sigquit_handler=None, numa_node=None, gc_threshold=None, host='0.0.0.0', port=30000, fastapi_root_path='', smg_grpc_mode=False, grpc_mode=False, grpc_port=None, sidecar=None, sidecar_args=None, skip_server_warmup=False, warmups=None, enable_http2=False, ssl_keyfile=None, ssl_certfile=None, ssl_ca_certs=None, ssl_keyfile_password=None, enable_ssl_refresh=False, api_key=None, admin_api_key=None, served_model_name='kimi-k3', weight_version='default', chat_template=None, hf_chat_template_name=None, completion_template=None, file_storage_path='sglang_storage', enable_cache_report=False, reasoning_parser=None, default_chat_template_kwargs=None, strip_thinking_cache=False, enable_strict_thinking=False, tool_call_parser=None, tool_server=None, sampling_defaults='model', asr_max_buffer_seconds=60, asr_max_concurrent_sessions=32, preferred_sampling_params=None, allow_auto_truncate=False, stream_interval=1, batch_notify_size=16, stream_response_default_include_usage=False, incremental_streaming_output=False, enable_streaming_session=False, enable_session_radix_cache=False, log_level='info', log_level_http=None, log_requests=False, log_requests_level=2, log_requests_format='text', log_requests_target=None, uvicorn_access_log_exclude_prefixes=[], crash_dump_folder=None, show_time_cost=False, enable_metrics=False, smg_http_sidecar_port=None, enable_mfu_metrics=False, enable_metrics_for_all_schedulers=False, load_snapshot_publish_interval=15, tokenizer_metrics_custom_labels_header='x-custom-labels', tokenizer_metrics_allowed_custom_labels=None, extra_metric_labels=None, bucket_time_to_first_token=None, bucket_inter_token_latency=None, bucket_e2e_request_latency=None, prompt_tokens_buckets=None, generation_tokens_buckets=None, gc_warning_threshold_secs=0.0, decode_log_interval=40, enable_request_time_stats_logging=False, kv_events_config=None, enable_forward_pass_metrics=False, forward_pass_metrics_worker_id='', forward_pass_metrics_ipc_name=None, enable_trace=False, trace_modules='request', otlp_traces_endpoint='localhost:4317', export_metrics_to_file=False, export_metrics_to_file_dir=None, stat_loggers=None, constrained_json_whitespace_pattern=None, constrained_json_disable_any_whitespace=False, attention_backend='triton', decode_attention_backend=None, prefill_attention_backend=None, sampling_backend='flashinfer', grammar_backend='xgrammar', radix_cache_backend=None, mm_attention_backend=None, fp8_gemm_runner_backend='auto', fp4_gemm_runner_backend='auto', bf16_gemm_backend='auto', dsa_prefill_backend=None, dsa_decode_backend=None, dsa_paged_mqa_logits_backend='auto', dsa_topk_backend='sgl-kernel', disable_flashinfer_autotune=False, flashinfer_autotune_skip_ops=None, mamba_backend='triton', cuda_graph_config=CudaGraphConfig(decode=PhaseConfig(backend='disabled', max_bs=16, bs=[1, 2, 4, 8, 12, 16], tc_compiler='eager', full_prefill_max_req=None), prefill=PhaseConfig(backend='disabled', max_bs=2048, bs=[4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], tc_compiler='eager', full_prefill_max_req=None)), cuda_graph_backend_decode=None, cuda_graph_backend_prefill=None, cuda_graph_max_bs_decode=16, cuda_graph_max_bs_prefill=None, cuda_graph_bs_decode=None, cuda_graph_bs_prefill=None, cuda_graph_tc_compiler=None, disable_prefill_cuda_graph=False, disable_decode_cuda_graph=False, disable_cuda_graph=False, disable_cuda_graph_padding=False, enable_profile_cuda_graph=False, enable_cudagraph_gc=False, debug_cuda_graph=False, enable_layerwise_nvtx_marker=False, enable_nccl_nvls=False, enable_symm_mem=False, triton_attention_reduce_in_fp32=False, triton_attention_num_kv_splits=8, triton_attention_split_tile_size=None, flashinfer_mla_disable_ragged=False, enable_fused_qk_norm_rope=False, enable_precise_embedding_interpolation=False, enable_fused_moe_sum_all_reduce=False, enable_deepseek_v4_fp4_indexer=False, disable_custom_all_reduce=False, enable_mscclpp=False, enable_torch_symm_mem=False, enable_scattered_sconv=False, pre_warm_nccl=False, enable_quant_communications=False, enable_flashinfer_allreduce_fusion=False, enforce_disable_flashinfer_allreduce_fusion=False, flashinfer_allreduce_fusion_backend=None, enable_aiter_allreduce_fusion=False, enable_torch_compile=False, enable_torch_compile_debug_mode=False, torch_compile_max_bs=32, torchao_config='', speculative_algorithm=None, speculative_draft_model_path=None, speculative_draft_model_revision=None, speculative_draft_load_format=None, speculative_num_steps=None, speculative_eagle_topk=None, speculative_num_draft_tokens=None, speculative_dflash_block_size=None, speculative_dspark_block_size=None, speculative_dspark_sps_table_path=None, speculative_dspark_confidence_sts_path=None, speculative_dspark_align_verify_tokens_to_graph_tier=False, speculative_accept_threshold_single=1.0, speculative_accept_threshold_acc=1.0, speculative_use_rejection_sampling=False, speculative_token_map=None, speculative_attention_mode='prefill', speculative_draft_attention_backend=None, speculative_draft_window_size=None, speculative_moe_runner_backend='flashinfer_mxfp4', speculative_moe_a2a_backend=None, speculative_draft_model_quantization=None, speculative_skip_dp_mlp_sync=False, enable_multi_layer_eagle=False, speculative_adaptive=False, speculative_adaptive_config=None, decoupled_spec_bind_endpoint=None, decoupled_spec_connect_endpoints=None, decoupled_spec_rank=None, decoupled_spec_role='null', spec_trace_dir=None, speculative_ngram_min_bfs_breadth=1, speculative_ngram_max_bfs_breadth=10, speculative_ngram_match_type='BFS', speculative_ngram_max_trie_depth=18, speculative_ngram_capacity=10000000, speculative_ngram_external_corpus_path=None, speculative_ngram_external_sam_budget=0, speculative_ngram_external_corpus_max_tokens=10000000, ep_size=4, moe_a2a_backend='none', moe_runner_backend='flashinfer_mxfp4', flashinfer_mxfp4_moe_precision='default', deepep_mode='auto', fuseep_mode=2, deepep_dispatcher_output_dtype='auto', ep_num_redundant_experts=0, ep_dispatch_algorithm=None, init_expert_location='trivial', enable_eplb=False, eplb_algorithm='auto', eplb_rebalance_num_iterations=1000, eplb_rebalance_layers_per_chunk=None, eplb_min_rebalancing_utilization_threshold=1.0, expert_distribution_recorder_mode=None, expert_distribution_recorder_buffer_size=1000, enable_expert_distribution_metrics=False, deepep_config=None, moe_dense_tp_size=None, elastic_ep_backend=None, enable_elastic_expert_backup=False, mooncake_ib_device=None, enable_waterfill=False, ep_join_mode=None, ep_join_rank_offset=0, elastic_ep_initial_size=None, max_ep_size=None, elastic_ep_scale_timeout=600, elastic_ep_rejoin=False, disable_flashinfer_cutlass_moe_fp4_allgather=False, disable_shared_experts_fusion=False, enforce_shared_experts_fusion=False, max_mamba_cache_size=None, mamba_ssm_dtype=None, mamba_max_states_per_path=-1, enable_mamba_cache_stochastic_rounding=False, mamba_cache_philox_rounds=0, mamba_full_memory_ratio=0.36, mamba_radix_cache_strategy='extra_buffer_lazy', uses_mamba_radix_cache=False, mamba_track_interval=256, enable_int8_mamba_checkpoint=False, int8_mamba_ckpt_size=None, linear_attn_backend='triton', linear_attn_decode_backend=None, linear_attn_prefill_backend=None, linear_attn_verify_backend=None, enable_linear_replayssm=False, linear_replayssm_cache_len=16, enable_linear_replayssm_spec=False, enable_hierarchical_cache=False, hicache_ratio=2.0, hicache_size=0, hicache_write_policy='write_through', hicache_io_backend='kernel', hicache_mem_layout='page_first', hicache_storage_backend=None, hicache_storage_prefetch_policy='timeout', hicache_storage_backend_extra_config=None, enable_hisparse=False, hisparse_config=None, enable_broadcast_mm_inputs_process=False, enable_prefix_mm_cache=False, mm_enable_dp_encoder=False, mm_process_config={}, mm_processor_worker_num=0, mm_io_worker_num=0, limit_mm_data_per_request=None, enable_mm_global_cache=False, disable_fast_image_processor=False, mm_feature_transport='cpu', keep_mm_feature_on_device=False, enable_lora=None, enable_lora_overlap_loading=None, max_lora_rank=None, lora_target_modules=None, lora_paths=None, max_loaded_loras=None, max_loras_per_batch=8, lora_eviction_policy='lru', lora_backend='csgmv', max_lora_chunk_size=16, experts_shared_outer_loras=None, lora_use_virtual_experts=False, lora_strict_loading=False, lora_drain_wait_threshold=0.0, enable_two_batch_overlap=False, enable_single_batch_overlap=False, tbo_token_distribution_threshold=0.48, cpu_offload_gb=0, offload_group_size=-1, offload_num_in_group=1, offload_prefetch_step=1, offload_mode='cpu', enable_lmcache=False, lmcache_config_file=None, enable_flexkv=False, flexkv_config_file=None, kt_weight_path=None, kt_method='AMXINT4', kt_cpuinfer=None, kt_threadpool_count=2, kt_num_gpu_experts=None, kt_max_deferred_experts_per_token=None, dllm_algorithm=None, dllm_algorithm_config=None, dllm_fdfo=True, disaggregation_mode='prefill', disaggregation_transfer_backend='mooncake', disaggregation_bootstrap_port=28800, disaggregation_ib_device='mlx5_0,mlx5_1,mlx5_2,mlx5_3', disaggregation_decode_enable_radix_cache=False, disaggregation_decode_enable_offload_kvcache=False, num_reserved_decode_tokens=512, disaggregation_decode_extra_slots=None, disaggregation_decode_polling_interval=1, optimistic_prefill_attempts=0, encoder_only=False, language_only=False, encoder_transfer_backend='zmq_to_tokenizer', encoder_urls=[], encoder_bootstrap_port=8997, encoder_register_urls=[], enable_adaptive_dispatch_to_encoder=False, enable_pdmux=False, pdmux_config_path=None, sm_group_num=8, custom_weight_loader=[], weight_loader_disable_mmap=False, weight_loader_prefetch_checkpoints=False, weight_loader_prefetch_num_threads=4, weight_loader_drop_cache_after_load=False, remote_instance_weight_loader_seed_instance_ip=None, remote_instance_weight_loader_seed_instance_service_port=None, remote_instance_weight_loader_send_weights_group_ports=None, remote_instance_weight_loader_backend='nccl', remote_instance_weight_loader_start_seed_via_transfer_engine=False, engine_info_bootstrap_port=6789, modelexpress_config=None, download_dir=None, model_checksum=None, delete_ckpt_after_loading=False, decrypted_config_file=None, decrypted_draft_config_file=None, checkpoint_engine_wait_weights_before_ready=False, enable_prefill_delayer=False, prefill_delayer_max_delay_passes=30, prefill_delayer_token_usage_low_watermark=None, prefill_delayer_forward_passes_buckets=None, prefill_delayer_wait_seconds_buckets=None, prefill_delayer_queue_min_ratio=None, prefill_delayer_max_delay_ms=None, min_free_slots_delay=None, enable_deterministic_inference=False, rl_on_policy_target=None, kv_canary='none', kv_canary_real_data='none', kv_canary_sweep_interval=0, enable_dynamic_batch_tokenizer=False, dynamic_batch_tokenizer_batch_size=32, dynamic_batch_tokenizer_batch_timeout=0.002, enable_tokenizer_batch_encode=False, disable_tokenizer_batch_decode=False, debug_tensor_dump_output_folder=None, debug_tensor_dump_layers=None, debug_tensor_dump_input_file=None, enable_memory_saver=False, enable_weights_cpu_backup=False, enable_draft_weights_cpu_backup=False, enable_custom_logit_processor=False, enable_return_hidden_states=False, enable_return_routed_experts=False, enable_return_indexer_topk=False, disable_outlines_disk_cache=False, enable_mis=False, weight_cache_mode='off', weight_cache_socket=None, weight_cache_timeout=1800, forward_hooks=None, msprobe_dump_config=None)
+[2026-08-27 04:28:46] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:46] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:46] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:46] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:46] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:46] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:46] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:46] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:02 PP3 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:02 PP3 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:03 PP3 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:03 PP3 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:03 PP3 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:03 PP3 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:03 PP2 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:03 PP2 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:03 PP2 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:03 PP2 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:04 PP2 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:04 PP2 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:04 PP2 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:04 PP2 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:04 PP3 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:04 PP3 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:05 PP3 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:05 PP2 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:05 PP3 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:05 PP3 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:05 PP2 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:06.044592   241 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:06.044694   241 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:06.044710   241 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.2 port: 12001
+I0827 04:29:06.044759   241 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.2:16343
+I0827 04:29:06.044883   241 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:06.053642   242 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:06.053738   242 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:06.053754   242 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.2 port: 12001
+I0827 04:29:06.053797   242 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.2:15170
+I0827 04:29:06.053925   242 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:06.058039   237 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:06.058135   237 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:06.058149   237 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.2 port: 12001
+I0827 04:29:06.058192   237 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.2:15486
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:06.058215   236 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:06.058310   236 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:06.058316   237 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:06.058326   236 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.2 port: 12001
+I0827 04:29:06.058367   236 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.2:16140
+I0827 04:29:06.058492   236 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:06.059734   239 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:06.059828   239 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:06.059844   239 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.2 port: 12001
+I0827 04:29:06.059890   239 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.2:15755
+I0827 04:29:06.060027   239 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:06.064433   237 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:06.064513   236 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:06.065588   239 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:06.069945   237 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:06.070852   236 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:06.071113   239 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:06.071978   241 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:06.079348   236 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:06.079967   242 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:06.080070   241 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:06.087044   241 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:06.087074   242 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:06.094024   242 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:06.095021   241 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:06.095822   237 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:06.097810   241 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:06.097839   241 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:06.097889   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.099939   239 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:06.100984   242 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:06.103683   242 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:06.103708   242 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:06.103756   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.103785   236 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:06.103833   237 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:06.106014   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:06.106982   236 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:06.107007   236 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:06.107053   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.107446   241 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:02
+I0827 04:29:06.107475   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.107936   239 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:06.107971   237 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:06.107997   237 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:06.108048   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.110760   239 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:06.110790   239 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:06.110847   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.111824   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:06.113214   242 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:02
+I0827 04:29:06.113242   242 rdma_context.cpp:205] Using SIEVE endpoint store
+[2026-08-27 04:29:06 PP2 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+I0827 04:29:06.114821   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:06.115196   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+[2026-08-27 04:29:06 PP2 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+I0827 04:29:06.116189   241 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:02
+I0827 04:29:06.116214   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.116621   237 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:02
+I0827 04:29:06.116650   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.118325   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:06.119874   239 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:02
+I0827 04:29:06.119906   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.121601   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:06.122952   242 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:02
+I0827 04:29:06.122977   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.127393   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:06.128841   239 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:02
+I0827 04:29:06.128867   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.132040   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:06.133428   236 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:02
+I0827 04:29:06.133458   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.136312   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:06.137817   239 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:02
+I0827 04:29:06.137846   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.144285   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:06.144373   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:06.145742   237 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:02
+I0827 04:29:06.145776   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.145833   241 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:02
+I0827 04:29:06.145866   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.152084   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:06.153820   242 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:02
+I0827 04:29:06.153860   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.154352   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:06.155728   241 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:02
+I0827 04:29:06.155835   241 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:06 PP3 TP2 EP2] Init torch distributed begin.
+I0827 04:29:06.163965   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:06.164443   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:06.164605   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:06.165319   236 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:02
+I0827 04:29:06.165345   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.165951   239 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:02
+I0827 04:29:06.166070   239 transfer_engine_impl.cpp:388] installTransport, type=rdma
+I0827 04:29:06.166186   242 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:02
+I0827 04:29:06.166311   242 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:06 PP3 TP0 EP0] Init torch distributed begin.
+[2026-08-27 04:29:06 PP3 TP3 EP3] Init torch distributed begin.
+I0827 04:29:06.172065   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:06.173699   237 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:02
+I0827 04:29:06.173743   237 rdma_context.cpp:205] Using SIEVE endpoint store
+[2026-08-27 04:29:06 PP3 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+I0827 04:29:06.192251   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:06.193763   236 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:02
+I0827 04:29:06.193797   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.200259   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:06.201810   237 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:02
+I0827 04:29:06.201937   237 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:06 PP2 TP2 EP2] Init torch distributed begin.
+I0827 04:29:06.220070   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:06.221505   236 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:02
+I0827 04:29:06.221653   236 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:06 PP2 TP1 EP1] Init torch distributed begin.
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:06.613179   235 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:06.613286   235 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:06.613301   235 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.2 port: 12001
+I0827 04:29:06.613344   235 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.2:16382
+I0827 04:29:06.613492   235 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:06.618471   235 topology.cpp:127] Device mlx5_0 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:06.618719   238 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:06.618824   238 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:06.618839   238 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.2 port: 12001
+I0827 04:29:06.618882   238 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.2:16170
+I0827 04:29:06.619020   238 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:06.624203   238 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:06.629066   238 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:06.633975   238 topology.cpp:127] Device mlx5_2 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:06.636366   240 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:06.636454   240 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:06.636466   240 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.2 port: 12001
+I0827 04:29:06.636509   240 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.2:15840
+I0827 04:29:06.636675   240 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:06.638876   238 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:06.640780   238 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:06.640802   238 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:06.640842   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.641396   240 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:06.643196   235 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:06.645701   240 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:06.649896   235 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:06.650310   240 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:06.654819   240 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:06.656375   240 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:06.656396   240 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:06.656432   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.657156   235 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:06.659863   235 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:06.659895   235 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:06.659950   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.661671   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:06.662956   240 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:02
+I0827 04:29:06.662979   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.664074   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:06.665762   238 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:02
+I0827 04:29:06.665796   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.667351   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:06.668808   235 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:02
+I0827 04:29:06.668838   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.676294   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:06.677721   235 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:02
+I0827 04:29:06.677748   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.692129   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:06.693621   240 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:02
+I0827 04:29:06.693648   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.696089   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:06.697444   238 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:02
+I0827 04:29:06.697470   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.707720   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:06.709362   235 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:02
+I0827 04:29:06.709394   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.719987   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:06.721463   240 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:02
+I0827 04:29:06.721489   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.728108   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:06.728855   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:06.729467   238 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:02
+I0827 04:29:06.729491   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:06.730392   240 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:02
+I0827 04:29:06.730533   240 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:06 PP3 TP1 EP1] Init torch distributed begin.
+I0827 04:29:06.736888   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:06.738317   238 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:02
+I0827 04:29:06.738422   238 transfer_engine_impl.cpp:388] installTransport, type=rdma
+I0827 04:29:06.738466   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+[2026-08-27 04:29:06 PP2 TP3 EP3] Init torch distributed begin.
+I0827 04:29:06.739869   235 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:02
+I0827 04:29:06.739982   235 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:06 PP2 TP0 EP0] Init torch distributed begin.
+[2026-08-27 04:29:26 PP3 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:26 PP2 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:27 PP2 TP3 EP3] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP2 TP1 EP1] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP2 TP2 EP2] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP2 TP0 EP0] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP3 TP3 EP3] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP3 TP0 EP0] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP3 TP2 EP2] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP3 TP1 EP1] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP3 TP3 EP3] Init torch distributed ends. elapsed=21.61 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP3 TP2 EP2] Init torch distributed ends. elapsed=21.62 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP3 TP1 EP1] Init torch distributed ends. elapsed=21.04 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP3 TP0 EP0] Init torch distributed ends. elapsed=21.61 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP2 TP2 EP2] Init torch distributed ends. elapsed=21.57 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP2 TP1 EP1] Init torch distributed ends. elapsed=21.55 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP2 TP0 EP0] Init torch distributed ends. elapsed=21.03 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP2 TP3 EP3] Init torch distributed ends. elapsed=21.04 s, mem usage=0.54 GB
+[2026-08-27 04:29:28 PP3 TP3 EP3] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP3 TP0 EP0] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP2 TP0 EP0] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP3 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP2 TP1 EP1] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP3 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP2 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP3 TP2 EP2] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP2 TP3 EP3] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP2 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP2 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP3 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP2 TP2 EP2] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP3 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP3 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP2 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP3 TP1 EP1] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP2 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP3 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP2 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP3 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP3 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP2 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP3 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP2 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP3 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP2 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP2 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP2 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP2 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP3 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP3 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:30:31 PP2 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:31 PP2 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:31 PP2 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:31 PP2 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:31 PP2 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:31 PP2 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:31 PP2 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:31 PP2 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:32 PP2 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:32 PP2 TP0 EP0] Load weight end. elapsed=64.92 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=36.14 GB, mem usage=45.91 GB.
+[2026-08-27 04:30:33 PP3 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:33 PP3 TP1 EP1] Load weight end. elapsed=65.31 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:34 PP2 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:34 PP2 TP1 EP1] Load weight end. elapsed=66.07 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=36.14 GB, mem usage=45.91 GB.
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO[TensorRT-LLM][INFO] Set logger level to INFO
+
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:34 PP3 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:35 PP3 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:35 PP3 TP0 EP0] Load weight end. elapsed=67.41 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:36 PP2 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:36 PP2 TP3 EP3] Load weight end. elapsed=68.04 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=36.14 GB, mem usage=45.91 GB.
+[2026-08-27 04:30:36 PP2 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:36 PP2 TP2 EP2] Load weight end. elapsed=68.23 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=36.14 GB, mem usage=45.91 GB.
+[2026-08-27 04:30:36 PP2 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:36 PP2 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:36 PP2 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:36 PP2 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:36 PP2 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:36 PP2 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:36 PP2 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:36 PP2 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:36 PP3 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:36 PP3 TP3 EP3] Load weight end. elapsed=68.86 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:37 PP3 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:37 PP3 TP2 EP2] Load weight end. elapsed=69.68 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:38 PP3 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:38 PP3 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:38 PP3 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:38 PP3 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:38 PP3 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:38 PP3 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:38 PP3 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:38 PP3 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:54 PP3 TP3 EP3] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP3 TP2 EP2] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP3 TP1 EP1] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP2 TP3 EP3] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP3 TP0 EP0] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP2 TP2 EP2] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP2 TP1 EP1] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP2 TP0 EP0] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP2 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:54 PP2 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:54 PP3 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP3 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP2 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:54 PP3 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP3 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP2 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:54 PP3 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP3 TP3 EP3] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP2 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP2 TP2 EP2] Memory pool end. avail mem=23.25 GB
+[2026-08-27 04:30:54 PP2 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP2 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP2 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP3 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP3 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP2 TP3 EP3] Memory pool end. avail mem=23.25 GB
+[2026-08-27 04:30:54 PP2 TP1 EP1] Memory pool end. avail mem=23.25 GB
+[2026-08-27 04:30:54 PP3 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP3 TP1 EP1] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP3 TP2 EP2] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP2 TP0 EP0] Memory pool end. avail mem=23.25 GB
+[2026-08-27 04:30:54 PP3 TP0 EP0] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP3 TP3 EP3] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp3_pp3_dp0.json
+[2026-08-27 04:30:54 PP2 TP1 EP1] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp1_pp2_dp0.json
+[2026-08-27 04:30:54 PP2 TP2 EP2] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp2_pp2_dp0.json
+[2026-08-27 04:30:54 PP3 TP1 EP1] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp1_pp3_dp0.json
+[2026-08-27 04:30:54 PP2 TP3 EP3] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp3_pp2_dp0.json
+[2026-08-27 04:30:54 PP3 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:30:54 PP3 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:30:54 PP3 TP2 EP2] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp2_pp3_dp0.json
+[2026-08-27 04:30:54 PP2 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:30:54 PP2 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:30:54 PP3 TP0 EP0] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp0_pp3_dp0.json
+[2026-08-27 04:30:54 PP2 TP0 EP0] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp0_pp2_dp0.json
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.65profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.60profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.79profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.51profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.86profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.27profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.62profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.98profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.86profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 12.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 13.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.93profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.92profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.09profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.63profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  6.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  7.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.33profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.58profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  8.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.56profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.23profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.63profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.36profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.09profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.97profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  8.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.30profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 13.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 13.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  8.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  8.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 11.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 11.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.23profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.48profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.23profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.53profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.27profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.59profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.62profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.97profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.60profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.46profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.12profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.68profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.77profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.62profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  6.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.87profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.30profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.40profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.83profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.94profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.59profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.66profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.29profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.23profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:01,  3.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.68profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.75profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  7.01profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.22profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.65profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.62profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.46profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.22profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  7.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:01<00:06,  1.03s/profile]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:01,  3.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.50profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.93profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.48profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.11profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.12profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.30profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.68profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  7.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 11.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  9.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 12.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 11.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 11.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 11.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.11profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.31profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.65profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.31profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.35profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.49profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:01<00:01,  2.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.65profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.68profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  7.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.09profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 11.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.88profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.98profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.50profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.62profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.51profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.91profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:01,  3.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.31profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.86profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.27profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.59profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.17profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.76profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.73profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  9.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.37profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.15profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.98profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.65profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.97profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.23profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.46profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  9.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.03profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.57profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.40profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.40profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.57profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.13profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.56profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  6.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.73profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.23profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  3.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.86profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.77profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.92profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.37profile/s]
+[2026-08-27 04:31:26 PP2 TP3 EP3] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.39profile/s]
+[2026-08-27 04:31:26 PP2 TP2 EP2] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.75profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.46profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.35profile/s]
+[2026-08-27 04:31:26 PP2 TP1 EP1] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.12profile/s]
+[2026-08-27 04:31:26 PP2 TP0 EP0] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.66profile/s][2026-08-27 04:31:26 PP2 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:26 PP2 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:26 PP2 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:26 PP2 TP0 EP0] max_total_num_tokens=3801408, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=54, context_len=1048576, available_gpu_mem=20.47 GB
+[2026-08-27 04:31:26 PP2 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:26 PP2 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:26 PP2 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:26 PP2 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:26 PP2 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.93profile/s]
+[2026-08-27 04:31:27 PP3 TP1 EP1] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.16profile/s]
+[2026-08-27 04:31:27 PP3 TP0 EP0] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.43profile/s]
+[2026-08-27 04:31:27 PP3 TP2 EP2] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.05profile/s]
+[2026-08-27 04:31:27 PP3 TP3 EP3] FlashInfer autotune completed.
+[2026-08-27 04:31:27 PP3 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:27 PP3 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:27 PP3 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:27 PP3 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:27 PP3 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:27 PP3 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:27 PP3 TP0 EP0] max_total_num_tokens=3801408, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=54, context_len=1048576, available_gpu_mem=16.54 GB
+[2026-08-27 04:31:27 PP3 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:27 PP3 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+W0827 04:31:30.590376  2521 memory_location.cpp:73] Failed to get NUMA node, addr: 0x513c5ec0, len: 6912: Operation not permitted [1]
+W0827 04:31:30.590432  2522 memory_location.cpp:73] Failed to get NUMA node, addr: 0x514a0900, len: 6912: Operation not permitted [1]
+W0827 04:31:30.590495  2525 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51226dc0, len: 55296: Operation not permitted [1]
+W0827 04:31:30.590492  2523 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5128db40, len: 6912: Operation not permitted [1]
+W0827 04:31:30.590533  2524 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5156ff80, len: 6912: Operation not permitted [1]
+W0827 04:31:30.590600  2526 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51320ac0, len: 55296: Operation not permitted [1]
+W0827 04:31:30.590636  2527 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51768540, len: 6912: Operation not permitted [1]
+W0827 04:31:30.590826  2528 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51483dc0, len: 13824: Operation not permitted [1]
+W0827 04:31:30.590914  2529 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51119380, len: 6912: Operation not permitted [1]
+W0827 04:31:30.590979  2530 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50fe1e00, len: 6912: Operation not permitted [1]
+W0827 04:31:30.846120  2547 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50ceebc0, len: 6912: Operation not permitted [1]
+W0827 04:31:30.846271  2548 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50c57640, len: 6912: Operation not permitted [1]
+W0827 04:31:30.846328  2549 memory_location.cpp:73] Failed to get NUMA node, addr: 0x502ed680, len: 6912: Operation not permitted [1]
+W0827 04:31:30.846354  2550 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50d1c800, len: 6912: Operation not permitted [1]
+W0827 04:31:30.846448  2551 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50510a40, len: 55296: Operation not permitted [1]
+W0827 04:31:30.846473  2552 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5051e280, len: 55296: Operation not permitted [1]
+W0827 04:31:30.846571  2553 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50d01340, len: 6912: Operation not permitted [1]
+W0827 04:31:30.846689  2554 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50ce3440, len: 13824: Operation not permitted [1]
+W0827 04:31:30.846858  2555 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5030f440, len: 6912: Operation not permitted [1]
+W0827 04:31:30.846926  2556 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50cae980, len: 6912: Operation not permitted [1]
+W0827 04:31:31.198441  2575 memory_location.cpp:73] Failed to get NUMA node, addr: 0x514f46c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.198501  2576 memory_location.cpp:73] Failed to get NUMA node, addr: 0x509c6400, len: 6912: Operation not permitted [1]
+W0827 04:31:31.198501  2577 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51513300, len: 6912: Operation not permitted [1]
+W0827 04:31:31.198679  2578 memory_location.cpp:73] Failed to get NUMA node, addr: 0x506eef40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.198787  2579 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50dc9180, len: 55296: Operation not permitted [1]
+W0827 04:31:31.198915  2580 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50904700, len: 55296: Operation not permitted [1]
+W0827 04:31:31.198982  2581 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50aa9440, len: 6912: Operation not permitted [1]
+W0827 04:31:31.199081  2582 memory_location.cpp:73] Failed to get NUMA node, addr: 0x515012c0, len: 13824: Operation not permitted [1]
+W0827 04:31:31.199148  2583 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5161ac80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.199185  2584 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50dd69c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.763885  2603 memory_location.cpp:73] Failed to get NUMA node, addr: 0x519f1500, len: 6912: Operation not permitted [1]
+W0827 04:31:31.763998  2605 memory_location.cpp:73] Failed to get NUMA node, addr: 0x517df940, len: 6912: Operation not permitted [1]
+W0827 04:31:31.763923  2604 memory_location.cpp:73] Failed to get NUMA node, addr: 0x514ab440, len: 6912: Operation not permitted [1]
+W0827 04:31:31.764139  2606 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51556180, len: 6912: Operation not permitted [1]
+W0827 04:31:31.764169  2608 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50ed8140, len: 55296: Operation not permitted [1]
+W0827 04:31:31.764184  2607 memory_location.cpp:73] Failed to get NUMA node, addr: 0x515df040, len: 55296: Operation not permitted [1]
+W0827 04:31:31.764356  2609 memory_location.cpp:73] Failed to get NUMA node, addr: 0x517e3580, len: 6912: Operation not permitted [1]
+W0827 04:31:31.764564  2610 memory_location.cpp:73] Failed to get NUMA node, addr: 0x519cdb80, len: 13824: Operation not permitted [1]
+W0827 04:31:31.764624  2611 memory_location.cpp:73] Failed to get NUMA node, addr: 0x519dbf80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.764714  2612 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50f08d80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.868036  2629 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50a7a680, len: 6912: Operation not permitted [1]
+W0827 04:31:31.868046  2630 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50b6e400, len: 6912: Operation not permitted [1]
+W0827 04:31:31.868139  2631 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50e5f580, len: 6912: Operation not permitted [1]
+W0827 04:31:31.868261  2632 memory_location.cpp:73] Failed to get NUMA node, addr: 0x503c7100, len: 6912: Operation not permitted [1]
+W0827 04:31:31.868330  2633 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50c94ac0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.868432  2635 memory_location.cpp:73] Failed to get NUMA node, addr: 0x509b2880, len: 6912: Operation not permitted [1]
+W0827 04:31:31.868502  2634 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50b14c80, len: 55296: Operation not permitted [1]
+W0827 04:31:31.868564  2636 memory_location.cpp:73] Failed to get NUMA node, addr: 0x503ee840, len: 13824: Operation not permitted [1]
+W0827 04:31:31.868708  2637 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51057d80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.868835  2638 memory_location.cpp:73] Failed to get NUMA node, addr: 0x509cd940, len: 6912: Operation not permitted [1]
+W0827 04:31:32.003983  2656 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50715b00, len: 6912: Operation not permitted [1]
+W0827 04:31:32.004096  2657 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51346e80, len: 6912: Operation not permitted [1]
+W0827 04:31:32.004005  2655 memory_location.cpp:73] Failed to get NUMA node, addr: 0x507176c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.004179  2658 memory_location.cpp:73] Failed to get NUMA node, addr: 0x507e3a40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.004302  2659 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5078aa80, len: 55296: Operation not permitted [1]
+W0827 04:31:32.004357  2660 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5135c980, len: 55296: Operation not permitted [1]
+W0827 04:31:32.004410  2661 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5099f180, len: 6912: Operation not permitted [1]
+W0827 04:31:32.004585  2662 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50b287c0, len: 13824: Operation not permitted [1]
+W0827 04:31:32.004739  2663 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5070b6c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.004812  2664 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50753540, len: 6912: Operation not permitted [1]
+W0827 04:31:32.144867  2688 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50aed700, len: 6912: Operation not permitted [1]
+W0827 04:31:32.144946  2689 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50792740, len: 6912: Operation not permitted [1]
+W0827 04:31:32.145030  2690 memory_location.cpp:73] Failed to get NUMA node, addr: 0x508fca40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.145135  2691 memory_location.cpp:73] Failed to get NUMA node, addr: 0x505d7c40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.145318  2693 memory_location.cpp:73] Failed to get NUMA node, addr: 0x505ec340, len: 55296: Operation not permitted [1]
+W0827 04:31:32.145387  2692 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50602cc0, len: 55296: Operation not permitted [1]
+W0827 04:31:32.145479  2694 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50748c00, len: 6912: Operation not permitted [1]
+W0827 04:31:32.145632  2695 memory_location.cpp:73] Failed to get NUMA node, addr: 0x507e3480, len: 13824: Operation not permitted [1]
+W0827 04:31:32.145767  2696 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50316400, len: 6912: Operation not permitted [1]
+W0827 04:31:32.145864  2697 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50788600, len: 6912: Operation not permitted [1]
+W0827 04:31:32.177534  2726 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fff3c80, len: 6912: Operation not permitted [1]
+W0827 04:31:32.177623  2728 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fad6780, len: 6912: Operation not permitted [1]
+W0827 04:31:32.177625  2727 memory_location.cpp:73] Failed to get NUMA node, addr: 0x501dc2c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.178279  2729 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50596480, len: 6912: Operation not permitted [1]
+W0827 04:31:32.178463  2730 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50222dc0, len: 55296: Operation not permitted [1]
+W0827 04:31:32.178498  2731 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5038f6c0, len: 55296: Operation not permitted [1]
+W0827 04:31:32.178581  2732 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4ffd7ec0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.178695  2733 memory_location.cpp:73] Failed to get NUMA node, addr: 0x506d84c0, len: 13824: Operation not permitted [1]
+W0827 04:31:32.178884  2734 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5078efc0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.178922  2735 memory_location.cpp:73] Failed to get NUMA node, addr: 0x502c7b00, len: 6912: Operation not permitted [1]
+[2026-08-27 04:31:32] Dummy health check server started in background thread at 0.0.0.0:30000
+[rank10]:[W827 04:32:32.605391940 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank11]:[W827 04:32:32.605815131 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank9]:[W827 04:32:32.606041691 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank8]:[W827 04:32:32.608313727 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank11]:[W827 04:33:19.698792619 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank8]:[W827 04:33:19.699392665 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank9]:[W827 04:33:19.699501937 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank10]:[W827 04:33:19.699589651 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank14]:[W827 04:33:19.728965278 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank12]:[W827 04:33:19.729026815 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank13]:[W827 04:33:19.729092243 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank15]:[W827 04:33:19.730463091 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank14]:[W827 04:33:25.763484084 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank13]:[W827 04:33:25.764518868 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank15]:[W827 04:33:25.764776298 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank12]:[W827 04:33:25.765271843 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[2026-08-27 04:35:10 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.29
+[2026-08-27 04:35:10 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.29
+[2026-08-27 04:35:29 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.39
+[2026-08-27 04:35:29 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.39
+[2026-08-27 04:35:31 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25.86
+[2026-08-27 04:35:31 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25.87
+[2026-08-27 04:43:23 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.14
+[2026-08-27 04:43:23 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.14
+[2026-08-27 04:48:13 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.22
+[2026-08-27 04:48:13 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.22
+[2026-08-27 04:48:14 PP2 TP1 EP1] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP2 TP2 EP2] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP2 TP0 EP0] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP3 TP2 EP2] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP3 TP0 EP0] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP3 TP1 EP1] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP2 TP3 EP3] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP3 TP3 EP3] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:57:14 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.12
+[2026-08-27 04:57:14 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.12
+[2026-08-27 05:04:21 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1088, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2.55
+[2026-08-27 05:04:21 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1088, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2.55
+[2026-08-27 05:04:41 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.19
+[2026-08-27 05:04:41 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.19
+[2026-08-27 05:05:39 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1.10
+[2026-08-27 05:05:39 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1.10
+[2026-08-27 05:07:32 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 72.23
+[2026-08-27 05:07:33 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 72.17
+[2026-08-27 05:07:33 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 78417.87
+[2026-08-27 05:07:33 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 501194.56
+[2026-08-27 05:07:33 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 30385.21
+[2026-08-27 05:07:33 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32725.49
+[2026-08-27 05:07:40 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8673, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1097.27
+[2026-08-27 05:07:40 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8673, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1082.51
+[2026-08-27 05:07:40 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 481, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 68995.34
+[2026-08-27 05:07:40 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 481, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 494490.91
+[2026-08-27 05:07:40 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33832.33
+[2026-08-27 05:07:40 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33008.60
+[2026-08-27 05:07:48 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8656, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1098.69
+[2026-08-27 05:07:48 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8656, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1082.21
+[2026-08-27 05:07:48 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 464, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 62757.93
+[2026-08-27 05:07:48 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 464, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 451735.55
+[2026-08-27 05:07:48 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 30632.65
+[2026-08-27 05:07:48 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 31628.03
+[2026-08-27 05:07:55 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8663, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1101.49
+[2026-08-27 05:07:55 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8663, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1087.16
+[2026-08-27 05:07:55 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 471, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 71370.09
+[2026-08-27 05:07:55 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 471, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 487386.35
+[2026-08-27 05:07:55 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 31311.94
+[2026-08-27 05:07:55 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 31458.64
+[2026-08-27 05:08:03 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8659, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1099.50
+[2026-08-27 05:08:03 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8659, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1082.83
+[2026-08-27 05:08:03 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 467, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 61030.58
+[2026-08-27 05:08:03 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 467, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 440213.50
+[2026-08-27 05:08:03 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33725.14
+[2026-08-27 05:08:03 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32738.01
+[2026-08-27 05:08:10 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8692, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1100.46
+[2026-08-27 05:08:11 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8692, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1085.28
+[2026-08-27 05:08:11 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 500, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 67432.08
+[2026-08-27 05:08:11 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 500, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 480206.29
+[2026-08-27 05:08:11 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32427.60
+[2026-08-27 05:08:11 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 30441.07
+[2026-08-27 05:08:18 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8557, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1101.25
+[2026-08-27 05:08:18 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8557, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1089.66
+[2026-08-27 05:08:18 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 365, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 85531.52
+[2026-08-27 05:08:18 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 365, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 467687.86
+[2026-08-27 05:08:18 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 22103.80
+[2026-08-27 05:08:18 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25225.78
+[2026-08-27 05:08:26 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8683, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1099.06
+[2026-08-27 05:08:26 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8683, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1084.12
+[2026-08-27 05:08:26 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 491, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 68454.69
+[2026-08-27 05:08:26 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 491, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 485556.59
+[2026-08-27 05:08:26 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33944.36
+[2026-08-27 05:08:26 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 30281.29
+[2026-08-27 05:11:45 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.32
+[2026-08-27 05:11:45 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.32
+[2026-08-27 05:14:42 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1088, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6.16
+[2026-08-27 05:14:42 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1088, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6.16
+[2026-08-27 05:17:47 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 44.24
+[2026-08-27 05:17:47 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 44.21
+[2026-08-27 05:17:47 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 74322.16
+[2026-08-27 05:17:47 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 505572.40
+[2026-08-27 05:17:47 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 29824.24
+[2026-08-27 05:17:47 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32304.11
+[2026-08-27 05:17:55 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8673, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1100.05
+[2026-08-27 05:17:55 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8673, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1085.77
+[2026-08-27 05:17:55 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 481, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 72167.09
+[2026-08-27 05:17:55 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 481, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 477186.86
+[2026-08-27 05:17:55 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 30671.49
+[2026-08-27 05:17:55 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 34433.71
+[2026-08-27 05:18:02 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8656, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1101.74
+[2026-08-27 05:18:02 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8656, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1087.80
+[2026-08-27 05:18:02 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 464, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 74681.51
+[2026-08-27 05:18:02 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 464, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 550093.83
+[2026-08-27 05:18:02 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33172.26
+[2026-08-27 05:18:02 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32409.88
+[2026-08-27 05:18:10 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8663, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1101.73
+[2026-08-27 05:18:10 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8663, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1087.58
+[2026-08-27 05:18:10 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 471, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 73073.93
+[2026-08-27 05:18:10 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 471, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 528029.69
+[2026-08-27 05:18:10 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32561.99
+[2026-08-27 05:18:10 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 31690.72
+[2026-08-27 05:18:17 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8659, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1099.69
+[2026-08-27 05:18:17 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8659, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1085.59
+[2026-08-27 05:18:17 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 467, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 72775.07
+[2026-08-27 05:18:17 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 467, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 520101.00
+[2026-08-27 05:18:17 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 30837.63
+[2026-08-27 05:18:18 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32670.85
+[2026-08-27 05:18:25 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8692, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1100.48
+[2026-08-27 05:18:25 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8692, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1086.22
+[2026-08-27 05:18:25 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 500, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 72864.89
+[2026-08-27 05:18:25 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 500, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 524671.87
+[2026-08-27 05:18:25 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 31161.52
+[2026-08-27 05:18:25 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 35156.96
+[2026-08-27 05:18:32 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8557, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1102.45
+[2026-08-27 05:18:33 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8557, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1090.72
+[2026-08-27 05:18:33 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 365, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 85680.60
+[2026-08-27 05:18:33 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 365, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 482538.50
+[2026-08-27 05:18:33 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 24798.12
+[2026-08-27 05:18:33 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 23973.27
+[2026-08-27 05:18:40 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8683, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1102.03
+[2026-08-27 05:18:40 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8683, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1087.43
+[2026-08-27 05:18:40 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 491, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 70828.52
+[2026-08-27 05:18:40 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 491, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 529856.70
+[2026-08-27 05:18:40 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32426.64
+[2026-08-27 05:18:40 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33626.30
+[2026-08-27 05:21:44 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 2, #pending-token: 8652, #bootstrap-req: 2, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 44.47
+[2026-08-27 05:21:45 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 3, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9902.45
+[2026-08-27 05:21:45 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 2, #pending-token: 8652, #bootstrap-req: 2, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 44.27
+[2026-08-27 05:21:46 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 3, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8141.53
+[2026-08-27 05:21:46 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 3, #pending-token: 9185, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7907.49
+[2026-08-27 05:21:47 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 3, #pending-token: 9185, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10345.23
+[2026-08-27 05:21:47 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 2, #pending-token: 993, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9860.59
+[2026-08-27 05:21:48 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 2, #pending-token: 993, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8351.60
+[2026-08-27 05:21:48 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 2, #pending-token: 9680, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8645.66
+[2026-08-27 05:21:49 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 2, #pending-token: 9680, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10382.09
+[2026-08-27 05:21:49 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1488, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10477.17
+[2026-08-27 05:21:50 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1488, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8583.82
+[2026-08-27 05:21:50 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 10199, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8562.77
+[2026-08-27 05:21:51 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 18858, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9978.66
+[2026-08-27 05:21:51 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 10199, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9847.23
+[2026-08-27 05:21:51 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 18858, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9275.72
+[2026-08-27 05:21:51 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 27591, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9156.16
+[2026-08-27 05:21:52 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 53023, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10350.08
+[2026-08-27 05:21:52 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 27591, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9601.22
+[2026-08-27 05:21:53 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 44876, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9145.49
+[2026-08-27 05:21:53 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 53023, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9622.28
+[2026-08-27 05:21:54 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 36684, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9799.45
+[2026-08-27 05:21:54 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 44876, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9677.55
+[2026-08-27 05:21:55 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 28504, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10333.04
+[2026-08-27 05:21:55 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 36684, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9394.13
+[2026-08-27 05:21:56 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 28504, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10860.67
+[2026-08-27 05:21:56 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 20312, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9527.15
+[2026-08-27 05:21:57 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 20312, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9357.72
+[2026-08-27 05:21:57 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 12139, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9361.75
+[2026-08-27 05:21:57 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 12139, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 15465.96
+[2026-08-27 05:21:57 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 3947, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 15452.41
+[2026-08-27 05:21:57 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 3947, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 310446.55
+[2026-08-27 05:21:57 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 3968, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 156702.42
+[2026-08-27 05:21:57 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 3968, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 139513.02
+[2026-08-27 05:22:04 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 0, #pending-token: 8608, #bootstrap-req: 4, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1132.42
+[2026-08-27 05:22:05 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 416, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10805.74
+[2026-08-27 05:22:05 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 0, #pending-token: 8608, #bootstrap-req: 4, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1019.78
+[2026-08-27 05:22:06 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 416, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7895.05
+[2026-08-27 05:22:06 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 9149, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7392.79
+[2026-08-27 05:22:07 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 957, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11340.60
+[2026-08-27 05:22:07 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 9149, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11099.21
+[2026-08-27 05:22:08 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 957, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8063.56
+[2026-08-27 05:22:08 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 9638, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7911.72
+[2026-08-27 05:22:09 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1446, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11076.15
+[2026-08-27 05:22:09 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 9638, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10949.54
+[2026-08-27 05:22:10 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1446, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8408.34
+[2026-08-27 05:22:10 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 10033, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8234.48
+[2026-08-27 05:22:10 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1841, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10913.26
+[2026-08-27 05:22:10 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 10033, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10431.88
+[2026-08-27 05:22:11 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1856, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2259.05
+[2026-08-27 05:22:11 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1841, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9802.26
+[2026-08-27 05:22:12 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8540, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8332.93
+[2026-08-27 05:22:12 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1856, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1872.96
+[2026-08-27 05:22:13 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 17066, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9894.47
+[2026-08-27 05:22:13 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8540, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10158.45
+[2026-08-27 05:22:14 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8910, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8389.00
+[2026-08-27 05:22:14 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 17066, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8419.18
+[2026-08-27 05:22:15 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 17466, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10331.84
+[2026-08-27 05:22:15 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8910, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10077.68
+[2026-08-27 05:22:16 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9324, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9373.73
+[2026-08-27 05:22:16 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 17466, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8567.43
+[2026-08-27 05:22:16 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 17751, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10615.28
+[2026-08-27 05:22:17 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9324, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11315.42
+[2026-08-27 05:22:17 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 17751, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8963.12
+[2026-08-27 05:22:17 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9579, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8324.62
+[2026-08-27 05:22:18 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1387, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 35673.33
+[2026-08-27 05:22:18 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9579, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 35404.74
+[2026-08-27 05:22:18 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1387, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 314704.56
+[2026-08-27 05:22:18 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1408, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 52789.70
+[2026-08-27 05:22:18 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1408, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 13474.13
+[2026-08-27 05:22:25 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8547, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1143.27
+[2026-08-27 05:22:26 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 355, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10109.37
+[2026-08-27 05:22:26 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8547, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1034.06
+[2026-08-27 05:22:27 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 355, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7893.67
+[2026-08-27 05:22:27 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8932, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7504.02
+[2026-08-27 05:22:28 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 740, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11535.13
+[2026-08-27 05:22:28 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8932, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10838.23
+[2026-08-27 05:22:29 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 740, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8155.37
+[2026-08-27 05:22:29 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9259, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7789.29
+[2026-08-27 05:22:29 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1067, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10733.11
+[2026-08-27 05:22:29 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9259, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10388.64
+[2026-08-27 05:22:30 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1067, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8544.20
+[2026-08-27 05:22:30 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9745, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8357.94
+[2026-08-27 05:22:31 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1553, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10706.17
+[2026-08-27 05:22:31 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9745, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10445.99
+[2026-08-27 05:22:32 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1600, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1952.39
+[2026-08-27 05:22:32 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1553, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9931.31
+[2026-08-27 05:22:33 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8602, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8370.54
+[2026-08-27 05:22:33 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1600, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1624.80
+[2026-08-27 05:22:34 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 410, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9785.83
+[2026-08-27 05:22:34 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8602, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9969.67
+[2026-08-27 05:22:35 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8696, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8591.23
+[2026-08-27 05:22:35 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 410, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8607.27
+[2026-08-27 05:22:35 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 504, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10402.38
+[2026-08-27 05:22:35 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8696, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10102.69
+[2026-08-27 05:22:36 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9164, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9308.92
+[2026-08-27 05:22:36 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 504, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8520.94
+[2026-08-27 05:22:37 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 972, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10664.47
+[2026-08-27 05:22:37 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9164, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11299.50
+[2026-08-27 05:22:38 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 972, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8956.53
+[2026-08-27 05:22:38 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9322, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8269.42
+[2026-08-27 05:22:38 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1130, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 43785.83
+[2026-08-27 05:22:38 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9322, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 43114.10
+[2026-08-27 05:22:38 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1130, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 348034.30
+[2026-08-27 05:22:38 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1152, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 43670.18
+[2026-08-27 05:22:38 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1152, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10950.18
+[2026-08-27 05:22:45 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8602, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1146.51
+[2026-08-27 05:22:46 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 410, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10113.13
+[2026-08-27 05:22:46 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8602, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1036.82
+[2026-08-27 05:22:47 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 410, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8059.88
+[2026-08-27 05:22:47 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 9105, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7668.92
+[2026-08-27 05:22:48 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 913, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11406.02
+[2026-08-27 05:22:48 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 9105, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11028.76
+[2026-08-27 05:22:49 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 913, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8034.52
+[2026-08-27 05:22:49 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9375, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7803.47
+[2026-08-27 05:22:50 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1183, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10701.85
+[2026-08-27 05:22:50 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9375, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10338.95
+[2026-08-27 05:22:51 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1183, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8629.50
+[2026-08-27 05:22:51 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9868, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8448.89
+[2026-08-27 05:22:52 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1676, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10696.20
+[2026-08-27 05:22:52 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9868, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10383.50
+[2026-08-27 05:22:52 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2088.70
+[2026-08-27 05:22:52 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1676, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9980.67
+[2026-08-27 05:22:53 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8467, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8474.75
+[2026-08-27 05:22:53 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1765.75
+[2026-08-27 05:22:54 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 275, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10047.97
+[2026-08-27 05:22:54 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8467, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10138.75
+[2026-08-27 05:22:55 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8988, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8471.06
+[2026-08-27 05:22:55 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 275, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8570.54
+[2026-08-27 05:22:56 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 796, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10303.51
+[2026-08-27 05:22:56 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8988, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9711.56
+[2026-08-27 05:22:57 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9363, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9482.47
+[2026-08-27 05:22:57 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 796, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8960.80
+[2026-08-27 05:22:58 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 1171, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10488.36
+[2026-08-27 05:22:58 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9363, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11295.90
+[2026-08-27 05:22:59 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 1171, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8985.86
+[2026-08-27 05:22:59 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9880, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8497.44
+[2026-08-27 05:22:59 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1688, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 30171.97
+[2026-08-27 05:22:59 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9880, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 29961.56
+[2026-08-27 05:22:59 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1688, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 384839.25
+[2026-08-27 05:22:59 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 77829.47
+[2026-08-27 05:22:59 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 17255.50
+[2026-08-27 05:23:06 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8601, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1148.36
+[2026-08-27 05:23:07 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 409, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10091.44
+[2026-08-27 05:23:07 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8601, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1039.48
+[2026-08-27 05:23:08 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 409, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8052.72
+[2026-08-27 05:23:08 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 9104, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7736.57
+[2026-08-27 05:23:09 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 912, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11209.86
+[2026-08-27 05:23:09 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 9104, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10701.52
+[2026-08-27 05:23:10 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 912, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8193.31
+[2026-08-27 05:23:10 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9485, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7925.16
+[2026-08-27 05:23:10 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1293, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10717.69
+[2026-08-27 05:23:10 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9485, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10383.19
+[2026-08-27 05:23:11 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1293, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8591.94
+[2026-08-27 05:23:11 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9900, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8353.36
+[2026-08-27 05:23:12 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1708, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10835.80
+[2026-08-27 05:23:12 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9900, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10393.61
+[2026-08-27 05:23:13 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2099.70
+[2026-08-27 05:23:13 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1708, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9919.38
+[2026-08-27 05:23:14 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8525, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8406.32
+[2026-08-27 05:23:14 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1756.57
+[2026-08-27 05:23:15 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 333, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10057.88
+[2026-08-27 05:23:15 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8525, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9966.29
+[2026-08-27 05:23:16 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8909, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8449.44
+[2026-08-27 05:23:16 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 333, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8768.58
+[2026-08-27 05:23:17 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 717, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10363.27
+[2026-08-27 05:23:17 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8909, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9712.12
+[2026-08-27 05:23:17 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9452, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9349.89
+[2026-08-27 05:23:17 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 717, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8950.85
+[2026-08-27 05:23:18 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 1260, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10634.96
+[2026-08-27 05:23:18 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9452, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11269.38
+[2026-08-27 05:23:19 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 1260, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9067.98
+[2026-08-27 05:23:19 PP2 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9830, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8519.49
+[2026-08-27 05:23:19 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1638, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33116.32
+[2026-08-27 05:23:19 PP3 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9830, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32958.23
+[2026-08-27 05:23:19 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1638, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 385069.93
+[2026-08-27 05:23:19 PP2 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1664, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 72026.72
+[2026-08-27 05:23:19 PP3 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1664, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 77715.90
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/logs/manual_p_node2.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/logs/manual_p_node2.log
new file mode 100644
index 0000000..f29b03e
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/logs/manual_p_node2.log
@@ -0,0 +1,1052 @@
+/usr/local/lib/python3.12/dist-packages/torchao/quantization/quant_api.py:1731: SyntaxWarning: invalid escape sequence '\.'
+  """Configuration class for applying different quantization configs to modules or parameters based on their fully qualified names (FQNs).
+/sgl-workspace/sglang/python/sglang/launch_server.py:56: UserWarning: 'python -m sglang.launch_server' is still supported, but 'sglang serve' is the recommended entrypoint.
+  Example: sglang serve --model-path <model> [options]
+  warnings.warn(
+[2026-08-27 04:28:43] Breakable CUDA graph is incompatible with MLA attention; disabling prefill CUDA graph.
+[2026-08-27 04:28:43] Attention backend not specified. Use triton backend by default.
+[2026-08-27 04:28:43] Pipeline parallelism is incompatible with overlap schedule.
+[2026-08-27 04:28:45] server_args=ServerArgs(model_path='/data/hf_models/Kimi-K3', tokenizer_path='/data/hf_models/Kimi-K3', tokenizer_mode='auto', tokenizer_backend='huggingface', tokenizer_worker_num=1, detokenizer_worker_num=1, skip_tokenizer_init=False, load_format='auto', model_loader_extra_config='{}', trust_remote_code=True, context_length=None, is_embedding=False, enable_multimodal=None, revision=None, model_impl='auto', model_config_parser='auto', json_model_override_args='{}', dtype='auto', quantization=None, quantization_param_path=None, kv_cache_dtype='auto', enable_fp32_lm_head=False, modelopt_quant=None, modelopt_checkpoint_restore_path=None, modelopt_checkpoint_save_path=None, modelopt_export_path=None, quantize_and_serve=False, rl_quant_profile=None, enable_tf32_matmul=False, mem_fraction_static=0.88, max_running_requests=None, max_queued_requests=None, max_total_tokens=None, chunked_prefill_size=8192, enable_dynamic_chunking=False, max_prefill_tokens=16384, prefill_max_requests=None, schedule_policy='fcfs', enable_priority_scheduling=False, disable_priority_preemption=False, default_priority_value=None, abort_on_priority_when_disabled=False, schedule_low_priority_values_first=False, priority_scheduling_preemption_threshold=10, retraction_policy='length', schedule_conservativeness=1.0, page_size=64, swa_full_tokens_ratio=0.8, disable_hybrid_swa_memory=False, radix_eviction_policy='lru', prefill_only_disable_kv_cache=False, disable_radix_cache=True, enable_page_major_kv_layout=False, enable_unified_memory=False, disable_chunked_prefix_cache=False, disable_overlap_schedule=True, num_continuous_decode_steps=1, scheduler_recv_interval=1, enable_mixed_chunk=False, nccl_port=None, dist_timeout=3600, dist_init_addr='174.1.60.1:20000', nnodes=4, node_rank=2, tp_size=4, dcp_size=1, pp_size=8, pp_max_micro_batch_size=None, pp_async_batch_depth=0, dp_size=1, load_balance_method='follow_bootstrap_room', attn_cp_size=1, moe_dp_size=1, dcp_comm_backend='ag_rs', dcp_replicate_q_proj=None, dwdp_size=1, enable_prefill_cp=False, cp_strategy=None, enable_dsa_cache_layer_split=False, enable_dsa_prefill_context_parallel=False, dsa_prefill_cp_mode='round-robin-split', enable_prefill_context_parallel=False, prefill_cp_mode='in-seq-split', enable_cp_decode_attn_tp=False, enable_dp_attention=False, enable_dp_attention_local_control_broadcast=False, enable_dp_lm_head=False, enable_attn_tp_input_scattered=False, disable_attn_tp_gather=False, enable_p2p_check=False, device='cuda', base_gpu_id=0, gpu_id_step=1, random_seed=336932344, watchdog_timeout=300, soft_watchdog_timeout=None, sleep_on_idle=False, use_ray=False, custom_sigquit_handler=None, numa_node=None, gc_threshold=None, host='0.0.0.0', port=30000, fastapi_root_path='', smg_grpc_mode=False, grpc_mode=False, grpc_port=None, sidecar=None, sidecar_args=None, skip_server_warmup=False, warmups=None, enable_http2=False, ssl_keyfile=None, ssl_certfile=None, ssl_ca_certs=None, ssl_keyfile_password=None, enable_ssl_refresh=False, api_key=None, admin_api_key=None, served_model_name='kimi-k3', weight_version='default', chat_template=None, hf_chat_template_name=None, completion_template=None, file_storage_path='sglang_storage', enable_cache_report=False, reasoning_parser=None, default_chat_template_kwargs=None, strip_thinking_cache=False, enable_strict_thinking=False, tool_call_parser=None, tool_server=None, sampling_defaults='model', asr_max_buffer_seconds=60, asr_max_concurrent_sessions=32, preferred_sampling_params=None, allow_auto_truncate=False, stream_interval=1, batch_notify_size=16, stream_response_default_include_usage=False, incremental_streaming_output=False, enable_streaming_session=False, enable_session_radix_cache=False, log_level='info', log_level_http=None, log_requests=False, log_requests_level=2, log_requests_format='text', log_requests_target=None, uvicorn_access_log_exclude_prefixes=[], crash_dump_folder=None, show_time_cost=False, enable_metrics=False, smg_http_sidecar_port=None, enable_mfu_metrics=False, enable_metrics_for_all_schedulers=False, load_snapshot_publish_interval=15, tokenizer_metrics_custom_labels_header='x-custom-labels', tokenizer_metrics_allowed_custom_labels=None, extra_metric_labels=None, bucket_time_to_first_token=None, bucket_inter_token_latency=None, bucket_e2e_request_latency=None, prompt_tokens_buckets=None, generation_tokens_buckets=None, gc_warning_threshold_secs=0.0, decode_log_interval=40, enable_request_time_stats_logging=False, kv_events_config=None, enable_forward_pass_metrics=False, forward_pass_metrics_worker_id='', forward_pass_metrics_ipc_name=None, enable_trace=False, trace_modules='request', otlp_traces_endpoint='localhost:4317', export_metrics_to_file=False, export_metrics_to_file_dir=None, stat_loggers=None, constrained_json_whitespace_pattern=None, constrained_json_disable_any_whitespace=False, attention_backend='triton', decode_attention_backend=None, prefill_attention_backend=None, sampling_backend='flashinfer', grammar_backend='xgrammar', radix_cache_backend=None, mm_attention_backend=None, fp8_gemm_runner_backend='auto', fp4_gemm_runner_backend='auto', bf16_gemm_backend='auto', dsa_prefill_backend=None, dsa_decode_backend=None, dsa_paged_mqa_logits_backend='auto', dsa_topk_backend='sgl-kernel', disable_flashinfer_autotune=False, flashinfer_autotune_skip_ops=None, mamba_backend='triton', cuda_graph_config=CudaGraphConfig(decode=PhaseConfig(backend='disabled', max_bs=16, bs=[1, 2, 4, 8, 12, 16], tc_compiler='eager', full_prefill_max_req=None), prefill=PhaseConfig(backend='disabled', max_bs=2048, bs=[4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], tc_compiler='eager', full_prefill_max_req=None)), cuda_graph_backend_decode=None, cuda_graph_backend_prefill=None, cuda_graph_max_bs_decode=16, cuda_graph_max_bs_prefill=None, cuda_graph_bs_decode=None, cuda_graph_bs_prefill=None, cuda_graph_tc_compiler=None, disable_prefill_cuda_graph=False, disable_decode_cuda_graph=False, disable_cuda_graph=False, disable_cuda_graph_padding=False, enable_profile_cuda_graph=False, enable_cudagraph_gc=False, debug_cuda_graph=False, enable_layerwise_nvtx_marker=False, enable_nccl_nvls=False, enable_symm_mem=False, triton_attention_reduce_in_fp32=False, triton_attention_num_kv_splits=8, triton_attention_split_tile_size=None, flashinfer_mla_disable_ragged=False, enable_fused_qk_norm_rope=False, enable_precise_embedding_interpolation=False, enable_fused_moe_sum_all_reduce=False, enable_deepseek_v4_fp4_indexer=False, disable_custom_all_reduce=False, enable_mscclpp=False, enable_torch_symm_mem=False, enable_scattered_sconv=False, pre_warm_nccl=False, enable_quant_communications=False, enable_flashinfer_allreduce_fusion=False, enforce_disable_flashinfer_allreduce_fusion=False, flashinfer_allreduce_fusion_backend=None, enable_aiter_allreduce_fusion=False, enable_torch_compile=False, enable_torch_compile_debug_mode=False, torch_compile_max_bs=32, torchao_config='', speculative_algorithm=None, speculative_draft_model_path=None, speculative_draft_model_revision=None, speculative_draft_load_format=None, speculative_num_steps=None, speculative_eagle_topk=None, speculative_num_draft_tokens=None, speculative_dflash_block_size=None, speculative_dspark_block_size=None, speculative_dspark_sps_table_path=None, speculative_dspark_confidence_sts_path=None, speculative_dspark_align_verify_tokens_to_graph_tier=False, speculative_accept_threshold_single=1.0, speculative_accept_threshold_acc=1.0, speculative_use_rejection_sampling=False, speculative_token_map=None, speculative_attention_mode='prefill', speculative_draft_attention_backend=None, speculative_draft_window_size=None, speculative_moe_runner_backend='flashinfer_mxfp4', speculative_moe_a2a_backend=None, speculative_draft_model_quantization=None, speculative_skip_dp_mlp_sync=False, enable_multi_layer_eagle=False, speculative_adaptive=False, speculative_adaptive_config=None, decoupled_spec_bind_endpoint=None, decoupled_spec_connect_endpoints=None, decoupled_spec_rank=None, decoupled_spec_role='null', spec_trace_dir=None, speculative_ngram_min_bfs_breadth=1, speculative_ngram_max_bfs_breadth=10, speculative_ngram_match_type='BFS', speculative_ngram_max_trie_depth=18, speculative_ngram_capacity=10000000, speculative_ngram_external_corpus_path=None, speculative_ngram_external_sam_budget=0, speculative_ngram_external_corpus_max_tokens=10000000, ep_size=4, moe_a2a_backend='none', moe_runner_backend='flashinfer_mxfp4', flashinfer_mxfp4_moe_precision='default', deepep_mode='auto', fuseep_mode=2, deepep_dispatcher_output_dtype='auto', ep_num_redundant_experts=0, ep_dispatch_algorithm=None, init_expert_location='trivial', enable_eplb=False, eplb_algorithm='auto', eplb_rebalance_num_iterations=1000, eplb_rebalance_layers_per_chunk=None, eplb_min_rebalancing_utilization_threshold=1.0, expert_distribution_recorder_mode=None, expert_distribution_recorder_buffer_size=1000, enable_expert_distribution_metrics=False, deepep_config=None, moe_dense_tp_size=None, elastic_ep_backend=None, enable_elastic_expert_backup=False, mooncake_ib_device=None, enable_waterfill=False, ep_join_mode=None, ep_join_rank_offset=0, elastic_ep_initial_size=None, max_ep_size=None, elastic_ep_scale_timeout=600, elastic_ep_rejoin=False, disable_flashinfer_cutlass_moe_fp4_allgather=False, disable_shared_experts_fusion=False, enforce_shared_experts_fusion=False, max_mamba_cache_size=None, mamba_ssm_dtype=None, mamba_max_states_per_path=-1, enable_mamba_cache_stochastic_rounding=False, mamba_cache_philox_rounds=0, mamba_full_memory_ratio=0.36, mamba_radix_cache_strategy='extra_buffer_lazy', uses_mamba_radix_cache=False, mamba_track_interval=256, enable_int8_mamba_checkpoint=False, int8_mamba_ckpt_size=None, linear_attn_backend='triton', linear_attn_decode_backend=None, linear_attn_prefill_backend=None, linear_attn_verify_backend=None, enable_linear_replayssm=False, linear_replayssm_cache_len=16, enable_linear_replayssm_spec=False, enable_hierarchical_cache=False, hicache_ratio=2.0, hicache_size=0, hicache_write_policy='write_through', hicache_io_backend='kernel', hicache_mem_layout='page_first', hicache_storage_backend=None, hicache_storage_prefetch_policy='timeout', hicache_storage_backend_extra_config=None, enable_hisparse=False, hisparse_config=None, enable_broadcast_mm_inputs_process=False, enable_prefix_mm_cache=False, mm_enable_dp_encoder=False, mm_process_config={}, mm_processor_worker_num=0, mm_io_worker_num=0, limit_mm_data_per_request=None, enable_mm_global_cache=False, disable_fast_image_processor=False, mm_feature_transport='cpu', keep_mm_feature_on_device=False, enable_lora=None, enable_lora_overlap_loading=None, max_lora_rank=None, lora_target_modules=None, lora_paths=None, max_loaded_loras=None, max_loras_per_batch=8, lora_eviction_policy='lru', lora_backend='csgmv', max_lora_chunk_size=16, experts_shared_outer_loras=None, lora_use_virtual_experts=False, lora_strict_loading=False, lora_drain_wait_threshold=0.0, enable_two_batch_overlap=False, enable_single_batch_overlap=False, tbo_token_distribution_threshold=0.48, cpu_offload_gb=0, offload_group_size=-1, offload_num_in_group=1, offload_prefetch_step=1, offload_mode='cpu', enable_lmcache=False, lmcache_config_file=None, enable_flexkv=False, flexkv_config_file=None, kt_weight_path=None, kt_method='AMXINT4', kt_cpuinfer=None, kt_threadpool_count=2, kt_num_gpu_experts=None, kt_max_deferred_experts_per_token=None, dllm_algorithm=None, dllm_algorithm_config=None, dllm_fdfo=True, disaggregation_mode='prefill', disaggregation_transfer_backend='mooncake', disaggregation_bootstrap_port=28800, disaggregation_ib_device='mlx5_0,mlx5_1,mlx5_2,mlx5_3', disaggregation_decode_enable_radix_cache=False, disaggregation_decode_enable_offload_kvcache=False, num_reserved_decode_tokens=512, disaggregation_decode_extra_slots=None, disaggregation_decode_polling_interval=1, optimistic_prefill_attempts=0, encoder_only=False, language_only=False, encoder_transfer_backend='zmq_to_tokenizer', encoder_urls=[], encoder_bootstrap_port=8997, encoder_register_urls=[], enable_adaptive_dispatch_to_encoder=False, enable_pdmux=False, pdmux_config_path=None, sm_group_num=8, custom_weight_loader=[], weight_loader_disable_mmap=False, weight_loader_prefetch_checkpoints=False, weight_loader_prefetch_num_threads=4, weight_loader_drop_cache_after_load=False, remote_instance_weight_loader_seed_instance_ip=None, remote_instance_weight_loader_seed_instance_service_port=None, remote_instance_weight_loader_send_weights_group_ports=None, remote_instance_weight_loader_backend='nccl', remote_instance_weight_loader_start_seed_via_transfer_engine=False, engine_info_bootstrap_port=6789, modelexpress_config=None, download_dir=None, model_checksum=None, delete_ckpt_after_loading=False, decrypted_config_file=None, decrypted_draft_config_file=None, checkpoint_engine_wait_weights_before_ready=False, enable_prefill_delayer=False, prefill_delayer_max_delay_passes=30, prefill_delayer_token_usage_low_watermark=None, prefill_delayer_forward_passes_buckets=None, prefill_delayer_wait_seconds_buckets=None, prefill_delayer_queue_min_ratio=None, prefill_delayer_max_delay_ms=None, min_free_slots_delay=None, enable_deterministic_inference=False, rl_on_policy_target=None, kv_canary='none', kv_canary_real_data='none', kv_canary_sweep_interval=0, enable_dynamic_batch_tokenizer=False, dynamic_batch_tokenizer_batch_size=32, dynamic_batch_tokenizer_batch_timeout=0.002, enable_tokenizer_batch_encode=False, disable_tokenizer_batch_decode=False, debug_tensor_dump_output_folder=None, debug_tensor_dump_layers=None, debug_tensor_dump_input_file=None, enable_memory_saver=False, enable_weights_cpu_backup=False, enable_draft_weights_cpu_backup=False, enable_custom_logit_processor=False, enable_return_hidden_states=False, enable_return_routed_experts=False, enable_return_indexer_topk=False, disable_outlines_disk_cache=False, enable_mis=False, weight_cache_mode='off', weight_cache_socket=None, weight_cache_timeout=1800, forward_hooks=None, msprobe_dump_config=None)
+[2026-08-27 04:28:45] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:45] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:45] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:45] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:45] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:45] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:45] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:45] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:01 PP4 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:01 PP4 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:01 PP5 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:01 PP5 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:01 PP4 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:01 PP4 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:01 PP5 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:01 PP5 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:01 PP4 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:01 PP4 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:01 PP4 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:01 PP4 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:02 PP5 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:02 PP5 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:02 PP5 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:02 PP5 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:03 PP5 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:03 PP5 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:03 PP4 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:03 PP4 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:03 PP5 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:03 PP5 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:03 PP4 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:03 PP4 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:05.195727   240 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:05.195828   240 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:05.195843   240 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.3 port: 12001
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:05.195847   236 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:05.195899   240 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.3:15652
+I0827 04:29:05.195935   236 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:05.195948   236 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.3 port: 12001
+I0827 04:29:05.195989   236 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.3:15032
+I0827 04:29:05.196039   240 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:05.196110   236 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:05.206050   241 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:05.206164   241 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:05.206182   241 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.3 port: 12001
+I0827 04:29:05.206233   241 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.3:16487
+I0827 04:29:05.206414   241 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:05.206894   238 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:05.207010   238 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:05.207029   238 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.3 port: 12001
+I0827 04:29:05.207082   238 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.3:15213
+I0827 04:29:05.207248   238 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:05.213657   238 topology.cpp:127] Device mlx5_0 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:05.215921   237 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:05.216045   237 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:05.216063   237 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.3 port: 12001
+I0827 04:29:05.216125   237 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.3:16450
+I0827 04:29:05.216308   237 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:05.219630   235 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:05.219745   235 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:05.219763   235 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.3 port: 12001
+I0827 04:29:05.219815   235 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.3:15658
+I0827 04:29:05.219810   236 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:05.220000   235 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:05.222787   237 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:05.223301   240 topology.cpp:127] Device mlx5_0 port 1 is available
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:05.223796   239 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+W0827 04:29:05.223798   242 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:05.223948   242 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:05.223948   239 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:05.223969   242 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.3 port: 12001
+I0827 04:29:05.223973   239 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.3 port: 12001
+I0827 04:29:05.224030   242 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.3:16182
+I0827 04:29:05.224030   239 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.3:15581
+I0827 04:29:05.224237   242 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:05.224242   239 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:05.226227   235 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:05.227162   236 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:05.228888   237 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:05.229261   240 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:05.232141   235 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:05.232787   239 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:05.232861   242 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:05.233434   241 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:05.233938   236 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:05.236400   237 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:05.236641   240 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:05.238986   235 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:05.241036   236 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:05.241355   238 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:05.241638   241 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:05.242061   239 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:05.242151   242 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:05.244179   236 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:05.244223   236 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:05.244287   237 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:05.244297   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.244899   240 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:05.245494   235 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:05.249099   237 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:05.249130   237 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:05.249187   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.251125   235 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:05.251137   240 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:05.251168   235 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:05.251168   240 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:05.251241   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.251241   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.251624   239 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:05.251729   242 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:05.251823   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:05.253651   236 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:03
+I0827 04:29:05.253687   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.256857   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:05.257697   239 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:05.258617   237 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:03
+I0827 04:29:05.258652   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.258757   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:05.258822   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:05.260108   239 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:05.260136   239 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:05.260191   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.260488   235 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:03
+I0827 04:29:05.260502   240 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:03
+I0827 04:29:05.260530   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.260533   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.262187   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:05.264094   236 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:03
+I0827 04:29:05.264137   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.266639   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:05.268234   237 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:03
+I0827 04:29:05.268261   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.268549   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:05.269187   241 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:05.269232   238 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:05.270207   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:05.270285   240 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:03
+I0827 04:29:05.270316   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.271827   235 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:03
+I0827 04:29:05.271884   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.275274   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:05.276836   238 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:05.276907   242 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:05.276996   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:05.277058   236 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:03
+I0827 04:29:05.277101   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.278084   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:05.278632   237 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:03
+I0827 04:29:05.278667   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.279789   240 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:03
+I0827 04:29:05.279817   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.282152   238 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:05.282193   238 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:05.282260   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.283128   242 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:05.283200   242 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:05.283263   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.286593   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:05.287469   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:05.288283   239 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:03
+I0827 04:29:05.288322   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.289064   240 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:03
+I0827 04:29:05.289194   240 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:05 PP5 TP1 EP1] Init torch distributed begin.
+I0827 04:29:05.290841   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:05.292457   242 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:03
+I0827 04:29:05.292488   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.292999   241 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:05.296203   241 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:05.296272   241 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:05.296341   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.300206   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:05.301280   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:05.301775   242 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:03
+I0827 04:29:05.301801   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.303004   235 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:03
+I0827 04:29:05.303042   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.303999   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:05.305142   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:05.305728   241 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:03
+I0827 04:29:05.305768   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.305974   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:05.306758   237 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:03
+I0827 04:29:05.306921   237 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:05 PP4 TP2 EP2] Init torch distributed begin.
+I0827 04:29:05.307497   236 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:03
+I0827 04:29:05.307627   236 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:05 PP4 TP1 EP1] Init torch distributed begin.
+I0827 04:29:05.309624   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:05.309773   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:05.311504   238 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:03
+I0827 04:29:05.311542   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.311590   242 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:03
+I0827 04:29:05.311633   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.313076   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:05.314762   239 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:03
+I0827 04:29:05.314813   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.319522   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:05.319566   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:05.321249   242 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:03
+I0827 04:29:05.321398   242 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:05 PP5 TP3 EP3] Init torch distributed begin.
+I0827 04:29:05.326727   238 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:03
+I0827 04:29:05.326759   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.329339   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:05.331007   235 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:03
+I0827 04:29:05.331142   235 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:05 PP4 TP0 EP0] Init torch distributed begin.
+I0827 04:29:05.334666   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:05.336285   238 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:03
+I0827 04:29:05.336313   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.337296   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:05.338828   241 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:03
+I0827 04:29:05.338877   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.344906   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:05.346731   239 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:03
+I0827 04:29:05.346776   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.365299   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:05.366928   238 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:03
+I0827 04:29:05.367062   238 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:05 PP4 TP3 EP3] Init torch distributed begin.
+I0827 04:29:05.368464   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:05.369644   241 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:03
+I0827 04:29:05.369663   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:05.377357   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:05.379019   239 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:03
+I0827 04:29:05.379146   239 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:05 PP5 TP0 EP0] Init torch distributed begin.
+I0827 04:29:05.396831   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:05.398531   241 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:03
+I0827 04:29:05.398662   241 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:05 PP5 TP2 EP2] Init torch distributed begin.
+[2026-08-27 04:29:26 PP5 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:26 PP4 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:27 PP4 TP1 EP1] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP4 TP3 EP3] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP4 TP2 EP2] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP4 TP0 EP0] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP5 TP2 EP2] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP5 TP3 EP3] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP5 TP0 EP0] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP5 TP1 EP1] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP5 TP0 EP0] Init torch distributed ends. elapsed=22.37 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP5 TP3 EP3] Init torch distributed ends. elapsed=22.42 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP5 TP2 EP2] Init torch distributed ends. elapsed=22.35 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP4 TP1 EP1] Init torch distributed ends. elapsed=22.44 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP5 TP1 EP1] Init torch distributed ends. elapsed=22.46 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP4 TP3 EP3] Init torch distributed ends. elapsed=22.38 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP4 TP2 EP2] Init torch distributed ends. elapsed=22.44 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP4 TP0 EP0] Init torch distributed ends. elapsed=22.41 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP5 TP0 EP0] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP4 TP1 EP1] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP4 TP3 EP3] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP4 TP2 EP2] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP5 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP4 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP4 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP5 TP2 EP2] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP4 TP0 EP0] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP4 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP5 TP3 EP3] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP5 TP1 EP1] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP4 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP5 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP5 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP5 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP4 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP5 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP4 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP5 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP5 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP4 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP4 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP5 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP5 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP4 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP4 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP4 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP5 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP5 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP4 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP5 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:30:33 PP4 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:33 PP4 TP2 EP2] Load weight end. elapsed=65.50 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:34 PP4 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:34 PP4 TP3 EP3] Load weight end. elapsed=66.61 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP5 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:36 PP4 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:36 PP4 TP1 EP1] Load weight end. elapsed=68.45 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:36 PP5 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:36 PP5 TP0 EP0] Load weight end. elapsed=69.00 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:37 PP5 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:37 PP5 TP3 EP3] Load weight end. elapsed=69.63 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:37 PP5 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:37 PP5 TP1 EP1] Load weight end. elapsed=69.72 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:38 PP5 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:38 PP5 TP2 EP2] Load weight end. elapsed=70.08 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:38 PP5 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:38 PP5 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:38 PP5 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:38 PP5 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:38 PP5 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:38 PP5 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:38 PP5 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:38 PP5 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:46 PP4 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:47 PP4 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:47 PP4 TP0 EP0] Load weight end. elapsed=79.31 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:47 PP4 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:47 PP4 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:47 PP4 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:47 PP4 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:47 PP4 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:47 PP4 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:47 PP4 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:47 PP4 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:54 PP5 TP3 EP3] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP5 TP1 EP1] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP5 TP2 EP2] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP5 TP0 EP0] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP4 TP3 EP3] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP4 TP2 EP2] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP4 TP1 EP1] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP4 TP0 EP0] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP5 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP5 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP4 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP4 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP4 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP5 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP4 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP5 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP5 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP5 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP4 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP5 TP3 EP3] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP4 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP4 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP4 TP1 EP1] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP5 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP5 TP2 EP2] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP4 TP3 EP3] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP4 TP0 EP0] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP5 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP5 TP1 EP1] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP5 TP0 EP0] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP4 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP4 TP2 EP2] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP4 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:30:54 PP4 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:30:54 PP4 TP0 EP0] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp0_pp4_dp0.json
+[2026-08-27 04:30:54 PP4 TP1 EP1] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp1_pp4_dp0.json
+[2026-08-27 04:30:54 PP5 TP2 EP2] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp2_pp5_dp0.json
+[2026-08-27 04:30:54 PP5 TP3 EP3] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp3_pp5_dp0.json
+[2026-08-27 04:30:54 PP5 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:30:54 PP5 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:30:54 PP5 TP0 EP0] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp0_pp5_dp0.json
+[2026-08-27 04:30:54 PP4 TP2 EP2] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp2_pp4_dp0.json
+[2026-08-27 04:30:54 PP4 TP3 EP3] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp3_pp4_dp0.json
+[2026-08-27 04:30:54 PP5 TP1 EP1] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp1_pp5_dp0.json
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO[TensorRT-LLM][INFO] Set logger level to INFO
+
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.37profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.12profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.87profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.32profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.91profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.77profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.94profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.88profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.20profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.87profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  7.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 10.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.46profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.67profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.91profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.29profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.49profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  7.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.30profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.94profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.47profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.12profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.06profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.35profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  9.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:01<00:02,  2.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.89profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.58profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.32profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.70profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.20profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.27profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  6.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  7.03profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.56profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.73profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.88profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.37profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:05,  1.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:01<00:02,  2.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.05profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.19profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.77profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.18profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.25profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  3.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.18profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.20profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.37profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  9.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.27profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  8.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.43profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.89profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.35profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.96profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.90profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.30profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.72profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.18profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.33profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.51profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  2.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.03profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.68profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.16profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.01profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.24profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 10.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:02,  2.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.06profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.83profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.35profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.87profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.19profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  3.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.06profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.86profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 13.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 13.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.13profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.39profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:01,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:01,  3.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.47profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  6.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.47profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.57profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.22profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.17profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.50profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.71profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.19profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.55profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.90profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:05,  1.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:01<00:06,  1.00s/profile]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.33profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.19profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.03profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.76profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.28profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.32profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.15profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.79profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.31profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.18profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 11.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:01,  3.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.06profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:01,  3.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  6.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  3.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.05profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.60profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.50profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.92profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.00profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.29profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 14.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 11.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.09profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.43profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.11profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.91profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  7.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:02,  2.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.18profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  6.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.78profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.75profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.89profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.64profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.05profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.47profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.47profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.65profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:05,  1.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.41profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.85profile/s]
+[2026-08-27 04:31:28 PP5 TP1 EP1] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.21profile/s]
+[2026-08-27 04:31:28 PP5 TP0 EP0] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.74profile/s]
+[2026-08-27 04:31:28 PP5 TP2 EP2] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.77profile/s]
+[2026-08-27 04:31:28 PP5 TP3 EP3] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.19profile/s][2026-08-27 04:31:28 PP5 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:28 PP5 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:28 PP5 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:28 PP5 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:28 PP5 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:28 PP5 TP0 EP0] max_total_num_tokens=3801408, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=54, context_len=1048576, available_gpu_mem=16.30 GB
+[2026-08-27 04:31:28 PP5 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:28 PP5 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:28 PP5 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.87profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.90profile/s]
+[2026-08-27 04:31:29 PP4 TP3 EP3] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.87profile/s]
+[2026-08-27 04:31:29 PP4 TP2 EP2] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.93profile/s]
+[2026-08-27 04:31:29 PP4 TP0 EP0] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.43profile/s]
+[2026-08-27 04:31:29 PP4 TP1 EP1] FlashInfer autotune completed.
+[2026-08-27 04:31:29 PP4 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:29 PP4 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:29 PP4 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:29 PP4 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:29 PP4 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:29 PP4 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:29 PP4 TP0 EP0] max_total_num_tokens=3801408, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=54, context_len=1048576, available_gpu_mem=16.30 GB
+[2026-08-27 04:31:29 PP4 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:29 PP4 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+W0827 04:31:30.513926  2527 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50a93b80, len: 6912: Operation not permitted [1]
+W0827 04:31:30.514082  2528 memory_location.cpp:73] Failed to get NUMA node, addr: 0x507df700, len: 6912: Operation not permitted [1]
+W0827 04:31:30.514106  2529 memory_location.cpp:73] Failed to get NUMA node, addr: 0x508aa940, len: 6912: Operation not permitted [1]
+W0827 04:31:30.514214  2530 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50f67bc0, len: 6912: Operation not permitted [1]
+W0827 04:31:30.514272  2531 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50a2f680, len: 55296: Operation not permitted [1]
+W0827 04:31:30.514393  2532 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50afecc0, len: 55296: Operation not permitted [1]
+W0827 04:31:30.514506  2533 memory_location.cpp:73] Failed to get NUMA node, addr: 0x502be580, len: 6912: Operation not permitted [1]
+W0827 04:31:30.514607  2534 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5054d3c0, len: 13824: Operation not permitted [1]
+W0827 04:31:30.514652  2535 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50ad4580, len: 6912: Operation not permitted [1]
+W0827 04:31:30.514743  2536 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50eb9880, len: 6912: Operation not permitted [1]
+W0827 04:31:31.680995  2556 memory_location.cpp:73] Failed to get NUMA node, addr: 0x501c3300, len: 6912: Operation not permitted [1]
+W0827 04:31:31.680985  2557 memory_location.cpp:73] Failed to get NUMA node, addr: 0x504ecec0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.680985  2555 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd4bac0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681169  2558 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50706900, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681272  2559 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5006d700, len: 55296: Operation not permitted [1]
+W0827 04:31:31.681360  2560 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50041cc0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.681468  2561 memory_location.cpp:73] Failed to get NUMA node, addr: 0x506f5a40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681562  2562 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fdfdf40, len: 13824: Operation not permitted [1]
+W0827 04:31:31.681607  2563 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fd16080, len: 6912: Operation not permitted [1]
+W0827 04:31:31.681685  2564 memory_location.cpp:73] Failed to get NUMA node, addr: 0x4fcfa300, len: 6912: Operation not permitted [1]
+W0827 04:31:31.733497  2583 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51554040, len: 6912: Operation not permitted [1]
+W0827 04:31:31.733552  2584 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5110a8c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.733618  2585 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51ca8e80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.733795  2586 memory_location.cpp:73] Failed to get NUMA node, addr: 0x52017680, len: 6912: Operation not permitted [1]
+W0827 04:31:31.733909  2587 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5174af80, len: 55296: Operation not permitted [1]
+W0827 04:31:31.734004  2588 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51770440, len: 55296: Operation not permitted [1]
+W0827 04:31:31.734061  2589 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51118000, len: 6912: Operation not permitted [1]
+W0827 04:31:31.734179  2590 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51564f40, len: 13824: Operation not permitted [1]
+W0827 04:31:31.734277  2591 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51f69480, len: 6912: Operation not permitted [1]
+W0827 04:31:31.734400  2592 memory_location.cpp:73] Failed to get NUMA node, addr: 0x517c2940, len: 6912: Operation not permitted [1]
+W0827 04:31:31.741586  2611 memory_location.cpp:73] Failed to get NUMA node, addr: 0x519520c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.741609  2612 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50ff8c40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.741679  2613 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51a2ad80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.741792  2614 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51b5b300, len: 6912: Operation not permitted [1]
+W0827 04:31:31.741938  2615 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50fe5b80, len: 55296: Operation not permitted [1]
+W0827 04:31:31.742043  2616 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51625140, len: 55296: Operation not permitted [1]
+W0827 04:31:31.742168  2617 memory_location.cpp:73] Failed to get NUMA node, addr: 0x516bf640, len: 6912: Operation not permitted [1]
+W0827 04:31:31.742278  2618 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51b73e80, len: 13824: Operation not permitted [1]
+W0827 04:31:31.742414  2619 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51ef60c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.742429  2620 memory_location.cpp:73] Failed to get NUMA node, addr: 0x513e95c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.900663  2641 memory_location.cpp:73] Failed to get NUMA node, addr: 0x517e8c80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.900679  2640 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5170b840, len: 6912: Operation not permitted [1]
+W0827 04:31:31.900738  2642 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5131d0c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.900669  2639 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51372400, len: 6912: Operation not permitted [1]
+W0827 04:31:31.900923  2643 memory_location.cpp:73] Failed to get NUMA node, addr: 0x512b7e00, len: 55296: Operation not permitted [1]
+W0827 04:31:31.900976  2644 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51380200, len: 55296: Operation not permitted [1]
+W0827 04:31:31.901000  2646 memory_location.cpp:73] Failed to get NUMA node, addr: 0x513151c0, len: 13824: Operation not permitted [1]
+W0827 04:31:31.901011  2645 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51194f40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.901038  2647 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51520740, len: 6912: Operation not permitted [1]
+W0827 04:31:31.901073  2648 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51724980, len: 6912: Operation not permitted [1]
+W0827 04:31:31.931138  2668 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51061000, len: 6912: Operation not permitted [1]
+W0827 04:31:31.931178  2669 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50d74c00, len: 6912: Operation not permitted [1]
+W0827 04:31:31.931098  2667 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50b2fb40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.931272  2670 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50de9000, len: 6912: Operation not permitted [1]
+W0827 04:31:31.931349  2671 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5097dc80, len: 55296: Operation not permitted [1]
+W0827 04:31:31.931519  2672 memory_location.cpp:73] Failed to get NUMA node, addr: 0x515aa000, len: 55296: Operation not permitted [1]
+W0827 04:31:31.931557  2675 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51269700, len: 6912: Operation not permitted [1]
+W0827 04:31:31.931556  2673 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5097bd00, len: 6912: Operation not permitted [1]
+W0827 04:31:31.931576  2674 memory_location.cpp:73] Failed to get NUMA node, addr: 0x514ee800, len: 13824: Operation not permitted [1]
+W0827 04:31:31.931759  2676 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50deac80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.983201  2696 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50819380, len: 6912: Operation not permitted [1]
+W0827 04:31:31.983201  2695 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50600a80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.983378  2697 memory_location.cpp:73] Failed to get NUMA node, addr: 0x506ddf40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.983438  2698 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50716800, len: 6912: Operation not permitted [1]
+W0827 04:31:31.983490  2699 memory_location.cpp:73] Failed to get NUMA node, addr: 0x501575c0, len: 55296: Operation not permitted [1]
+W0827 04:31:31.983624  2700 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50863340, len: 55296: Operation not permitted [1]
+W0827 04:31:31.983737  2701 memory_location.cpp:73] Failed to get NUMA node, addr: 0x508d7a40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.983930  2702 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50668c40, len: 13824: Operation not permitted [1]
+W0827 04:31:31.984035  2703 memory_location.cpp:73] Failed to get NUMA node, addr: 0x506841c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.984107  2704 memory_location.cpp:73] Failed to get NUMA node, addr: 0x508f16c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.001853  2725 memory_location.cpp:73] Failed to get NUMA node, addr: 0x507c8d40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.001852  2724 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50b39240, len: 6912: Operation not permitted [1]
+W0827 04:31:32.002048  2727 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50a8ecc0, len: 55296: Operation not permitted [1]
+W0827 04:31:32.001888  2726 memory_location.cpp:73] Failed to get NUMA node, addr: 0x512a8140, len: 6912: Operation not permitted [1]
+W0827 04:31:32.002044  2728 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50a68880, len: 55296: Operation not permitted [1]
+W0827 04:31:32.002257  2729 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5087b880, len: 6912: Operation not permitted [1]
+W0827 04:31:32.002264  2731 memory_location.cpp:73] Failed to get NUMA node, addr: 0x50fc6980, len: 6912: Operation not permitted [1]
+W0827 04:31:32.002266  2730 memory_location.cpp:73] Failed to get NUMA node, addr: 0x51331100, len: 13824: Operation not permitted [1]
+W0827 04:31:32.002470  2732 memory_location.cpp:73] Failed to get NUMA node, addr: 0x513384c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.002780  2723 memory_location.cpp:73] Failed to get NUMA node, addr: 0x512c6940, len: 6912: Operation not permitted [1]
+[2026-08-27 04:31:32] Dummy health check server started in background thread at 0.0.0.0:30000
+[rank16]:[W827 04:33:25.972222579 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank17]:[W827 04:33:25.980800100 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank19]:[W827 04:33:25.981413562 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank18]:[W827 04:33:26.644294178 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank16]:[W827 04:34:10.467297472 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank18]:[W827 04:34:10.467837454 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank17]:[W827 04:34:10.468151853 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank19]:[W827 04:34:10.468427857 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank22]:[W827 04:34:10.495396864 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank23]:[W827 04:34:10.495456053 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank21]:[W827 04:34:10.495795969 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank20]:[W827 04:34:10.495814123 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank22]:[W827 04:34:16.703081720 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank20]:[W827 04:34:16.703915664 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank21]:[W827 04:34:16.704033167 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank23]:[W827 04:34:16.704314444 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[2026-08-27 04:35:10 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.29
+[2026-08-27 04:35:10 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.29
+[2026-08-27 04:35:29 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.39
+[2026-08-27 04:35:29 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.39
+[2026-08-27 04:35:31 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25.89
+[2026-08-27 04:35:31 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25.91
+[2026-08-27 04:43:23 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.14
+[2026-08-27 04:43:23 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.14
+[2026-08-27 04:48:13 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.22
+[2026-08-27 04:48:14 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.22
+[2026-08-27 04:48:14 PP4 TP1 EP1] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP4 TP3 EP3] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP4 TP2 EP2] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP4 TP0 EP0] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP5 TP1 EP1] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP5 TP2 EP2] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP5 TP0 EP0] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP5 TP3 EP3] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:57:14 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.12
+[2026-08-27 04:57:14 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.12
+[2026-08-27 05:04:21 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1088, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2.55
+[2026-08-27 05:04:21 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1088, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2.55
+[2026-08-27 05:04:41 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.19
+[2026-08-27 05:04:41 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.19
+[2026-08-27 05:05:39 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1.10
+[2026-08-27 05:05:39 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1.10
+[2026-08-27 05:07:33 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 72.17
+[2026-08-27 05:07:33 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 72.17
+[2026-08-27 05:07:33 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 437051.18
+[2026-08-27 05:07:33 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 375850.16
+[2026-08-27 05:07:33 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25700.77
+[2026-08-27 05:07:33 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33225.98
+[2026-08-27 05:07:40 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8673, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1083.39
+[2026-08-27 05:07:40 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8673, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1083.24
+[2026-08-27 05:07:40 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 481, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 429004.90
+[2026-08-27 05:07:40 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 481, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 414364.03
+[2026-08-27 05:07:40 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 30845.47
+[2026-08-27 05:07:40 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 34555.92
+[2026-08-27 05:07:48 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8656, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1082.66
+[2026-08-27 05:07:48 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8656, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1082.26
+[2026-08-27 05:07:48 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 464, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 397827.41
+[2026-08-27 05:07:48 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 464, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 437968.14
+[2026-08-27 05:07:48 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32533.07
+[2026-08-27 05:07:48 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 31758.27
+[2026-08-27 05:07:55 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8663, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1087.59
+[2026-08-27 05:07:55 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8663, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1087.59
+[2026-08-27 05:07:55 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 471, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 394787.42
+[2026-08-27 05:07:55 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 471, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 405862.61
+[2026-08-27 05:07:55 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32866.57
+[2026-08-27 05:07:55 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32717.90
+[2026-08-27 05:08:03 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8659, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1082.97
+[2026-08-27 05:08:03 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8659, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1083.02
+[2026-08-27 05:08:03 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 467, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 419865.10
+[2026-08-27 05:08:03 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 467, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 410377.15
+[2026-08-27 05:08:03 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33296.73
+[2026-08-27 05:08:03 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 28803.86
+[2026-08-27 05:08:11 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8692, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1085.66
+[2026-08-27 05:08:11 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8692, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1085.97
+[2026-08-27 05:08:11 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 500, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 394417.45
+[2026-08-27 05:08:11 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 500, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 404424.58
+[2026-08-27 05:08:11 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33136.89
+[2026-08-27 05:08:11 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 31089.88
+[2026-08-27 05:08:18 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8557, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1090.03
+[2026-08-27 05:08:18 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8557, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1089.95
+[2026-08-27 05:08:18 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 365, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 409432.91
+[2026-08-27 05:08:18 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 365, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 521781.11
+[2026-08-27 05:08:18 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 30469.13
+[2026-08-27 05:08:18 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 22116.29
+[2026-08-27 05:08:26 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8683, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1083.95
+[2026-08-27 05:08:26 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8683, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1084.25
+[2026-08-27 05:08:26 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 491, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 406971.55
+[2026-08-27 05:08:26 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 491, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 413225.93
+[2026-08-27 05:08:26 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33768.16
+[2026-08-27 05:08:26 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32002.71
+[2026-08-27 05:11:45 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.32
+[2026-08-27 05:11:45 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.32
+[2026-08-27 05:14:42 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1088, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6.16
+[2026-08-27 05:14:42 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1088, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6.16
+[2026-08-27 05:17:47 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 44.22
+[2026-08-27 05:17:47 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 44.22
+[2026-08-27 05:17:47 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 414289.62
+[2026-08-27 05:17:47 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 418259.05
+[2026-08-27 05:17:47 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 31189.29
+[2026-08-27 05:17:47 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 34799.42
+[2026-08-27 05:17:55 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8673, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1086.30
+[2026-08-27 05:17:55 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8673, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1086.07
+[2026-08-27 05:17:55 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 481, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 412548.65
+[2026-08-27 05:17:55 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 481, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 453654.88
+[2026-08-27 05:17:55 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33203.44
+[2026-08-27 05:17:55 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32631.53
+[2026-08-27 05:18:02 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8656, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1088.44
+[2026-08-27 05:18:02 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8656, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1088.37
+[2026-08-27 05:18:02 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 464, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 420591.31
+[2026-08-27 05:18:02 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 464, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 421477.03
+[2026-08-27 05:18:02 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33878.95
+[2026-08-27 05:18:02 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 35064.03
+[2026-08-27 05:18:10 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8663, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1088.01
+[2026-08-27 05:18:10 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8663, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1087.84
+[2026-08-27 05:18:10 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 471, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 429396.76
+[2026-08-27 05:18:10 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 471, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 403808.84
+[2026-08-27 05:18:10 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 30291.78
+[2026-08-27 05:18:10 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 35867.18
+[2026-08-27 05:18:18 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8659, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1086.23
+[2026-08-27 05:18:18 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8659, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1086.07
+[2026-08-27 05:18:18 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 467, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 416829.87
+[2026-08-27 05:18:18 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 467, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 367035.35
+[2026-08-27 05:18:18 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 27946.41
+[2026-08-27 05:18:18 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32311.83
+[2026-08-27 05:18:25 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8692, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1087.22
+[2026-08-27 05:18:25 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8692, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1087.19
+[2026-08-27 05:18:25 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 500, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 454585.15
+[2026-08-27 05:18:25 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 500, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 478466.41
+[2026-08-27 05:18:25 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 34596.06
+[2026-08-27 05:18:25 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 36601.41
+[2026-08-27 05:18:33 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8557, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1090.87
+[2026-08-27 05:18:33 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8557, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1090.67
+[2026-08-27 05:18:33 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 365, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 423632.77
+[2026-08-27 05:18:33 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 365, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 427798.84
+[2026-08-27 05:18:33 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 24938.13
+[2026-08-27 05:18:33 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 26814.84
+[2026-08-27 05:18:40 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8683, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1087.87
+[2026-08-27 05:18:40 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8683, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1087.73
+[2026-08-27 05:18:40 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 491, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 435790.11
+[2026-08-27 05:18:40 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 491, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 437235.07
+[2026-08-27 05:18:40 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33178.64
+[2026-08-27 05:18:40 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 30137.19
+[2026-08-27 05:21:46 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 2, #pending-token: 8652, #bootstrap-req: 2, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 44.04
+[2026-08-27 05:21:47 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 3, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9796.50
+[2026-08-27 05:21:47 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 2, #pending-token: 8652, #bootstrap-req: 2, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 43.85
+[2026-08-27 05:21:48 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 3, #pending-token: 9185, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8786.35
+[2026-08-27 05:21:48 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 3, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8753.58
+[2026-08-27 05:21:49 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 2, #pending-token: 993, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9748.57
+[2026-08-27 05:21:49 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 3, #pending-token: 9185, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9863.29
+[2026-08-27 05:21:50 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 2, #pending-token: 9680, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8707.37
+[2026-08-27 05:21:50 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 2, #pending-token: 993, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8549.77
+[2026-08-27 05:21:51 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1488, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9553.86
+[2026-08-27 05:21:51 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 2, #pending-token: 9680, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9316.43
+[2026-08-27 05:21:52 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1488, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9810.72
+[2026-08-27 05:21:52 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 10199, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8877.35
+[2026-08-27 05:21:52 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 18858, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10683.27
+[2026-08-27 05:21:52 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 10199, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9914.60
+[2026-08-27 05:21:53 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 27591, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9125.55
+[2026-08-27 05:21:53 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 18858, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9121.64
+[2026-08-27 05:21:54 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 53023, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9612.99
+[2026-08-27 05:21:54 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 27591, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9653.85
+[2026-08-27 05:21:55 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 44876, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9352.82
+[2026-08-27 05:21:55 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 53023, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9371.66
+[2026-08-27 05:21:56 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 36684, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9652.74
+[2026-08-27 05:21:56 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 44876, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9597.62
+[2026-08-27 05:21:57 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 28504, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10734.61
+[2026-08-27 05:21:57 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 36684, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9728.96
+[2026-08-27 05:21:57 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 20312, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 15406.33
+[2026-08-27 05:21:57 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 28504, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 18481.89
+[2026-08-27 05:21:57 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 20312, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 294820.71
+[2026-08-27 05:21:57 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 12139, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 258255.44
+[2026-08-27 05:21:57 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 3947, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 296280.34
+[2026-08-27 05:21:57 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 12139, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 273258.91
+[2026-08-27 05:21:57 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 3968, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 41900.16
+[2026-08-27 05:21:57 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 3947, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 83041.18
+[2026-08-27 05:21:58 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 3968, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7314.63
+[2026-08-27 05:22:06 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 0, #pending-token: 8608, #bootstrap-req: 4, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 922.15
+[2026-08-27 05:22:07 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 416, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9857.85
+[2026-08-27 05:22:07 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 0, #pending-token: 8608, #bootstrap-req: 4, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 892.94
+[2026-08-27 05:22:08 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 9149, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8818.94
+[2026-08-27 05:22:08 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 416, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8761.69
+[2026-08-27 05:22:09 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 957, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10110.54
+[2026-08-27 05:22:09 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 9149, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10200.43
+[2026-08-27 05:22:10 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 9638, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8782.22
+[2026-08-27 05:22:10 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 957, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8750.32
+[2026-08-27 05:22:10 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1446, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9843.04
+[2026-08-27 05:22:10 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 9638, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9909.13
+[2026-08-27 05:22:11 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1446, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9535.57
+[2026-08-27 05:22:11 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 10033, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9304.25
+[2026-08-27 05:22:12 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1841, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8648.37
+[2026-08-27 05:22:12 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 10033, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8117.84
+[2026-08-27 05:22:13 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1841, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10198.16
+[2026-08-27 05:22:13 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1856, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2161.72
+[2026-08-27 05:22:14 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8540, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8634.86
+[2026-08-27 05:22:14 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1856, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1928.73
+[2026-08-27 05:22:15 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8540, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9967.52
+[2026-08-27 05:22:15 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 17066, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9942.20
+[2026-08-27 05:22:16 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8910, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8633.09
+[2026-08-27 05:22:16 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 17066, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8614.10
+[2026-08-27 05:22:17 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 17466, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10316.02
+[2026-08-27 05:22:17 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8910, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10031.51
+[2026-08-27 05:22:18 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9324, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9877.92
+[2026-08-27 05:22:18 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 17466, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9424.86
+[2026-08-27 05:22:18 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 17751, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 36929.95
+[2026-08-27 05:22:18 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9324, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 50710.16
+[2026-08-27 05:22:18 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 17751, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 310670.60
+[2026-08-27 05:22:18 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9579, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 230464.10
+[2026-08-27 05:22:18 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1387, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 75748.54
+[2026-08-27 05:22:18 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9579, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 70984.49
+[2026-08-27 05:22:18 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1408, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2600.96
+[2026-08-27 05:22:18 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1387, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 15177.10
+[2026-08-27 05:22:19 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1408, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1879.05
+[2026-08-27 05:22:27 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8547, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 982.50
+[2026-08-27 05:22:28 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 355, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9870.85
+[2026-08-27 05:22:28 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8547, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 973.15
+[2026-08-27 05:22:29 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8932, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8633.06
+[2026-08-27 05:22:29 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 355, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8626.74
+[2026-08-27 05:22:29 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 740, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10223.25
+[2026-08-27 05:22:29 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8932, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10140.49
+[2026-08-27 05:22:30 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 740, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8655.21
+[2026-08-27 05:22:30 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9259, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8412.98
+[2026-08-27 05:22:31 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1067, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10205.64
+[2026-08-27 05:22:31 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9259, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9957.18
+[2026-08-27 05:22:32 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1067, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9648.18
+[2026-08-27 05:22:32 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9745, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9472.33
+[2026-08-27 05:22:33 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1553, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8576.88
+[2026-08-27 05:22:33 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9745, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8156.56
+[2026-08-27 05:22:34 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1600, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1870.62
+[2026-08-27 05:22:34 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1553, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9817.59
+[2026-08-27 05:22:35 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8602, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8886.75
+[2026-08-27 05:22:35 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1600, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1744.17
+[2026-08-27 05:22:36 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 410, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9856.51
+[2026-08-27 05:22:36 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8602, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9924.95
+[2026-08-27 05:22:36 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8696, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8614.79
+[2026-08-27 05:22:36 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 410, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8588.26
+[2026-08-27 05:22:37 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 504, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10212.49
+[2026-08-27 05:22:37 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8696, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10054.15
+[2026-08-27 05:22:38 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9164, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9916.25
+[2026-08-27 05:22:38 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 504, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9469.58
+[2026-08-27 05:22:38 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 972, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 45807.47
+[2026-08-27 05:22:38 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9164, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 65745.12
+[2026-08-27 05:22:38 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 972, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 271786.52
+[2026-08-27 05:22:38 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9322, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 228784.17
+[2026-08-27 05:22:38 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1130, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 76741.41
+[2026-08-27 05:22:38 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9322, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 73462.62
+[2026-08-27 05:22:39 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1152, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2130.27
+[2026-08-27 05:22:39 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1130, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 15175.73
+[2026-08-27 05:22:40 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1152, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1541.83
+[2026-08-27 05:22:47 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8602, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 986.08
+[2026-08-27 05:22:48 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 410, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9910.97
+[2026-08-27 05:22:48 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8602, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 976.33
+[2026-08-27 05:22:49 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 9105, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8736.80
+[2026-08-27 05:22:49 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 410, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8755.48
+[2026-08-27 05:22:50 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 913, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10163.94
+[2026-08-27 05:22:50 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 9105, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10126.04
+[2026-08-27 05:22:51 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9375, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8789.11
+[2026-08-27 05:22:51 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 913, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8802.92
+[2026-08-27 05:22:52 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1183, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9769.42
+[2026-08-27 05:22:52 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9375, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9789.67
+[2026-08-27 05:22:52 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9868, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9680.86
+[2026-08-27 05:22:52 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1183, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9551.84
+[2026-08-27 05:22:53 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1676, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8479.61
+[2026-08-27 05:22:53 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9868, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8580.43
+[2026-08-27 05:22:54 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1676, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9905.15
+[2026-08-27 05:22:54 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2025.45
+[2026-08-27 05:22:55 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8467, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8799.98
+[2026-08-27 05:22:55 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1790.28
+[2026-08-27 05:22:56 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8467, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9998.59
+[2026-08-27 05:22:56 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 275, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9869.57
+[2026-08-27 05:22:57 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8988, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8731.62
+[2026-08-27 05:22:57 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 275, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8714.49
+[2026-08-27 05:22:58 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 796, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10327.91
+[2026-08-27 05:22:58 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8988, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10150.84
+[2026-08-27 05:22:59 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9363, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9995.05
+[2026-08-27 05:22:59 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 796, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9408.40
+[2026-08-27 05:22:59 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 1171, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 30814.55
+[2026-08-27 05:22:59 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9363, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 40670.98
+[2026-08-27 05:22:59 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 1171, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 312400.93
+[2026-08-27 05:22:59 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9880, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 273542.20
+[2026-08-27 05:22:59 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1688, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 81674.04
+[2026-08-27 05:22:59 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9880, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 78852.53
+[2026-08-27 05:23:00 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3180.06
+[2026-08-27 05:23:00 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1688, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 15111.86
+[2026-08-27 05:23:00 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2313.52
+[2026-08-27 05:23:08 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8601, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 987.24
+[2026-08-27 05:23:09 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 409, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9965.41
+[2026-08-27 05:23:09 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8601, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 978.45
+[2026-08-27 05:23:10 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 9104, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8686.72
+[2026-08-27 05:23:10 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 409, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8681.63
+[2026-08-27 05:23:10 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 912, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10219.36
+[2026-08-27 05:23:10 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 9104, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10129.55
+[2026-08-27 05:23:11 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9485, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8726.43
+[2026-08-27 05:23:11 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 912, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8797.95
+[2026-08-27 05:23:12 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1293, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9968.13
+[2026-08-27 05:23:12 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9485, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9970.37
+[2026-08-27 05:23:13 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1293, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9561.18
+[2026-08-27 05:23:13 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9900, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9322.88
+[2026-08-27 05:23:14 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1708, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8660.45
+[2026-08-27 05:23:14 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9900, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8445.20
+[2026-08-27 05:23:15 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1708, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9863.39
+[2026-08-27 05:23:15 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2040.28
+[2026-08-27 05:23:16 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8525, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8723.80
+[2026-08-27 05:23:16 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1809.05
+[2026-08-27 05:23:17 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8525, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9978.38
+[2026-08-27 05:23:17 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 333, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9952.63
+[2026-08-27 05:23:18 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 333, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8701.62
+[2026-08-27 05:23:18 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8909, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8696.74
+[2026-08-27 05:23:18 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 717, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10354.85
+[2026-08-27 05:23:18 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8909, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10184.48
+[2026-08-27 05:23:19 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9452, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10071.93
+[2026-08-27 05:23:19 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 717, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9491.81
+[2026-08-27 05:23:19 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 1260, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33416.21
+[2026-08-27 05:23:19 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9452, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 44390.95
+[2026-08-27 05:23:19 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 1260, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 436736.09
+[2026-08-27 05:23:19 PP4 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9830, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 336391.66
+[2026-08-27 05:23:19 PP5 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9830, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 339855.11
+[2026-08-27 05:23:19 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1638, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 394477.69
+[2026-08-27 05:23:19 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1638, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 408438.83
+[2026-08-27 05:23:19 PP4 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1664, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 78605.25
+[2026-08-27 05:23:19 PP5 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1664, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 99592.51
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/logs/manual_p_node3.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/logs/manual_p_node3.log
new file mode 100644
index 0000000..ece91b3
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/logs/manual_p_node3.log
@@ -0,0 +1,1055 @@
+/usr/local/lib/python3.12/dist-packages/torchao/quantization/quant_api.py:1731: SyntaxWarning: invalid escape sequence '\.'
+  """Configuration class for applying different quantization configs to modules or parameters based on their fully qualified names (FQNs).
+/sgl-workspace/sglang/python/sglang/launch_server.py:56: UserWarning: 'python -m sglang.launch_server' is still supported, but 'sglang serve' is the recommended entrypoint.
+  Example: sglang serve --model-path <model> [options]
+  warnings.warn(
+[2026-08-27 04:28:46] Breakable CUDA graph is incompatible with MLA attention; disabling prefill CUDA graph.
+[2026-08-27 04:28:47] Attention backend not specified. Use triton backend by default.
+[2026-08-27 04:28:47] Pipeline parallelism is incompatible with overlap schedule.
+[2026-08-27 04:28:49] server_args=ServerArgs(model_path='/data/hf_models/Kimi-K3', tokenizer_path='/data/hf_models/Kimi-K3', tokenizer_mode='auto', tokenizer_backend='huggingface', tokenizer_worker_num=1, detokenizer_worker_num=1, skip_tokenizer_init=False, load_format='auto', model_loader_extra_config='{}', trust_remote_code=True, context_length=None, is_embedding=False, enable_multimodal=None, revision=None, model_impl='auto', model_config_parser='auto', json_model_override_args='{}', dtype='auto', quantization=None, quantization_param_path=None, kv_cache_dtype='auto', enable_fp32_lm_head=False, modelopt_quant=None, modelopt_checkpoint_restore_path=None, modelopt_checkpoint_save_path=None, modelopt_export_path=None, quantize_and_serve=False, rl_quant_profile=None, enable_tf32_matmul=False, mem_fraction_static=0.88, max_running_requests=None, max_queued_requests=None, max_total_tokens=None, chunked_prefill_size=8192, enable_dynamic_chunking=False, max_prefill_tokens=16384, prefill_max_requests=None, schedule_policy='fcfs', enable_priority_scheduling=False, disable_priority_preemption=False, default_priority_value=None, abort_on_priority_when_disabled=False, schedule_low_priority_values_first=False, priority_scheduling_preemption_threshold=10, retraction_policy='length', schedule_conservativeness=1.0, page_size=64, swa_full_tokens_ratio=0.8, disable_hybrid_swa_memory=False, radix_eviction_policy='lru', prefill_only_disable_kv_cache=False, disable_radix_cache=True, enable_page_major_kv_layout=False, enable_unified_memory=False, disable_chunked_prefix_cache=False, disable_overlap_schedule=True, num_continuous_decode_steps=1, scheduler_recv_interval=1, enable_mixed_chunk=False, nccl_port=None, dist_timeout=3600, dist_init_addr='174.1.60.1:20000', nnodes=4, node_rank=3, tp_size=4, dcp_size=1, pp_size=8, pp_max_micro_batch_size=None, pp_async_batch_depth=0, dp_size=1, load_balance_method='follow_bootstrap_room', attn_cp_size=1, moe_dp_size=1, dcp_comm_backend='ag_rs', dcp_replicate_q_proj=None, dwdp_size=1, enable_prefill_cp=False, cp_strategy=None, enable_dsa_cache_layer_split=False, enable_dsa_prefill_context_parallel=False, dsa_prefill_cp_mode='round-robin-split', enable_prefill_context_parallel=False, prefill_cp_mode='in-seq-split', enable_cp_decode_attn_tp=False, enable_dp_attention=False, enable_dp_attention_local_control_broadcast=False, enable_dp_lm_head=False, enable_attn_tp_input_scattered=False, disable_attn_tp_gather=False, enable_p2p_check=False, device='cuda', base_gpu_id=0, gpu_id_step=1, random_seed=927276655, watchdog_timeout=300, soft_watchdog_timeout=None, sleep_on_idle=False, use_ray=False, custom_sigquit_handler=None, numa_node=None, gc_threshold=None, host='0.0.0.0', port=30000, fastapi_root_path='', smg_grpc_mode=False, grpc_mode=False, grpc_port=None, sidecar=None, sidecar_args=None, skip_server_warmup=False, warmups=None, enable_http2=False, ssl_keyfile=None, ssl_certfile=None, ssl_ca_certs=None, ssl_keyfile_password=None, enable_ssl_refresh=False, api_key=None, admin_api_key=None, served_model_name='kimi-k3', weight_version='default', chat_template=None, hf_chat_template_name=None, completion_template=None, file_storage_path='sglang_storage', enable_cache_report=False, reasoning_parser=None, default_chat_template_kwargs=None, strip_thinking_cache=False, enable_strict_thinking=False, tool_call_parser=None, tool_server=None, sampling_defaults='model', asr_max_buffer_seconds=60, asr_max_concurrent_sessions=32, preferred_sampling_params=None, allow_auto_truncate=False, stream_interval=1, batch_notify_size=16, stream_response_default_include_usage=False, incremental_streaming_output=False, enable_streaming_session=False, enable_session_radix_cache=False, log_level='info', log_level_http=None, log_requests=False, log_requests_level=2, log_requests_format='text', log_requests_target=None, uvicorn_access_log_exclude_prefixes=[], crash_dump_folder=None, show_time_cost=False, enable_metrics=False, smg_http_sidecar_port=None, enable_mfu_metrics=False, enable_metrics_for_all_schedulers=False, load_snapshot_publish_interval=15, tokenizer_metrics_custom_labels_header='x-custom-labels', tokenizer_metrics_allowed_custom_labels=None, extra_metric_labels=None, bucket_time_to_first_token=None, bucket_inter_token_latency=None, bucket_e2e_request_latency=None, prompt_tokens_buckets=None, generation_tokens_buckets=None, gc_warning_threshold_secs=0.0, decode_log_interval=40, enable_request_time_stats_logging=False, kv_events_config=None, enable_forward_pass_metrics=False, forward_pass_metrics_worker_id='', forward_pass_metrics_ipc_name=None, enable_trace=False, trace_modules='request', otlp_traces_endpoint='localhost:4317', export_metrics_to_file=False, export_metrics_to_file_dir=None, stat_loggers=None, constrained_json_whitespace_pattern=None, constrained_json_disable_any_whitespace=False, attention_backend='triton', decode_attention_backend=None, prefill_attention_backend=None, sampling_backend='flashinfer', grammar_backend='xgrammar', radix_cache_backend=None, mm_attention_backend=None, fp8_gemm_runner_backend='auto', fp4_gemm_runner_backend='auto', bf16_gemm_backend='auto', dsa_prefill_backend=None, dsa_decode_backend=None, dsa_paged_mqa_logits_backend='auto', dsa_topk_backend='sgl-kernel', disable_flashinfer_autotune=False, flashinfer_autotune_skip_ops=None, mamba_backend='triton', cuda_graph_config=CudaGraphConfig(decode=PhaseConfig(backend='disabled', max_bs=16, bs=[1, 2, 4, 8, 12, 16], tc_compiler='eager', full_prefill_max_req=None), prefill=PhaseConfig(backend='disabled', max_bs=2048, bs=[4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048], tc_compiler='eager', full_prefill_max_req=None)), cuda_graph_backend_decode=None, cuda_graph_backend_prefill=None, cuda_graph_max_bs_decode=16, cuda_graph_max_bs_prefill=None, cuda_graph_bs_decode=None, cuda_graph_bs_prefill=None, cuda_graph_tc_compiler=None, disable_prefill_cuda_graph=False, disable_decode_cuda_graph=False, disable_cuda_graph=False, disable_cuda_graph_padding=False, enable_profile_cuda_graph=False, enable_cudagraph_gc=False, debug_cuda_graph=False, enable_layerwise_nvtx_marker=False, enable_nccl_nvls=False, enable_symm_mem=False, triton_attention_reduce_in_fp32=False, triton_attention_num_kv_splits=8, triton_attention_split_tile_size=None, flashinfer_mla_disable_ragged=False, enable_fused_qk_norm_rope=False, enable_precise_embedding_interpolation=False, enable_fused_moe_sum_all_reduce=False, enable_deepseek_v4_fp4_indexer=False, disable_custom_all_reduce=False, enable_mscclpp=False, enable_torch_symm_mem=False, enable_scattered_sconv=False, pre_warm_nccl=False, enable_quant_communications=False, enable_flashinfer_allreduce_fusion=False, enforce_disable_flashinfer_allreduce_fusion=False, flashinfer_allreduce_fusion_backend=None, enable_aiter_allreduce_fusion=False, enable_torch_compile=False, enable_torch_compile_debug_mode=False, torch_compile_max_bs=32, torchao_config='', speculative_algorithm=None, speculative_draft_model_path=None, speculative_draft_model_revision=None, speculative_draft_load_format=None, speculative_num_steps=None, speculative_eagle_topk=None, speculative_num_draft_tokens=None, speculative_dflash_block_size=None, speculative_dspark_block_size=None, speculative_dspark_sps_table_path=None, speculative_dspark_confidence_sts_path=None, speculative_dspark_align_verify_tokens_to_graph_tier=False, speculative_accept_threshold_single=1.0, speculative_accept_threshold_acc=1.0, speculative_use_rejection_sampling=False, speculative_token_map=None, speculative_attention_mode='prefill', speculative_draft_attention_backend=None, speculative_draft_window_size=None, speculative_moe_runner_backend='flashinfer_mxfp4', speculative_moe_a2a_backend=None, speculative_draft_model_quantization=None, speculative_skip_dp_mlp_sync=False, enable_multi_layer_eagle=False, speculative_adaptive=False, speculative_adaptive_config=None, decoupled_spec_bind_endpoint=None, decoupled_spec_connect_endpoints=None, decoupled_spec_rank=None, decoupled_spec_role='null', spec_trace_dir=None, speculative_ngram_min_bfs_breadth=1, speculative_ngram_max_bfs_breadth=10, speculative_ngram_match_type='BFS', speculative_ngram_max_trie_depth=18, speculative_ngram_capacity=10000000, speculative_ngram_external_corpus_path=None, speculative_ngram_external_sam_budget=0, speculative_ngram_external_corpus_max_tokens=10000000, ep_size=4, moe_a2a_backend='none', moe_runner_backend='flashinfer_mxfp4', flashinfer_mxfp4_moe_precision='default', deepep_mode='auto', fuseep_mode=2, deepep_dispatcher_output_dtype='auto', ep_num_redundant_experts=0, ep_dispatch_algorithm=None, init_expert_location='trivial', enable_eplb=False, eplb_algorithm='auto', eplb_rebalance_num_iterations=1000, eplb_rebalance_layers_per_chunk=None, eplb_min_rebalancing_utilization_threshold=1.0, expert_distribution_recorder_mode=None, expert_distribution_recorder_buffer_size=1000, enable_expert_distribution_metrics=False, deepep_config=None, moe_dense_tp_size=None, elastic_ep_backend=None, enable_elastic_expert_backup=False, mooncake_ib_device=None, enable_waterfill=False, ep_join_mode=None, ep_join_rank_offset=0, elastic_ep_initial_size=None, max_ep_size=None, elastic_ep_scale_timeout=600, elastic_ep_rejoin=False, disable_flashinfer_cutlass_moe_fp4_allgather=False, disable_shared_experts_fusion=False, enforce_shared_experts_fusion=False, max_mamba_cache_size=None, mamba_ssm_dtype=None, mamba_max_states_per_path=-1, enable_mamba_cache_stochastic_rounding=False, mamba_cache_philox_rounds=0, mamba_full_memory_ratio=0.36, mamba_radix_cache_strategy='extra_buffer_lazy', uses_mamba_radix_cache=False, mamba_track_interval=256, enable_int8_mamba_checkpoint=False, int8_mamba_ckpt_size=None, linear_attn_backend='triton', linear_attn_decode_backend=None, linear_attn_prefill_backend=None, linear_attn_verify_backend=None, enable_linear_replayssm=False, linear_replayssm_cache_len=16, enable_linear_replayssm_spec=False, enable_hierarchical_cache=False, hicache_ratio=2.0, hicache_size=0, hicache_write_policy='write_through', hicache_io_backend='kernel', hicache_mem_layout='page_first', hicache_storage_backend=None, hicache_storage_prefetch_policy='timeout', hicache_storage_backend_extra_config=None, enable_hisparse=False, hisparse_config=None, enable_broadcast_mm_inputs_process=False, enable_prefix_mm_cache=False, mm_enable_dp_encoder=False, mm_process_config={}, mm_processor_worker_num=0, mm_io_worker_num=0, limit_mm_data_per_request=None, enable_mm_global_cache=False, disable_fast_image_processor=False, mm_feature_transport='cpu', keep_mm_feature_on_device=False, enable_lora=None, enable_lora_overlap_loading=None, max_lora_rank=None, lora_target_modules=None, lora_paths=None, max_loaded_loras=None, max_loras_per_batch=8, lora_eviction_policy='lru', lora_backend='csgmv', max_lora_chunk_size=16, experts_shared_outer_loras=None, lora_use_virtual_experts=False, lora_strict_loading=False, lora_drain_wait_threshold=0.0, enable_two_batch_overlap=False, enable_single_batch_overlap=False, tbo_token_distribution_threshold=0.48, cpu_offload_gb=0, offload_group_size=-1, offload_num_in_group=1, offload_prefetch_step=1, offload_mode='cpu', enable_lmcache=False, lmcache_config_file=None, enable_flexkv=False, flexkv_config_file=None, kt_weight_path=None, kt_method='AMXINT4', kt_cpuinfer=None, kt_threadpool_count=2, kt_num_gpu_experts=None, kt_max_deferred_experts_per_token=None, dllm_algorithm=None, dllm_algorithm_config=None, dllm_fdfo=True, disaggregation_mode='prefill', disaggregation_transfer_backend='mooncake', disaggregation_bootstrap_port=28800, disaggregation_ib_device='mlx5_0,mlx5_1,mlx5_2,mlx5_3', disaggregation_decode_enable_radix_cache=False, disaggregation_decode_enable_offload_kvcache=False, num_reserved_decode_tokens=512, disaggregation_decode_extra_slots=None, disaggregation_decode_polling_interval=1, optimistic_prefill_attempts=0, encoder_only=False, language_only=False, encoder_transfer_backend='zmq_to_tokenizer', encoder_urls=[], encoder_bootstrap_port=8997, encoder_register_urls=[], enable_adaptive_dispatch_to_encoder=False, enable_pdmux=False, pdmux_config_path=None, sm_group_num=8, custom_weight_loader=[], weight_loader_disable_mmap=False, weight_loader_prefetch_checkpoints=False, weight_loader_prefetch_num_threads=4, weight_loader_drop_cache_after_load=False, remote_instance_weight_loader_seed_instance_ip=None, remote_instance_weight_loader_seed_instance_service_port=None, remote_instance_weight_loader_send_weights_group_ports=None, remote_instance_weight_loader_backend='nccl', remote_instance_weight_loader_start_seed_via_transfer_engine=False, engine_info_bootstrap_port=6789, modelexpress_config=None, download_dir=None, model_checksum=None, delete_ckpt_after_loading=False, decrypted_config_file=None, decrypted_draft_config_file=None, checkpoint_engine_wait_weights_before_ready=False, enable_prefill_delayer=False, prefill_delayer_max_delay_passes=30, prefill_delayer_token_usage_low_watermark=None, prefill_delayer_forward_passes_buckets=None, prefill_delayer_wait_seconds_buckets=None, prefill_delayer_queue_min_ratio=None, prefill_delayer_max_delay_ms=None, min_free_slots_delay=None, enable_deterministic_inference=False, rl_on_policy_target=None, kv_canary='none', kv_canary_real_data='none', kv_canary_sweep_interval=0, enable_dynamic_batch_tokenizer=False, dynamic_batch_tokenizer_batch_size=32, dynamic_batch_tokenizer_batch_timeout=0.002, enable_tokenizer_batch_encode=False, disable_tokenizer_batch_decode=False, debug_tensor_dump_output_folder=None, debug_tensor_dump_layers=None, debug_tensor_dump_input_file=None, enable_memory_saver=False, enable_weights_cpu_backup=False, enable_draft_weights_cpu_backup=False, enable_custom_logit_processor=False, enable_return_hidden_states=False, enable_return_routed_experts=False, enable_return_indexer_topk=False, disable_outlines_disk_cache=False, enable_mis=False, weight_cache_mode='off', weight_cache_socket=None, weight_cache_timeout=1800, forward_hooks=None, msprobe_dump_config=None)
+[2026-08-27 04:28:49] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:49] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:49] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:49] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:49] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:49] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:49] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[2026-08-27 04:28:49] User lacks permission to set NUMA affinity, skipping NUMA node configuration for GPU. If using docker, try adding --cap-add SYS_NICE to your docker run command.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:05 PP6 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:05 PP6 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:07 PP6 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:07 PP6 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:07 PP6 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:07 PP7 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:07 PP7 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] `KimiK3Processor` defines `image_processor_class = 'AutoImageProcessor'`, which is deprecated. Register the correct mapping in `AutoImageProcessor` instead.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[transformers] The `use_fast` parameter is deprecated and will be removed in a future version. Use `backend="torchvision"` instead of `use_fast=True`, or `backend="pil"` instead of `use_fast=False`.
+[2026-08-27 04:29:07 PP7 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:07 PP7 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:07 PP6 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:07 PP6 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:07.662256   238 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:07.662379   238 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:07.662392   238 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.4 port: 12001
+I0827 04:29:07.662429   238 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.4:15914
+I0827 04:29:07.662514   238 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:07.685189   238 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:07.689950   238 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:07.695539   238 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:07.701294   238 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:07.703691   238 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:07.703715   238 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:07.703760   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:07.710492   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:07.711953   238 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:04
+I0827 04:29:07.711989   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:07.735716   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:07.737195   238 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:04
+I0827 04:29:07.737223   238 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:07.763653   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:07.765146   238 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:04
+I0827 04:29:07.765172   238 rdma_context.cpp:205] Using SIEVE endpoint store
+[2026-08-27 04:29:07 PP7 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:07 PP7 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:07 PP7 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:07 PP7 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+I0827 04:29:07.791589   238 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:07.793025   238 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:04
+I0827 04:29:07.793139   238 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:07 PP6 TP3 EP3] Init torch distributed begin.
+[2026-08-27 04:29:07 PP6 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:29:07 PP6 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:29:08 PP6 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:08.916008   235 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:08.916072   235 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:08.916082   235 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.4 port: 12001
+I0827 04:29:08.916121   235 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.4:15737
+I0827 04:29:08.916215   235 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:08.941840   235 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:08.950320   235 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:08.979278   235 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:08.988204   235 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:08.991855   235 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:08.991879   235 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:08.991928   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:08.999985   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:09.001416   235 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:04
+I0827 04:29:09.001446   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:09.027477   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:09.028954   235 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:04
+I0827 04:29:09.028980   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:09.055524   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:09.056981   235 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:04
+I0827 04:29:09.057020   235 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:09.083544   235 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:09.084911   235 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:04
+I0827 04:29:09.085031   235 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:09 PP6 TP0 EP0] Init torch distributed begin.
+[2026-08-27 04:29:09 PP7 TP3 EP3] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:09 PP6 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:09 PP7 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:09 PP6 TP2 EP2] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:09 PP7 TP1 EP1] Multimodal data loading enabled with 16 worker threads (auto).
+[2026-08-27 04:29:09 PP7 TP0 EP0] Multimodal data loading enabled with 16 worker threads (auto).
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:10.000486   242 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:10.000609   242 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:10.000628   242 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.4 port: 12001
+I0827 04:29:10.000689   242 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.4:16890
+I0827 04:29:10.000851   242 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:10.006592   236 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:10.006711   236 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:10.006728   236 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.4 port: 12001
+I0827 04:29:10.006786   236 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.4:15893
+I0827 04:29:10.006963   236 transfer_engine_impl.cpp:258] Auto-discovering topology...
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:10.022238   241 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:10.022354   241 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:10.022370   241 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.4 port: 12001
+I0827 04:29:10.022423   241 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.4:16277
+I0827 04:29:10.022593   241 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:10.032052   242 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:10.032101   236 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:10.039772   236 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:10.039858   242 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:10.047292   236 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:10.047341   242 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:10.047785   241 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:10.054806   242 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:10.054838   236 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:10.055280   241 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:10.059140   236 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:10.059171   236 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:10.059175   242 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:10.059201   242 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:10.059227   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.059260   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.062610   241 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:10.066787   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:10.066826   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:10.068500   242 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:04
+I0827 04:29:10.068532   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.068558   236 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:04
+I0827 04:29:10.068590   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.069866   241 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:10.072566   241 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:10.072592   241 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:10.072649   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.080149   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:10.081809   241 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:04
+I0827 04:29:10.081843   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.095701   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:10.097368   236 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:04
+I0827 04:29:10.097400   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.099974   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:10.101672   242 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:04
+I0827 04:29:10.101717   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.107705   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:10.109316   241 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:04
+I0827 04:29:10.109347   241 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.123868   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:10.125504   236 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:04
+I0827 04:29:10.125536   236 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.127957   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:10.129673   242 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:04
+I0827 04:29:10.129703   242 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.139813   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:10.141404   241 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:04
+I0827 04:29:10.141431   241 rdma_context.cpp:205] Using SIEVE endpoint store
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:10.150050   237 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:10.150159   237 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:10.150173   237 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.4 port: 12001
+I0827 04:29:10.150219   237 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.4:15413
+I0827 04:29:10.150358   237 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:10.151880   236 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:10.153475   236 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:04
+I0827 04:29:10.153601   236 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:10 PP6 TP1 EP1] Init torch distributed begin.
+I0827 04:29:10.155849   242 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:10.155864   237 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:10.157462   242 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:04
+I0827 04:29:10.157585   242 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:10 PP7 TP3 EP3] Init torch distributed begin.
+I0827 04:29:10.171923   241 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:10.172516   240 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:10.172667   240 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:10.172688   240 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.4 port: 12001
+I0827 04:29:10.172762   240 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.4:15705
+I0827 04:29:10.172977   240 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:10.173573   241 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:04
+I0827 04:29:10.173692   241 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:10 PP7 TP2 EP2] Init torch distributed begin.
+WARNING: Logging before InitGoogleLogging() is written to STDERR
+W0827 04:29:10.177717   239 transfer_engine_py.cpp:100] Using default malloc/free for protocol: rdma
+I0827 04:29:10.177865   239 transfer_engine_impl.cpp:971] Metrics reporting is disabled (set MC_TE_METRIC=1 to enable)
+I0827 04:29:10.177886   239 transfer_engine_impl.cpp:115] Transfer Engine parseHostNameWithPort. server_name: 174.1.60.4 port: 12001
+I0827 04:29:10.177956   239 transfer_engine_impl.cpp:182] Transfer Engine RPC using P2P handshake, listening on 174.1.60.4:16192
+I0827 04:29:10.178146   239 transfer_engine_impl.cpp:258] Auto-discovering topology...
+I0827 04:29:10.183638   237 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:10.187090   239 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:10.191001   237 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:10.195258   239 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:10.198271   237 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:10.200984   237 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:10.201025   237 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:10.201079   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.203014   239 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:10.203156   240 topology.cpp:127] Device mlx5_0 port 1 is available
+I0827 04:29:10.209589   240 topology.cpp:127] Device mlx5_1 port 1 is available
+I0827 04:29:10.210305   239 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:10.212793   239 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:10.212821   239 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:10.212878   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.216092   240 topology.cpp:127] Device mlx5_2 port 1 is available
+I0827 04:29:10.219946   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:10.221297   239 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:04
+I0827 04:29:10.221323   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.222187   240 topology.cpp:127] Device mlx5_3 port 1 is available
+I0827 04:29:10.224284   240 transfer_engine_impl.cpp:273] Topology discovery complete. Found 4 HCAs.
+I0827 04:29:10.224304   240 rdma_transport.cpp:76] [RDMA] Relaxed ordering disabled via MC_IB_PCI_RELAXED_ORDERING=0. Falling back to strict ordering.
+I0827 04:29:10.224344   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.230640   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:10.231778   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_2/
 (with network device)
+I0827 04:29:10.231937   240 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:04
+I0827 04:29:10.231962   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.233362   237 rdma_context.cpp:288] RDMA device: mlx5_2, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:17:04
+I0827 04:29:10.233397   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.236478   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:10.237651   240 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:04
+I0827 04:29:10.237670   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.250197   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:10.251711   239 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:04
+I0827 04:29:10.251737   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.263743   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_3/
 (with network device)
+I0827 04:29:10.265311   237 rdma_context.cpp:288] RDMA device: mlx5_3, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:18:04
+I0827 04:29:10.265339   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.267139   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:10.268607   240 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:04
+I0827 04:29:10.268632   240 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.273075   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:10.274682   237 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:04
+I0827 04:29:10.274710   237 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.276602   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_0/
 (with network device)
+I0827 04:29:10.278275   239 rdma_context.cpp:288] RDMA device: mlx5_0, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:15:04
+I0827 04:29:10.278314   239 rdma_context.cpp:205] Using SIEVE endpoint store
+I0827 04:29:10.295709   240 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:10.297385   240 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:04
+I0827 04:29:10.297510   240 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:10 PP7 TP1 EP1] Init torch distributed begin.
+I0827 04:29:10.299893   237 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:10.301563   237 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:04
+I0827 04:29:10.301689   237 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:10 PP6 TP2 EP2] Init torch distributed begin.
+I0827 04:29:10.303737   239 rdma_context.cpp:1353] Using user-specified GID index: 3 on mlx5_1/
 (with network device)
+I0827 04:29:10.305307   239 rdma_context.cpp:288] RDMA device: mlx5_1, LID: 0, GID: (GID_Index 3) 00:00:00:00:00:00:00:00:00:00:ff:ff:0a:64:16:04
+I0827 04:29:10.305433   239 transfer_engine_impl.cpp:388] installTransport, type=rdma
+[2026-08-27 04:29:10 PP7 TP0 EP0] Init torch distributed begin.
+[2026-08-27 04:29:26 PP7 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:26 PP6 TP0 EP0] sglang is using nccl==2.28.9
+[2026-08-27 04:29:27 PP6 TP0 EP0] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP6 TP3 EP3] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP6 TP2 EP2] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP6 TP1 EP1] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP7 TP1 EP1] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP7 TP0 EP0] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP7 TP3 EP3] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP7 TP2 EP2] CustomAllreduce is disabled because it's not supported on more than two PCIe-only GPUs. To silence this warning, specify disable_custom_all_reduce=True explicitly.
+[2026-08-27 04:29:27 PP7 TP1 EP1] Init torch distributed ends. elapsed=17.45 s, mem usage=0.46 GB
+[2026-08-27 04:29:27 PP7 TP3 EP3] Init torch distributed ends. elapsed=17.59 s, mem usage=0.46 GB
+[2026-08-27 04:29:27 PP7 TP2 EP2] Init torch distributed ends. elapsed=17.57 s, mem usage=0.46 GB
+[2026-08-27 04:29:27 PP7 TP0 EP0] Init torch distributed ends. elapsed=17.44 s, mem usage=0.46 GB
+[2026-08-27 04:29:27 PP6 TP1 EP1] Init torch distributed ends. elapsed=17.59 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP6 TP3 EP3] Init torch distributed ends. elapsed=19.95 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP6 TP0 EP0] Init torch distributed ends. elapsed=18.66 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP6 TP2 EP2] Init torch distributed ends. elapsed=17.44 s, mem usage=0.54 GB
+[2026-08-27 04:29:27 PP6 TP1 EP1] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP7 TP1 EP1] Load weight begin. avail mem=82.12 GB
+[2026-08-27 04:29:27 PP6 TP0 EP0] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP6 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP7 TP1 EP1] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP7 TP3 EP3] Load weight begin. avail mem=82.12 GB
+[2026-08-27 04:29:27 PP6 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP7 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:27 PP6 TP2 EP2] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:27 PP6 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP7 TP2 EP2] Load weight begin. avail mem=82.12 GB
+[2026-08-27 04:29:28 PP7 TP2 EP2] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP6 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP7 TP1 EP1] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP6 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP7 TP1 EP1] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP7 TP0 EP0] Load weight begin. avail mem=82.12 GB
+[2026-08-27 04:29:28 PP6 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP7 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP7 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP6 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP7 TP0 EP0] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP6 TP3 EP3] Load weight begin. avail mem=82.04 GB
+[2026-08-27 04:29:28 PP6 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP6 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP6 TP3 EP3] Kimi-K3 vision attention uses shape-aware auto selection on B300/GB300 (Triton for small workloads, FA4 otherwise).
+[2026-08-27 04:29:28 PP7 TP2 EP2] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP7 TP2 EP2] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP7 TP0 EP0] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP6 TP3 EP3] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
+[2026-08-27 04:29:28 PP7 TP0 EP0] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:29:28 PP6 TP3 EP3] FlashInfer TRTLLM MoE deferred finalize is disabled (moe_runner_backend=flashinfer_mxfp4, quant_method=Mxfp4MoEMethod).
+[2026-08-27 04:30:40 PP7 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:40 PP7 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:40 PP7 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:40 PP7 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:40 PP7 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:40 PP7 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:40 PP7 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:40 PP7 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:41 PP7 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:41 PP7 TP1 EP1] Load weight end. elapsed=73.17 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=31.91 GB, mem usage=50.21 GB.
+[2026-08-27 04:30:41 PP7 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:41 PP7 TP0 EP0] Load weight end. elapsed=73.80 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=31.91 GB, mem usage=50.21 GB.
+[2026-08-27 04:30:42 PP6 TP1 EP1] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:42 PP6 TP1 EP1] Load weight end. elapsed=74.47 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:45 PP6 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:45 PP6 TP2 EP2] Load weight end. elapsed=77.20 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:50 PP7 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:50 PP7 TP3 EP3] Load weight end. elapsed=82.79 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=31.91 GB, mem usage=50.21 GB.
+[2026-08-27 04:30:51 PP6 TP3 EP3] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:51 PP6 TP3 EP3] Load weight end. elapsed=83.19 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:51 PP7 TP2 EP2] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:51 PP7 TP2 EP2] Load weight end. elapsed=83.68 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=31.91 GB, mem usage=50.21 GB.
+[2026-08-27 04:30:51 PP7 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:51 PP7 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:52 PP7 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:52 PP7 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:52 PP7 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:52 PP7 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:52 PP7 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:52 PP7 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:52 PP6 TP0 EP0] K3 fused KDA decode disabled: unexpected conv/A_log/dt_bias layout (conv (9216, 4), A_log (1, 1, 24, 1), dt_bias (3072,))
+[2026-08-27 04:30:53 PP6 TP0 EP0] Precompiled dynamic-token fused K3 vision RoPE kernel
+[2026-08-27 04:30:53 PP6 TP0 EP0] Load weight end. elapsed=85.47 s, type=KimiK3ForConditionalGeneration, quant=compressed-tensors, avail mem=32.29 GB, mem usage=49.75 GB.
+[2026-08-27 04:30:53 PP6 TP0 EP0] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:53 PP6 TP0 EP0] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:53 PP6 TP3 EP3] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:53 PP6 TP3 EP3] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:53 PP6 TP2 EP2] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:53 PP6 TP2 EP2] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:53 PP6 TP1 EP1] Reloaded tiktoken model from /data/hf_models/Kimi-K3/tiktoken.model
+[2026-08-27 04:30:53 PP6 TP1 EP1] #words: 163840 - BOS ID: 163584 - EOS ID: 163585
+[2026-08-27 04:30:54 PP7 TP3 EP3] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP7 TP2 EP2] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP7 TP1 EP1] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP7 TP0 EP0] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP6 TP3 EP3] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP6 TP2 EP2] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP6 TP1 EP1] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP6 TP0 EP0] max_running_requests is capped to 54 by the mamba state cache (max_mamba_cache_size=54, 1 state slots per request). To raise it: increase --mamba-full-memory-ratio or --max-mamba-cache-size, or halve the state size with --mamba-ssm-dtype bfloat16.
+[2026-08-27 04:30:54 PP6 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP6 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP7 TP2 EP2] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:54 PP6 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP7 TP3 EP3] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:54 PP7 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:54 PP7 TP1 EP1] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.02GB, ssm_state size: 0.64GB 
+[2026-08-27 04:30:54 PP6 TP0 EP0] Mamba Cache is allocated. max_mamba_cache_size: 54, conv_state size: 0.03GB, ssm_state size: 0.73GB 
+[2026-08-27 04:30:54 PP6 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP6 TP3 EP3] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP6 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP6 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP6 TP2 EP2] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP6 TP1 EP1] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP7 TP2 EP2] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 16.31 GB
+[2026-08-27 04:30:54 PP7 TP2 EP2] Memory pool end. avail mem=14.95 GB
+[2026-08-27 04:30:54 PP6 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 12.24 GB
+[2026-08-27 04:30:54 PP6 TP0 EP0] Memory pool end. avail mem=19.33 GB
+[2026-08-27 04:30:54 PP7 TP3 EP3] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 16.31 GB
+[2026-08-27 04:30:54 PP7 TP0 EP0] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 16.31 GB
+[2026-08-27 04:30:54 PP7 TP1 EP1] KV Cache is allocated. dtype: torch.bfloat16, #tokens: 3801408, KV size: 16.31 GB
+[2026-08-27 04:30:54 PP7 TP3 EP3] Memory pool end. avail mem=14.95 GB
+[2026-08-27 04:30:54 PP7 TP0 EP0] Memory pool end. avail mem=14.95 GB
+[2026-08-27 04:30:54 PP7 TP1 EP1] Memory pool end. avail mem=14.95 GB
+[2026-08-27 04:30:54 PP7 TP2 EP2] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp2_pp7_dp0.json
+[2026-08-27 04:30:54 PP6 TP3 EP3] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp3_pp6_dp0.json
+[2026-08-27 04:30:54 PP6 TP1 EP1] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp1_pp6_dp0.json
+[2026-08-27 04:30:54 PP7 TP3 EP3] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp3_pp7_dp0.json
+[2026-08-27 04:30:54 PP6 TP2 EP2] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp2_pp6_dp0.json
+[2026-08-27 04:30:54 PP6 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:30:54 PP6 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:30:54 PP6 TP0 EP0] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp0_pp6_dp0.json
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[TensorRT-LLM][INFO] Set logger level to INFO
+[2026-08-27 04:30:54 PP7 TP0 EP0] Linear attention kernel backend: decode=triton, prefill=triton, verify=triton
+[2026-08-27 04:30:54 PP7 TP0 EP0] KDA kernel dispatcher: decode=TritonKDAKernel, verify=TritonKDAKernel, extend=TritonKDAKernel packed_decode=True
+[2026-08-27 04:30:54 PP7 TP0 EP0] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp0_pp7_dp0.json
+[2026-08-27 04:30:54 PP7 TP1 EP1] Running FlashInfer autotune with cache: /root/.cache/sglang/flashinfer/autotune/0.6.18/sm120/e367314fa83c5b67/rank_tp1_pp7_dp0.json
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 11.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  9.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 13.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.76profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.62profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.66profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.72profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.68profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.55profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.12profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  6.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  6.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  7.16profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.92profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.15profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.59profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.56profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 19.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  8.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.45profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.67profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.40profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.88profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  8.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  6.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.83profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.50profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.32profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.92profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.77profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.94profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.57profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 12.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 12.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 13.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.36profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.66profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  6.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.72profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  6.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.22profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.65profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.13profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.03profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.46profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.59profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.16profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  6.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.96profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.76profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.67profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  8.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.27profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.88profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.01profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.57profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.12profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  5.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  6.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 11.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 10.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 11.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  2.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.68profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.90profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  8.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.80profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.43profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.92profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  8.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  8.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.66profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  7.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  6.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:00<00:00,  7.13profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.32profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.71profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:05,  1.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.16profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.31profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.77profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.81profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.28profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.48profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.87profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.40profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 13.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.35profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.98profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.88profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.73profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  3.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.88profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.09profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.20profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.97profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.83profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.85profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.47profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.40profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.60profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  8.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  7.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.97profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  9.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.05profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.05profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  6.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.49profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.72profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.19profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 10.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  6.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 10.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  8.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  5.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.91profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.64profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.58profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  3.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.92profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.35profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.06profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  9.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.07profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00, 12.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.09profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:02,  2.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 12.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 10.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.15profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.76profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.49profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.03profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 10.79profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.96profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.36profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.47profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.65profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  2.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.07profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.62profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.55profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.84profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  3.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.03profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  9.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  9.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.63profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.58profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.63profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.98profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.54profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:03,  1.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.61profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:05,  1.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.97profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  2.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:01,  3.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.50profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  4.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  6.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.90profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.94profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.13profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.09profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.78profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.56profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.88profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.53profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.36profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.30profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  3.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.35profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.26profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  3.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.56profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00, 15.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  3.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  6.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00, 14.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.44profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  8.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00, 12.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.36profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.52profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00, 12.34profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.79profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.42profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.05profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.69profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:01<00:06,  1.06s/profile]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:01<00:01,  3.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  8.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.66profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.52profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  4.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.24profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  6.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.00profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.82profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.04profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.14profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  6.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.91profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.94profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.38profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.02profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.84profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.04profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:02,  2.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.35profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.10profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.95profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.57profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.34profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.66profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.70profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.28profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.08profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  8.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.81profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:04,  1.29profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:01<00:00,  4.22profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  7.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:01,  4.51profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.67profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:02,  2.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  6.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.19profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  7.72profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.96profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:01<00:00,  4.47profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.26profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.83profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:01<00:00,  5.23profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.02profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:01<00:00,  5.33profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.85profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.01profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.64profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.66profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.44profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.13profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.53profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  7.59profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  8.49profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:00,  7.73profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.20profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  4.21profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.83profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  14%|█▍        | 1/7 [00:00<00:02,  2.08profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:00,  8.75profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.12profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.78profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  8.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  5.89profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  29%|██▊       | 2/7 [00:00<00:01,  3.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.68profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.71profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.60profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  43%|████▎     | 3/7 [00:00<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00, 10.15profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  57%|█████▋    | 4/7 [00:00<00:00,  8.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  6.43profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  6.91profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.59profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  71%|███████▏  | 5/7 [00:00<00:00,  7.61profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.76profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  7.46profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1:  86%|████████▌ | 6/7 [00:00<00:00,  9.54profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.48profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  8.71profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.14profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:00<00:00,  9.10profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.87profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.93profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  6.29profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  7.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  7.25profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm1: 100%|██████████| 7/7 [00:01<00:00,  5.97profile/s]
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:   0%|          | 0/7 [00:00<?, ?profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.74profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  4.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:00,  7.03profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  14%|█▍        | 1/7 [00:00<00:01,  3.86profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  4.38profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  7.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.11profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.88profile/s]
+[2026-08-27 04:31:27 PP6 TP3 EP3] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.42profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.86profile/s]
+[2026-08-27 04:31:27 PP6 TP1 EP1] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  4.27profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.55profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:00,  5.13profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  7.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  29%|██▊       | 2/7 [00:00<00:01,  3.39profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.32profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:00,  5.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.16profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  7.41profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  5.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  6.45profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.17profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.24profile/s]
+[2026-08-27 04:31:27 PP6 TP2 EP2] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  43%|████▎     | 3/7 [00:00<00:01,  3.77profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.31profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  6.65profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  3.98profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.33profile/s]
+[2026-08-27 04:31:27 PP6 TP0 EP0] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:00<00:00,  5.80profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  57%|█████▋    | 4/7 [00:00<00:00,  4.56profile/s][2026-08-27 04:31:27 PP6 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:27 PP6 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:27 PP6 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:27 PP6 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:27 PP6 TP0 EP0] max_total_num_tokens=3801408, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=54, context_len=1048576, available_gpu_mem=16.52 GB
+[2026-08-27 04:31:27 PP6 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:27 PP6 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:27 PP6 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:27 PP6 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  6.37profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:00<00:00,  6.01profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  71%|███████▏  | 5/7 [00:01<00:00,  5.21profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  5.40profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.18profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.82profile/s]
+[2026-08-27 04:31:28 PP7 TP2 EP2] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.95profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  6.32profile/s]
+[2026-08-27 04:31:28 PP7 TP3 EP3] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.50profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  5.60profile/s]
+[2026-08-27 04:31:28 PP7 TP0 EP0] FlashInfer autotune completed.
+
[AutoTuner]: Tuning trtllm::fused_moe::gemm2:  86%|████████▌ | 6/7 [00:01<00:00,  4.69profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.99profile/s]
[AutoTuner]: Tuning trtllm::fused_moe::gemm2: 100%|██████████| 7/7 [00:01<00:00,  4.60profile/s]
+[2026-08-27 04:31:28 PP7 TP1 EP1] FlashInfer autotune completed.
+[2026-08-27 04:31:28 PP7 TP2 EP2] multimem all-gather disabled (no multicast for world_size=4)
+[2026-08-27 04:31:28 PP7 TP3 EP3] multimem all-gather disabled (no multicast for world_size=4)
+[2026-08-27 04:31:28 PP7 TP1 EP1] multimem all-gather disabled (no multicast for world_size=4)
+[2026-08-27 04:31:28 PP7 TP0 EP0] multimem all-gather disabled (no multicast for world_size=4)
+[2026-08-27 04:31:28 PP7 TP2 EP2] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:28 PP7 TP3 EP3] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:28 PP7 TP1 EP1] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:28 PP7 TP3 EP3] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:28 PP7 TP0 EP0] Disable prefill CUDA graph because cuda_graph_config resolved prefill.backend='disabled' (e.g. via --cuda-graph-backend-prefill=disabled or auto-disable rules).
+[2026-08-27 04:31:28 PP7 TP2 EP2] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:28 PP7 TP0 EP0] max_total_num_tokens=3801408, chunked_prefill_size=8192, max_prefill_tokens=16384, max_running_requests=54, context_len=1048576, available_gpu_mem=12.14 GB
+[2026-08-27 04:31:28 PP7 TP0 EP0] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+[2026-08-27 04:31:28 PP7 TP1 EP1] Tree cache initialized: source=default impl=ChunkCache hybrid_swa=False hybrid_ssm=True hierarchical=False streaming_wrapped=False
+W0827 04:31:31.959518  2596 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5efc8ec0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.959637  2598 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ee654c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.959640  2597 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5eda3c80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.959542  2595 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5efc6f80, len: 6912: Operation not permitted [1]
+W0827 04:31:31.959741  2599 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5eeaf500, len: 55296: Operation not permitted [1]
+W0827 04:31:31.959794  2600 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5ef4e380, len: 55296: Operation not permitted [1]
+W0827 04:31:31.959827  2601 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5f548500, len: 6912: Operation not permitted [1]
+W0827 04:31:31.959918  2602 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5eec3bc0, len: 13824: Operation not permitted [1]
+W0827 04:31:31.960085  2603 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5f01bcc0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.960222  2604 memory_location.cpp:73] Failed to get NUMA node, addr: 0x5f4c8ac0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.980162  2623 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60ba35c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.980152  2624 memory_location.cpp:73] Failed to get NUMA node, addr: 0x609ccf40, len: 6912: Operation not permitted [1]
+W0827 04:31:31.980242  2625 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60d38600, len: 6912: Operation not permitted [1]
+W0827 04:31:31.980361  2626 memory_location.cpp:73] Failed to get NUMA node, addr: 0x609546c0, len: 6912: Operation not permitted [1]
+W0827 04:31:31.980516  2628 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60c49a00, len: 55296: Operation not permitted [1]
+W0827 04:31:31.980537  2627 memory_location.cpp:73] Failed to get NUMA node, addr: 0x607c6800, len: 55296: Operation not permitted [1]
+W0827 04:31:31.980547  2629 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60d11800, len: 6912: Operation not permitted [1]
+W0827 04:31:31.980834  2630 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60a1b740, len: 13824: Operation not permitted [1]
+W0827 04:31:31.981053  2632 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60ce5000, len: 6912: Operation not permitted [1]
+W0827 04:31:31.981062  2631 memory_location.cpp:73] Failed to get NUMA node, addr: 0x60aef300, len: 6912: Operation not permitted [1]
+W0827 04:31:32.117022  2651 memory_location.cpp:73] Failed to get NUMA node, addr: 0x74c79d40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.117132  2652 memory_location.cpp:73] Failed to get NUMA node, addr: 0x74f99700, len: 6912: Operation not permitted [1]
+W0827 04:31:32.117317  2653 memory_location.cpp:73] Failed to get NUMA node, addr: 0x7546c740, len: 6912: Operation not permitted [1]
+W0827 04:31:32.117331  2654 memory_location.cpp:73] Failed to get NUMA node, addr: 0x74f4f140, len: 6912: Operation not permitted [1]
+W0827 04:31:32.117414  2656 memory_location.cpp:73] Failed to get NUMA node, addr: 0x750037c0, len: 55296: Operation not permitted [1]
+W0827 04:31:32.117445  2655 memory_location.cpp:73] Failed to get NUMA node, addr: 0x74bf1200, len: 55296: Operation not permitted [1]
+W0827 04:31:32.117705  2657 memory_location.cpp:73] Failed to get NUMA node, addr: 0x74f80080, len: 6912: Operation not permitted [1]
+W0827 04:31:32.117738  2658 memory_location.cpp:73] Failed to get NUMA node, addr: 0x74fbd2c0, len: 13824: Operation not permitted [1]
+W0827 04:31:32.117790  2659 memory_location.cpp:73] Failed to get NUMA node, addr: 0x74fa5e40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.118012  2660 memory_location.cpp:73] Failed to get NUMA node, addr: 0x74c24880, len: 6912: Operation not permitted [1]
+W0827 04:31:32.137427  2677 memory_location.cpp:73] Failed to get NUMA node, addr: 0x685bea40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.137485  2678 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6865c240, len: 6912: Operation not permitted [1]
+W0827 04:31:32.137612  2680 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6832f280, len: 6912: Operation not permitted [1]
+W0827 04:31:32.137683  2679 memory_location.cpp:73] Failed to get NUMA node, addr: 0x68511ec0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.137791  2681 memory_location.cpp:73] Failed to get NUMA node, addr: 0x687ae280, len: 55296: Operation not permitted [1]
+W0827 04:31:32.137800  2682 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6863aec0, len: 55296: Operation not permitted [1]
+W0827 04:31:32.137928  2683 memory_location.cpp:73] Failed to get NUMA node, addr: 0x685f4540, len: 6912: Operation not permitted [1]
+W0827 04:31:32.138145  2685 memory_location.cpp:73] Failed to get NUMA node, addr: 0x68b70f80, len: 6912: Operation not permitted [1]
+W0827 04:31:32.138144  2684 memory_location.cpp:73] Failed to get NUMA node, addr: 0x686e87c0, len: 13824: Operation not permitted [1]
+W0827 04:31:32.138219  2686 memory_location.cpp:73] Failed to get NUMA node, addr: 0x685491c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.187127  2706 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8eaeff40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.187187  2707 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8eaa1740, len: 6912: Operation not permitted [1]
+W0827 04:31:32.187136  2705 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8e5f2600, len: 6912: Operation not permitted [1]
+W0827 04:31:32.187413  2709 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ec4eb00, len: 55296: Operation not permitted [1]
+W0827 04:31:32.187405  2708 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ea3b800, len: 6912: Operation not permitted [1]
+W0827 04:31:32.187651  2710 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ed92ec0, len: 55296: Operation not permitted [1]
+W0827 04:31:32.187799  2711 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8eb3fbc0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.188133  2712 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8ed0e1c0, len: 13824: Operation not permitted [1]
+W0827 04:31:32.188231  2713 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8e9d20c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.188272  2714 memory_location.cpp:73] Failed to get NUMA node, addr: 0x8f2c7a80, len: 6912: Operation not permitted [1]
+W0827 04:31:32.252096  2732 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72b6b180, len: 6912: Operation not permitted [1]
+W0827 04:31:32.252050  2731 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72be0600, len: 6912: Operation not permitted [1]
+W0827 04:31:32.252147  2733 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72716c00, len: 6912: Operation not permitted [1]
+W0827 04:31:32.252276  2734 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72739940, len: 6912: Operation not permitted [1]
+W0827 04:31:32.252379  2735 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72da8a80, len: 55296: Operation not permitted [1]
+W0827 04:31:32.252439  2736 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72d15800, len: 55296: Operation not permitted [1]
+W0827 04:31:32.252678  2737 memory_location.cpp:73] Failed to get NUMA node, addr: 0x7315db40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.252701  2738 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72e0e700, len: 13824: Operation not permitted [1]
+W0827 04:31:32.252745  2739 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72f5f480, len: 6912: Operation not permitted [1]
+W0827 04:31:32.252832  2740 memory_location.cpp:73] Failed to get NUMA node, addr: 0x72b295c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.268508  2760 memory_location.cpp:73] Failed to get NUMA node, addr: 0x69eefc00, len: 6912: Operation not permitted [1]
+W0827 04:31:32.268532  2759 memory_location.cpp:73] Failed to get NUMA node, addr: 0x69ec8680, len: 6912: Operation not permitted [1]
+W0827 04:31:32.268577  2761 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6a02e280, len: 6912: Operation not permitted [1]
+W0827 04:31:32.268708  2762 memory_location.cpp:73] Failed to get NUMA node, addr: 0x69d75dc0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.268766  2763 memory_location.cpp:73] Failed to get NUMA node, addr: 0x69fc4880, len: 55296: Operation not permitted [1]
+W0827 04:31:32.268886  2764 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6a0c8540, len: 55296: Operation not permitted [1]
+W0827 04:31:32.268975  2765 memory_location.cpp:73] Failed to get NUMA node, addr: 0x69c75940, len: 6912: Operation not permitted [1]
+W0827 04:31:32.269125  2766 memory_location.cpp:73] Failed to get NUMA node, addr: 0x698ae1c0, len: 13824: Operation not permitted [1]
+W0827 04:31:32.269232  2767 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6a5457c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.269354  2768 memory_location.cpp:73] Failed to get NUMA node, addr: 0x69c8fe80, len: 6912: Operation not permitted [1]
+W0827 04:31:32.281724  2785 memory_location.cpp:73] Failed to get NUMA node, addr: 0x678e47c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.281751  2786 memory_location.cpp:73] Failed to get NUMA node, addr: 0x67a39b80, len: 6912: Operation not permitted [1]
+W0827 04:31:32.281852  2787 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6798d9c0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.281994  2788 memory_location.cpp:73] Failed to get NUMA node, addr: 0x67981dc0, len: 6912: Operation not permitted [1]
+W0827 04:31:32.282085  2789 memory_location.cpp:73] Failed to get NUMA node, addr: 0x67823300, len: 55296: Operation not permitted [1]
+W0827 04:31:32.282146  2790 memory_location.cpp:73] Failed to get NUMA node, addr: 0x67a3bb00, len: 55296: Operation not permitted [1]
+W0827 04:31:32.282214  2791 memory_location.cpp:73] Failed to get NUMA node, addr: 0x67cdad40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.282285  2792 memory_location.cpp:73] Failed to get NUMA node, addr: 0x68028e40, len: 13824: Operation not permitted [1]
+W0827 04:31:32.282544  2794 memory_location.cpp:73] Failed to get NUMA node, addr: 0x6796cf40, len: 6912: Operation not permitted [1]
+W0827 04:31:32.282545  2793 memory_location.cpp:73] Failed to get NUMA node, addr: 0x67fa6600, len: 6912: Operation not permitted [1]
+[2026-08-27 04:31:32] Dummy health check server started in background thread at 0.0.0.0:30000
+[rank27]:[W827 04:34:16.127713860 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank25]:[W827 04:34:16.128748517 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank24]:[W827 04:34:16.129831915 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank26]:[W827 04:34:16.136239904 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank26]:[W827 04:35:04.964905528 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank25]:[W827 04:35:04.966225881 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank27]:[W827 04:35:04.966239242 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank24]:[W827 04:35:04.966301664 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank31]:[W827 04:35:04.990701389 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank28]:[W827 04:35:04.991914270 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank29]:[W827 04:35:04.994546241 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank30]:[W827 04:35:04.999020079 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank31]:[W827 04:35:10.873647776 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank30]:[W827 04:35:10.873901815 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[rank28]:[W827 04:35:10.874227564 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py:1484: UserWarning: The given buffer is not writable, and PyTorch does not support non-writable tensors. This means you can write to the underlying (supposedly non-writable) buffer using the tensor. You may want to copy the buffer to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:1586.)
+  object_tensor = torch.frombuffer(pickle.dumps(obj), dtype=torch.uint8)
+[rank29]:[W827 04:35:10.885318244 ProcessGroupNCCL.cpp:4094] Warning: An unbatched P2P op (send/recv) was called on this ProcessGroup with size 8.  In lazy initialization mode, this will result in a new 2-rank NCCL communicator to be created. (function operator())
+[2026-08-27 04:35:10 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.29
+[2026-08-27 04:35:10 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.29
+[2026-08-27 04:35:29 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.39
+[2026-08-27 04:35:29 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.39
+[2026-08-27 04:35:31 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25.93
+[2026-08-27 04:35:31 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 25.95
+[2026-08-27 04:43:23 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.14
+[2026-08-27 04:43:23 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.14
+[2026-08-27 04:48:14 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.22
+[2026-08-27 04:48:14 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.22
+[2026-08-27 04:48:14 PP6 TP1 EP1] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP6 TP2 EP2] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP7 TP2 EP2] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP6 TP0 EP0] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP7 TP0 EP0] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP7 TP1 EP1] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP6 TP3 EP3] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:48:14 PP7 TP3 EP3] Using Mamba state slice transfer for different TP sizes between prefill and decode. Prefill attn_tp_size=4, Decode attn_tp_size=32. Performance may be affected.
+[2026-08-27 04:57:14 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.12
+[2026-08-27 04:57:14 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.12
+[2026-08-27 05:04:21 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1088, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2.55
+[2026-08-27 05:04:21 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1088, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2.55
+[2026-08-27 05:04:41 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.19
+[2026-08-27 05:04:41 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 3.18
+[2026-08-27 05:05:39 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1.10
+[2026-08-27 05:05:39 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1.10
+[2026-08-27 05:07:33 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 72.18
+[2026-08-27 05:07:33 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 72.18
+[2026-08-27 05:07:33 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 389087.41
+[2026-08-27 05:07:33 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 459604.95
+[2026-08-27 05:07:33 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32068.03
+[2026-08-27 05:07:33 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 36484.58
+[2026-08-27 05:07:40 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8673, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1083.15
+[2026-08-27 05:07:40 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8673, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1082.87
+[2026-08-27 05:07:40 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 481, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 423255.51
+[2026-08-27 05:07:40 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 481, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 454364.97
+[2026-08-27 05:07:40 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 34494.75
+[2026-08-27 05:07:40 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33090.16
+[2026-08-27 05:07:48 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8656, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1082.19
+[2026-08-27 05:07:48 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8656, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1082.17
+[2026-08-27 05:07:48 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 464, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 443415.27
+[2026-08-27 05:07:48 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 464, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 422320.55
+[2026-08-27 05:07:48 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 30852.82
+[2026-08-27 05:07:48 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33407.06
+[2026-08-27 05:07:55 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8663, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1087.54
+[2026-08-27 05:07:55 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8663, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1087.42
+[2026-08-27 05:07:55 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 471, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 437681.61
+[2026-08-27 05:07:55 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 471, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 442367.54
+[2026-08-27 05:07:55 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 31561.40
+[2026-08-27 05:07:55 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 31433.36
+[2026-08-27 05:08:03 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8659, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1083.03
+[2026-08-27 05:08:03 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8659, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1082.87
+[2026-08-27 05:08:03 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 467, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 376910.00
+[2026-08-27 05:08:03 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 467, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 408825.56
+[2026-08-27 05:08:03 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32102.57
+[2026-08-27 05:08:03 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33296.68
+[2026-08-27 05:08:11 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8692, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1085.73
+[2026-08-27 05:08:11 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8692, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1085.61
+[2026-08-27 05:08:11 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 500, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 434356.57
+[2026-08-27 05:08:11 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 500, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 383780.71
+[2026-08-27 05:08:11 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 27941.29
+[2026-08-27 05:08:11 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32876.54
+[2026-08-27 05:08:18 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8557, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1090.32
+[2026-08-27 05:08:18 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8557, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1090.65
+[2026-08-27 05:08:18 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 365, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 477756.06
+[2026-08-27 05:08:18 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 365, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 457694.49
+[2026-08-27 05:08:18 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 27080.22
+[2026-08-27 05:08:18 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 27388.57
+[2026-08-27 05:08:26 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8683, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1083.79
+[2026-08-27 05:08:26 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8683, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1083.51
+[2026-08-27 05:08:26 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 491, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 434750.14
+[2026-08-27 05:08:26 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 491, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 447315.13
+[2026-08-27 05:08:26 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33020.44
+[2026-08-27 05:08:26 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33110.34
+[2026-08-27 05:11:45 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.32
+[2026-08-27 05:11:45 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 64, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 0.32
+[2026-08-27 05:14:42 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1088, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6.16
+[2026-08-27 05:14:42 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1088, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6.16
+[2026-08-27 05:17:47 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 44.21
+[2026-08-27 05:17:47 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8652, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 44.21
+[2026-08-27 05:17:47 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 461237.58
+[2026-08-27 05:17:47 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 443473.87
+[2026-08-27 05:17:47 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 30106.68
+[2026-08-27 05:17:47 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32700.87
+[2026-08-27 05:17:55 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8673, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1086.24
+[2026-08-27 05:17:55 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8673, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1086.14
+[2026-08-27 05:17:55 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 481, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 438855.61
+[2026-08-27 05:17:55 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 481, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 448853.01
+[2026-08-27 05:17:55 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32917.40
+[2026-08-27 05:17:55 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 35669.59
+[2026-08-27 05:18:02 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8656, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1088.38
+[2026-08-27 05:18:02 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8656, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1088.12
+[2026-08-27 05:18:02 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 464, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 432657.94
+[2026-08-27 05:18:02 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 464, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 441680.19
+[2026-08-27 05:18:02 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32420.83
+[2026-08-27 05:18:02 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33219.83
+[2026-08-27 05:18:10 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8663, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1087.85
+[2026-08-27 05:18:10 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8663, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1087.68
+[2026-08-27 05:18:10 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 471, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 433552.65
+[2026-08-27 05:18:10 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 471, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 448022.33
+[2026-08-27 05:18:10 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32265.40
+[2026-08-27 05:18:10 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 32427.35
+[2026-08-27 05:18:18 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8659, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1086.07
+[2026-08-27 05:18:18 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8659, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1085.68
+[2026-08-27 05:18:18 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 467, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 365245.86
+[2026-08-27 05:18:18 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 467, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 360696.15
+[2026-08-27 05:18:18 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 27416.04
+[2026-08-27 05:18:18 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 35182.90
+[2026-08-27 05:18:25 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8692, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1087.81
+[2026-08-27 05:18:25 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8692, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1087.87
+[2026-08-27 05:18:25 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 500, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 494405.53
+[2026-08-27 05:18:25 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 500, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 406159.77
+[2026-08-27 05:18:25 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 28692.54
+[2026-08-27 05:18:25 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 35731.06
+[2026-08-27 05:18:33 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8557, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1090.88
+[2026-08-27 05:18:33 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8557, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1090.72
+[2026-08-27 05:18:33 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 365, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 476480.55
+[2026-08-27 05:18:33 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 365, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 465689.14
+[2026-08-27 05:18:33 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 24743.50
+[2026-08-27 05:18:33 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 384, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 26986.87
+[2026-08-27 05:18:40 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8683, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1087.65
+[2026-08-27 05:18:40 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 8683, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1087.47
+[2026-08-27 05:18:40 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 491, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 444510.88
+[2026-08-27 05:18:40 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 491, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 381334.71
+[2026-08-27 05:18:40 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 27475.75
+[2026-08-27 05:18:40 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 512, #cached-token: 0, full token usage: 0.00, mamba usage: 0.02, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 35546.91
+[2026-08-27 05:21:48 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 2, #pending-token: 8652, #bootstrap-req: 2, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 43.63
+[2026-08-27 05:21:49 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 3, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9876.33
+[2026-08-27 05:21:50 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 2, #pending-token: 8652, #bootstrap-req: 2, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 43.26
+[2026-08-27 05:21:50 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 3, #pending-token: 9185, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8052.60
+[2026-08-27 05:21:50 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 3, #pending-token: 460, #bootstrap-req: 0, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9996.10
+[2026-08-27 05:21:51 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 2, #pending-token: 993, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9980.46
+[2026-08-27 05:21:51 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 3, #pending-token: 9185, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8828.21
+[2026-08-27 05:21:52 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 2, #pending-token: 9680, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9798.99
+[2026-08-27 05:21:52 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 2, #pending-token: 993, #bootstrap-req: 0, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9906.89
+[2026-08-27 05:21:52 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1488, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9933.78
+[2026-08-27 05:21:53 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 2, #pending-token: 9680, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9164.38
+[2026-08-27 05:21:53 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 10199, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9100.99
+[2026-08-27 05:21:54 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1488, #bootstrap-req: 0, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9595.39
+[2026-08-27 05:21:54 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 18858, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9669.02
+[2026-08-27 05:21:55 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 10199, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9348.80
+[2026-08-27 05:21:55 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 27591, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9301.63
+[2026-08-27 05:21:56 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 18858, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9457.16
+[2026-08-27 05:21:56 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 53023, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9526.90
+[2026-08-27 05:21:57 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 27591, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9435.09
+[2026-08-27 05:21:57 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 44876, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9774.31
+[2026-08-27 05:21:57 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 53023, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 15230.20
+[2026-08-27 05:21:57 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 36684, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 18786.53
+[2026-08-27 05:21:57 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 44876, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 521404.18
+[2026-08-27 05:21:57 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 28504, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 355637.74
+[2026-08-27 05:21:57 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 36684, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 388735.22
+[2026-08-27 05:21:57 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 20312, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 74956.78
+[2026-08-27 05:21:57 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 28504, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 73931.58
+[2026-08-27 05:21:58 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 12139, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 16102.75
+[2026-08-27 05:21:58 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 20312, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 14970.35
+[2026-08-27 05:21:58 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 3947, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 12109.59
+[2026-08-27 05:21:59 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 12139, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10923.81
+[2026-08-27 05:21:59 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 3968, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 5070.20
+[2026-08-27 05:21:59 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 3947, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11542.46
+[2026-08-27 05:22:00 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 3968, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 4785.41
+[2026-08-27 05:22:08 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 0, #pending-token: 8608, #bootstrap-req: 4, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 940.83
+[2026-08-27 05:22:09 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 416, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10204.68
+[2026-08-27 05:22:09 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 0, #pending-token: 8608, #bootstrap-req: 4, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 868.76
+[2026-08-27 05:22:10 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 9149, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8740.29
+[2026-08-27 05:22:10 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 416, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 29834.98
+[2026-08-27 05:22:10 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 957, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9976.76
+[2026-08-27 05:22:11 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 9149, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7962.19
+[2026-08-27 05:22:11 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 9638, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9468.71
+[2026-08-27 05:22:12 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 957, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6057.13
+[2026-08-27 05:22:12 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1446, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8241.57
+[2026-08-27 05:22:13 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 9638, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10874.51
+[2026-08-27 05:22:13 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 10033, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10016.49
+[2026-08-27 05:22:14 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1446, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7794.33
+[2026-08-27 05:22:14 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1841, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8525.06
+[2026-08-27 05:22:15 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 10033, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10735.67
+[2026-08-27 05:22:15 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1856, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2257.01
+[2026-08-27 05:22:16 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1841, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8114.90
+[2026-08-27 05:22:16 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8540, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8613.20
+[2026-08-27 05:22:16 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1856, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2421.25
+[2026-08-27 05:22:17 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 17066, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10036.34
+[2026-08-27 05:22:17 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8540, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8278.93
+[2026-08-27 05:22:18 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8910, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9399.94
+[2026-08-27 05:22:18 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 17066, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 36997.59
+[2026-08-27 05:22:18 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 17466, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 51312.40
+[2026-08-27 05:22:18 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8910, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 411464.53
+[2026-08-27 05:22:18 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9324, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 66779.98
+[2026-08-27 05:22:18 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 17466, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 68447.00
+[2026-08-27 05:22:18 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 17751, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 16323.38
+[2026-08-27 05:22:18 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9324, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 15086.19
+[2026-08-27 05:22:19 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9579, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 12098.16
+[2026-08-27 05:22:19 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 17751, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10943.79
+[2026-08-27 05:22:20 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1387, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10522.42
+[2026-08-27 05:22:20 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9579, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11560.43
+[2026-08-27 05:22:20 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1408, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2179.97
+[2026-08-27 05:22:21 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1387, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9951.54
+[2026-08-27 05:22:21 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1408, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1996.17
+[2026-08-27 05:22:29 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8547, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1014.77
+[2026-08-27 05:22:29 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 355, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10133.87
+[2026-08-27 05:22:30 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8547, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 936.03
+[2026-08-27 05:22:30 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8932, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8653.72
+[2026-08-27 05:22:30 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 355, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 34268.89
+[2026-08-27 05:22:31 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 740, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10007.94
+[2026-08-27 05:22:31 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8932, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7704.93
+[2026-08-27 05:22:32 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9259, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9600.28
+[2026-08-27 05:22:33 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 740, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6120.45
+[2026-08-27 05:22:33 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1067, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8368.71
+[2026-08-27 05:22:34 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9259, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10965.05
+[2026-08-27 05:22:34 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9745, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9275.86
+[2026-08-27 05:22:35 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1067, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7846.08
+[2026-08-27 05:22:35 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1553, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9172.31
+[2026-08-27 05:22:35 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9745, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10985.13
+[2026-08-27 05:22:36 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1600, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1929.98
+[2026-08-27 05:22:36 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1553, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7933.67
+[2026-08-27 05:22:36 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8602, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8619.09
+[2026-08-27 05:22:37 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1600, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2101.55
+[2026-08-27 05:22:37 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 410, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10046.41
+[2026-08-27 05:22:38 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8602, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8223.98
+[2026-08-27 05:22:38 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8696, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9435.25
+[2026-08-27 05:22:38 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 410, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 45497.16
+[2026-08-27 05:22:38 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 504, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 66281.64
+[2026-08-27 05:22:38 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8696, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 425064.62
+[2026-08-27 05:22:38 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9164, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 67906.51
+[2026-08-27 05:22:38 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 504, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 69370.77
+[2026-08-27 05:22:39 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 972, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 16327.79
+[2026-08-27 05:22:39 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9164, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 15103.13
+[2026-08-27 05:22:40 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9322, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 12119.18
+[2026-08-27 05:22:40 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 972, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10991.29
+[2026-08-27 05:22:40 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1130, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10534.55
+[2026-08-27 05:22:40 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9322, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11588.54
+[2026-08-27 05:22:41 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1152, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1784.29
+[2026-08-27 05:22:41 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1130, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10014.78
+[2026-08-27 05:22:42 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1152, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1638.04
+[2026-08-27 05:22:49 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8602, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1020.06
+[2026-08-27 05:22:50 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 410, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10100.36
+[2026-08-27 05:22:51 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8602, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 939.88
+[2026-08-27 05:22:51 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 9105, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8814.04
+[2026-08-27 05:22:51 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 410, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 31402.72
+[2026-08-27 05:22:52 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 913, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9823.70
+[2026-08-27 05:22:52 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 9105, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7950.01
+[2026-08-27 05:22:52 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9375, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9520.99
+[2026-08-27 05:22:53 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 913, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6088.53
+[2026-08-27 05:22:53 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1183, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8593.45
+[2026-08-27 05:22:54 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9375, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11034.51
+[2026-08-27 05:22:54 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9868, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9888.23
+[2026-08-27 05:22:55 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1183, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7825.73
+[2026-08-27 05:22:55 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1676, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8507.30
+[2026-08-27 05:22:56 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9868, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10702.83
+[2026-08-27 05:22:56 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2096.94
+[2026-08-27 05:22:57 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1676, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8175.44
+[2026-08-27 05:22:57 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8467, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8745.10
+[2026-08-27 05:22:58 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2235.70
+[2026-08-27 05:22:58 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 275, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9728.14
+[2026-08-27 05:22:59 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8467, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8423.31
+[2026-08-27 05:22:59 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8988, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9773.55
+[2026-08-27 05:22:59 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 275, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 30877.72
+[2026-08-27 05:22:59 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 796, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 41016.53
+[2026-08-27 05:22:59 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8988, #bootstrap-req: 1, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 458929.33
+[2026-08-27 05:22:59 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9363, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 73474.43
+[2026-08-27 05:22:59 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 796, #bootstrap-req: 1, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 74700.60
+[2026-08-27 05:22:59 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 1171, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 16297.37
+[2026-08-27 05:23:00 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9363, #bootstrap-req: 2, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 15107.75
+[2026-08-27 05:23:00 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9880, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 12178.33
+[2026-08-27 05:23:00 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 1171, #bootstrap-req: 2, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11020.95
+[2026-08-27 05:23:01 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1688, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10567.86
+[2026-08-27 05:23:01 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9880, #bootstrap-req: 3, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 11615.07
+[2026-08-27 05:23:02 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2677.95
+[2026-08-27 05:23:02 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1688, #bootstrap-req: 3, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10008.16
+[2026-08-27 05:23:02 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 4, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2457.28
+[2026-08-27 05:23:10 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8601, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 1020.50
+[2026-08-27 05:23:10 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 409, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10128.83
+[2026-08-27 05:23:11 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 8601, #bootstrap-req: 4, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 941.05
+[2026-08-27 05:23:11 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 9104, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8792.53
+[2026-08-27 05:23:11 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.07, #running-req: 0, #queue-req: 1, #pending-token: 409, #bootstrap-req: 3, #inflight-req: 0, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 31651.22
+[2026-08-27 05:23:12 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 912, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9996.90
+[2026-08-27 05:23:12 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 9104, #bootstrap-req: 3, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7936.01
+[2026-08-27 05:23:13 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9485, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9519.44
+[2026-08-27 05:23:14 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 1, #pending-token: 912, #bootstrap-req: 2, #inflight-req: 1, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 6079.95
+[2026-08-27 05:23:14 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1293, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8474.47
+[2026-08-27 05:23:15 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9485, #bootstrap-req: 2, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10878.89
+[2026-08-27 05:23:15 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9900, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9816.19
+[2026-08-27 05:23:16 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 1, #pending-token: 1293, #bootstrap-req: 1, #inflight-req: 2, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 7901.51
+[2026-08-27 05:23:16 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1708, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8613.59
+[2026-08-27 05:23:16 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9900, #bootstrap-req: 1, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10766.20
+[2026-08-27 05:23:17 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2099.82
+[2026-08-27 05:23:17 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 1, #pending-token: 1708, #bootstrap-req: 0, #inflight-req: 3, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8087.47
+[2026-08-27 05:23:18 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8525, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8707.95
+[2026-08-27 05:23:18 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1728, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 2244.91
+[2026-08-27 05:23:18 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 333, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 10166.80
+[2026-08-27 05:23:19 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8525, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 8506.15
+[2026-08-27 05:23:19 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8909, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 9459.25
+[2026-08-27 05:23:19 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 333, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 33322.73
+[2026-08-27 05:23:19 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 717, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 45670.17
+[2026-08-27 05:23:19 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.04, mamba usage: 0.15, #running-req: 0, #queue-req: 0, #pending-token: 8909, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 667975.55
+[2026-08-27 05:23:19 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9452, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 458625.25
+[2026-08-27 05:23:19 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 717, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 444740.14
+[2026-08-27 05:23:19 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.13, #running-req: 0, #queue-req: 0, #pending-token: 9452, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 446180.29
+[2026-08-27 05:23:19 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 1260, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 337352.56
+[2026-08-27 05:23:19 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 1260, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 382567.64
+[2026-08-27 05:23:19 PP6 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9830, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 429427.21
+[2026-08-27 05:23:19 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1638, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 448443.92
+[2026-08-27 05:23:19 PP7 TP0 EP0] Prefill batch, #new-seq: 2, #new-token: 8192, #cached-token: 0, full token usage: 0.03, mamba usage: 0.11, #running-req: 0, #queue-req: 0, #pending-token: 9830, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 396809.23
+[2026-08-27 05:23:19 PP6 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1664, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 93514.06
+[2026-08-27 05:23:19 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 8192, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 1638, #bootstrap-req: 0, #inflight-req: 4, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 464006.76
+[2026-08-27 05:23:20 PP7 TP0 EP0] Prefill batch, #new-seq: 1, #new-token: 1664, #cached-token: 0, full token usage: 0.02, mamba usage: 0.09, #running-req: 0, #queue-req: 0, #pending-token: 0, #bootstrap-req: 0, #inflight-req: 5, #optimistic-req: 0, cuda graph: False, input throughput (token/s): 108432.29
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/logs/manual_router.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/logs/manual_router.log
new file mode 100644
index 0000000..719f01f
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/logs/manual_router.log
@@ -0,0 +1,126 @@
+MiniLB is only for debugging purposes, it only supports random policy!
+[MiniLB] Overriding policy to random
+INFO:     Started server process [1]
+INFO:     Waiting for application startup.
+INFO:     Application startup complete.
+INFO:     Uvicorn running on http://0.0.0.0:31000 (Press CTRL+C to quit)
+INFO:     174.1.60.1:51774 - "GET /health HTTP/1.1" 200 OK
+INFO:     174.1.60.1:39128 - "GET /health HTTP/1.1" 200 OK
+INFO:     174.1.60.1:43568 - "GET /health HTTP/1.1" 200 OK
+INFO:     174.1.60.1:58114 - "GET /v1/models HTTP/1.1" 200 OK
+/usr/local/lib/python3.12/dist-packages/fastapi/routing.py:350: FastAPIDeprecationWarning: ORJSONResponse is deprecated, FastAPI now serializes data directly to JSON bytes via Pydantic when a return type or response model is set, which is faster and doesn't need a custom response class. Read more in the FastAPI docs: https://fastapi.tiangolo.com/advanced/custom-response/#orjson-or-response-model and https://fastapi.tiangolo.com/tutorial/response-model/
+  return await dependant.call(**values)
+INFO:     174.1.60.1:58122 - "GET /v1/models HTTP/1.1" 200 OK
+INFO:     174.1.60.1:34030 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:52970 - "GET /server_info HTTP/1.1" 200 OK
+INFO:     174.1.60.1:52972 - "GET /server_info HTTP/1.1" 200 OK
+INFO:     174.1.60.1:43934 - "GET /health HTTP/1.1" 200 OK
+INFO:     174.1.60.1:57120 - "GET /v1/models HTTP/1.1" 200 OK
+INFO:     174.1.60.1:57126 - "GET /v1/models HTTP/1.1" 200 OK
+INFO:     174.1.60.1:55592 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:55598 - "GET /server_info HTTP/1.1" 200 OK
+INFO:     174.1.60.1:55600 - "GET /server_info HTTP/1.1" 200 OK
+INFO:     174.1.60.1:39092 - "GET /v1/models HTTP/1.1" 200 OK
+INFO:     174.1.60.1:39100 - "GET /v1/models HTTP/1.1" 200 OK
+INFO:     174.1.60.1:42280 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:51506 - "GET /server_info HTTP/1.1" 200 OK
+INFO:     174.1.60.1:51522 - "GET /server_info HTTP/1.1" 200 OK
+INFO:     174.1.60.1:59130 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:54978 - "GET /v1/models HTTP/1.1" 200 OK
+/usr/local/lib/python3.12/dist-packages/sglang_router/mini_lb.py:390: FastAPIDeprecationWarning: ORJSONResponse is deprecated, FastAPI now serializes data directly to JSON bytes via Pydantic when a return type or response model is set, which is faster and doesn't need a custom response class. Read more in the FastAPI docs: https://fastapi.tiangolo.com/advanced/custom-response/#orjson-or-response-model and https://fastapi.tiangolo.com/tutorial/response-model/
+  return await lb.generate(
+INFO:     174.1.60.1:54992 - "GET /v1/models HTTP/1.1" 200 OK
+INFO:     174.1.60.1:45422 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:49916 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:52084 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:52096 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:45122 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:35498 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:50726 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:52480 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:52490 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:40770 - "GET /server_info HTTP/1.1" 200 OK
+INFO:     174.1.60.1:40776 - "GET /server_info HTTP/1.1" 200 OK
+INFO:     174.1.60.1:41252 - "GET /health HTTP/1.1" 200 OK
+INFO:     174.1.60.1:53822 - "GET /v1/models HTTP/1.1" 200 OK
+INFO:     174.1.60.1:53832 - "GET /v1/models HTTP/1.1" 200 OK
+INFO:     174.1.60.1:36760 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:36764 - "GET /server_info HTTP/1.1" 200 OK
+INFO:     174.1.60.1:36772 - "GET /server_info HTTP/1.1" 200 OK
+INFO:     174.1.60.1:47514 - "GET /v1/models HTTP/1.1" 200 OK
+INFO:     174.1.60.1:47520 - "GET /v1/models HTTP/1.1" 200 OK
+INFO:     174.1.60.1:35562 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:35578 - "GET /server_info HTTP/1.1" 200 OK
+INFO:     174.1.60.1:35586 - "GET /server_info HTTP/1.1" 200 OK
+INFO:     174.1.60.1:40864 - "GET /v1/models HTTP/1.1" 200 OK
+INFO:     174.1.60.1:40878 - "GET /v1/models HTTP/1.1" 200 OK
+INFO:     174.1.60.1:35998 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:47822 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:41780 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:49022 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:49034 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:36558 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:49056 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:50776 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:50790 - "GET /server_info HTTP/1.1" 200 OK
+INFO:     174.1.60.1:50802 - "GET /server_info HTTP/1.1" 200 OK
+INFO:     174.1.60.1:45642 - "GET /v1/models HTTP/1.1" 200 OK
+INFO:     174.1.60.1:45656 - "GET /v1/models HTTP/1.1" 200 OK
+INFO:     174.1.60.1:47786 - "POST /generate HTTP/1.1" 200 OK
+Unclosed connection
+INFO:     174.1.60.1:47794 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:47804 - "POST /generate HTTP/1.1" 200 OK
+client_connection: Connection<ConnectionKey(host='174.1.60.1', port=30000, is_ssl=False, ssl=True, proxy=None, proxy_auth=None, proxy_headers_hash=None, server_hostname=None)>
+Unclosed connection
+client_connection: Connection<ConnectionKey(host='174.1.60.1', port=30000, is_ssl=False, ssl=True, proxy=None, proxy_auth=None, proxy_headers_hash=None, server_hostname=None)>
+INFO:     174.1.60.1:47806 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:47812 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:47816 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:47822 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:47836 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:47582 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:47594 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:47602 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:47604 - "POST /generate HTTP/1.1" 200 OK
+Unclosed connection
+client_connection: Connection<ConnectionKey(host='174.1.60.1', port=30000, is_ssl=False, ssl=True, proxy=None, proxy_auth=None, proxy_headers_hash=None, server_hostname=None)>
+INFO:     174.1.60.1:47618 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:47626 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:47640 - "POST /generate HTTP/1.1" 200 OK
+Unclosed connection
+client_connection: Connection<ConnectionKey(host='174.1.60.1', port=30000, is_ssl=False, ssl=True, proxy=None, proxy_auth=None, proxy_headers_hash=None, server_hostname=None)>
+Unclosed connection
+client_connection: Connection<ConnectionKey(host='174.1.60.1', port=30000, is_ssl=False, ssl=True, proxy=None, proxy_auth=None, proxy_headers_hash=None, server_hostname=None)>
+INFO:     174.1.60.1:47644 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:50356 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:50366 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:50374 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:50376 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:50388 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:50402 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:50414 - "POST /generate HTTP/1.1" 200 OK
+Unclosed connection
+client_connection: Connection<ConnectionKey(host='174.1.60.1', port=30000, is_ssl=False, ssl=True, proxy=None, proxy_auth=None, proxy_headers_hash=None, server_hostname=None)>
+INFO:     174.1.60.1:46792 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:53680 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:53696 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:53708 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:53714 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:53718 - "POST /generate HTTP/1.1" 200 OK
+Unclosed connection
+client_connection: Connection<ConnectionKey(host='174.1.60.1', port=30000, is_ssl=False, ssl=True, proxy=None, proxy_auth=None, proxy_headers_hash=None, server_hostname=None)>
+INFO:     174.1.60.1:53728 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:53742 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:57586 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:53888 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:53898 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:53900 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:53906 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:53920 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:53934 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:47682 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:47698 - "POST /generate HTTP/1.1" 200 OK
+INFO:     174.1.60.1:41500 - "GET /server_info HTTP/1.1" 200 OK
+INFO:     174.1.60.1:41510 - "GET /server_info HTTP/1.1" 200 OK
+Unclosed connection
+client_connection: Connection<ConnectionKey(host='174.1.60.1', port=30000, is_ssl=False, ssl=True, proxy=None, proxy_auth=None, proxy_headers_hash=None, server_hostname=None)>
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/metadata/fatal_pattern_scan.txt b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/metadata/fatal_pattern_scan.txt
new file mode 100644
index 0000000..e69de29
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/metadata/status_after_validation.txt b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/metadata/status_after_validation.txt
new file mode 100644
index 0000000..dd1ec3d
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/metadata/status_after_validation.txt
@@ -0,0 +1,85 @@
+===== node 1 =====
+kimi3_pro6000_pd_pp8_standard_prefill_node0|local/sglang:kimi-k3-sm120-flashinfer-mxfp4-pd1|Up About an hour
+kimi3_pro6000_pd_pp8_standard_mc_master|local/sglang:kimi-k3-sm120-flashinfer-mxfp4-pd1|Up About an hour
+0, 59711 MiB, 0 %
+1, 59711 MiB, 0 %
+2, 59711 MiB, 0 %
+3, 59711 MiB, 0 %
+4, 67385 MiB, 0 %
+5, 67389 MiB, 0 %
+6, 67383 MiB, 0 %
+7, 67395 MiB, 0 %
+===== node 2 =====
+kimi3_pro6000_pd_pp8_standard_prefill_node1|local/sglang:kimi-k3-sm120-flashinfer-mxfp4-pd1|Up About an hour
+0, 70433 MiB, 0 %
+1, 70439 MiB, 0 %
+2, 70439 MiB, 0 %
+3, 70439 MiB, 0 %
+4, 74811 MiB, 0 %
+5, 74811 MiB, 0 %
+6, 74811 MiB, 0 %
+7, 74811 MiB, 0 %
+===== node 3 =====
+kimi3_pro6000_pd_pp8_standard_prefill_node2|local/sglang:kimi-k3-sm120-flashinfer-mxfp4-pd1|Up About an hour
+0, 75679 MiB, 0 %
+1, 75675 MiB, 0 %
+2, 75675 MiB, 0 %
+3, 75675 MiB, 0 %
+4, 76687 MiB, 0 %
+5, 76687 MiB, 0 %
+6, 76683 MiB, 0 %
+7, 76687 MiB, 0 %
+===== node 4 =====
+kimi3_pro6000_pd_pp8_standard_prefill_node3|local/sglang:kimi-k3-sm120-flashinfer-mxfp4-pd1|Up About an hour
+0, 77185 MiB, 0 %
+1, 77185 MiB, 0 %
+2, 77185 MiB, 0 %
+3, 77189 MiB, 0 %
+4, 79665 MiB, 0 %
+5, 79665 MiB, 0 %
+6, 79665 MiB, 0 %
+7, 79665 MiB, 0 %
+===== node 5 =====
+kimi3_pro6000_pd_pp8_standard_router|local/sglang:kimi-k3-sm120-flashinfer-mxfp4-pd1|Up 49 minutes
+kimi3_pro6000_pd_pp8_standard_decode_node0|local/sglang:kimi-k3-sm120-flashinfer-mxfp4-pd1|Up 57 minutes
+0, 76277 MiB, 0 %
+1, 76261 MiB, 0 %
+2, 76245 MiB, 0 %
+3, 76293 MiB, 0 %
+4, 76317 MiB, 0 %
+5, 76261 MiB, 0 %
+6, 76245 MiB, 0 %
+7, 76333 MiB, 0 %
+===== node 6 =====
+kimi3_pro6000_pd_pp8_standard_decode_node1|local/sglang:kimi-k3-sm120-flashinfer-mxfp4-pd1|Up 57 minutes
+0, 76317 MiB, 0 %
+1, 76261 MiB, 0 %
+2, 76245 MiB, 0 %
+3, 76333 MiB, 0 %
+4, 76317 MiB, 0 %
+5, 76261 MiB, 0 %
+6, 76245 MiB, 0 %
+7, 76333 MiB, 0 %
+===== node 7 =====
+kimi3_pro6000_pd_pp8_standard_decode_node2|local/sglang:kimi-k3-sm120-flashinfer-mxfp4-pd1|Up 57 minutes
+0, 76317 MiB, 0 %
+1, 76261 MiB, 0 %
+2, 76245 MiB, 0 %
+3, 76333 MiB, 0 %
+4, 76317 MiB, 0 %
+5, 76261 MiB, 0 %
+6, 76245 MiB, 0 %
+7, 76333 MiB, 0 %
+===== node 8 =====
+kimi3_pro6000_pd_pp8_standard_decode_node3|local/sglang:kimi-k3-sm120-flashinfer-mxfp4-pd1|Up 57 minutes
+0, 76317 MiB, 0 %
+1, 76261 MiB, 0 %
+2, 76245 MiB, 0 %
+3, 76333 MiB, 0 %
+4, 76277 MiB, 0 %
+5, 76261 MiB, 0 %
+6, 76245 MiB, 0 %
+7, 76293 MiB, 0 %
+p=healthy
+d=healthy
+router=healthy
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/orchestrator.log b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/orchestrator.log
new file mode 100644
index 0000000..80670ab
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/orchestrator.log
@@ -0,0 +1,4 @@
+[2026-08-27 13:08:51] case=smoke_64_8_c1 isl=64 osl=8 c=1 prompts=1
+[2026-08-27 13:11:49] case=smoke_1k_16_c1 isl=1024 osl=16 c=1 prompts=1
+[2026-08-27 13:14:46] case=prefill_16k_1_c1 isl=16384 osl=1 c=1 prompts=8
+[2026-08-27 13:18:44] case=prefill_16k_1_c8 isl=16384 osl=1 c=8 prompts=40
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/summary.csv b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/summary.csv
new file mode 100644
index 0000000..444233d
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/results/kimi3-pd-validation-final2-20260827-1310/summary.csv
@@ -0,0 +1,5 @@
+case,completed,input_throughput,request_throughput,median_ttft_ms,p95_ttft_ms,median_tpot_ms,p95_tpot_ms,median_e2e_latency_ms,p95_e2e_latency_ms
+prefill_16k_1_c1,8,2161.228874738624,0.1319109420616836,7565.384251996875,7614.432096167002,0.0,0.0,7565.415543504059,7614.480598270893
+prefill_16k_1_c8,40,6364.310472074438,0.38844668408657457,20554.67213492375,21344.942523411006,0.0,0.0,20554.723028908484,21344.9914853787
+smoke_1k_16_c1,1,508.0875601313434,0.49617925794076506,1955.715852091089,1955.715852091089,-0.0075165337572495146,-0.0075165337572495146,1955.6031040847301,1955.6031040847301
+smoke_64_8_c1,1,42.17906575206403,0.6590479023760004,1216.788083082065,1216.788083082065,33.2736177037337,33.2736177037337,1449.7034070082009,1449.7034070082009
diff --git a/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/run_pd_validation.sh b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/run_pd_validation.sh
new file mode 100755
index 0000000..c01e313
--- /dev/null
+++ b/experiments/pro6000/kimi3_pro6000_pd_pp8_standard/run_pd_validation.sh
@@ -0,0 +1,149 @@
+#!/usr/bin/env bash
+# Validate real end-to-end generation through the PD router.
+set -Eeuo pipefail
+
+SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
+# shellcheck source=/dev/null
+source "${SCRIPT_DIR}/config.env"
+
+RUN_ID="${RUN_ID:-kimi3-pd-validation-$(date '+%Y%m%d-%H%M%S')}"
+RESULT_ROOT="${RESULT_ROOT:-${SCRIPT_DIR}/results/${RUN_ID}}"
+SUDO_PASSWORD_FILE="${SUDO_PASSWORD_FILE:-/data/hzy/.sudo_password}"
+mkdir -p "${RESULT_ROOT}"/{bench,metadata}
+
+if [[ -z "${SUDO_PASSWORD:-}" ]]; then
+  [[ -r "$SUDO_PASSWORD_FILE" ]] || {
+    echo "ERROR: set SUDO_PASSWORD or create ${SUDO_PASSWORD_FILE}" >&2
+    exit 2
+  }
+  IFS= read -r SUDO_PASSWORD <"$SUDO_PASSWORD_FILE"
+fi
+
+sudo_local() {
+  printf '%s\n' "$SUDO_PASSWORD" | sudo -S -p '' -- "$@"
+}
+
+router_url="http://${ROUTER_HOST}:${ROUTER_PORT}"
+curl --fail --silent --show-error --max-time 10 "${router_url}/health" >/dev/null
+
+run_case() {
+  local name="$1" input_len="$2" output_len="$3" concurrency="$4" prompts="$5"
+  local output_file="${RESULT_ROOT}/bench/${name}.jsonl"
+  local log_file="${RESULT_ROOT}/bench/${name}.log"
+  rm -f "$output_file"
+  printf '[%s] case=%s isl=%s osl=%s c=%s prompts=%s\n' \
+    "$(date '+%F %T')" "$name" "$input_len" "$output_len" "$concurrency" "$prompts" \
+    | tee -a "${RESULT_ROOT}/orchestrator.log"
+
+  local -a cmd=(
+    docker run --rm --network host
+    -v "${MODEL_PATH}:${MODEL_PATH}:ro"
+    -v "${RESULT_ROOT}:${RESULT_ROOT}"
+    --entrypoint python3 "$PD_IMAGE"
+    -m sglang.benchmark.serving
+    --backend sglang --host "$ROUTER_HOST" --port "$ROUTER_PORT"
+    --tokenizer "$MODEL_PATH" --dataset-name random-ids
+    --random-input-len "$input_len" --random-output-len "$output_len"
+    --random-range-ratio 1.0 --num-prompts "$prompts"
+    --max-concurrency "$concurrency" --request-rate 10000
+    --warmup-requests 0 --output-file "$output_file"
+    --output-details --disable-tqdm
+  )
+  # The 1K correctness smoke validates exact 16-token generation.  Keep it
+  # non-streaming to avoid benchmark-client stream completion hangs observed
+  # with random token IDs; the TTFT performance cases below remain streaming.
+  if [[ "$name" == "smoke_1k_16_c1" ]]; then
+    cmd+=(--disable-stream)
+  fi
+  printf '%q ' "${cmd[@]}" >"${RESULT_ROOT}/bench/${name}.cmd.txt"
+  printf '\n' >>"${RESULT_ROOT}/bench/${name}.cmd.txt"
+  sudo_local "${cmd[@]}" >"$log_file" 2>&1
+
+  python3 - "$output_file" "$prompts" <<'PY'
+import json
+import sys
+
+path, expected = sys.argv[1], int(sys.argv[2])
+rows = [json.loads(line) for line in open(path, encoding="utf-8") if line.strip()]
+assert len(rows) == 1, (path, len(rows))
+row = rows[0]
+assert row.get("completed") == expected, row.get("completed")
+errors = [error for error in (row.get("errors") or []) if error]
+assert not errors, errors
+print(json.dumps({
+    "completed": row.get("completed"),
+    "input_throughput": row.get("input_throughput"),
+    "median_ttft_ms": row.get("median_ttft_ms"),
+    "p95_ttft_ms": row.get("p95_ttft_ms"),
+    "median_tpot_ms": row.get("median_tpot_ms"),
+    "request_throughput": row.get("request_throughput"),
+}, ensure_ascii=False))
+PY
+}
+
+case "${1:-all}" in
+  all)
+    run_case smoke_64_8_c1 64 8 1 1
+    run_case smoke_1k_16_c1 1024 16 1 1
+    run_case prefill_16k_1_c1 16384 1 1 8
+    run_case prefill_16k_1_c8 16384 1 8 40
+    ;;
+  decode512)
+    run_case decode_16k_512_c1 16384 512 1 1
+    ;;
+  decode512-c8)
+    run_case decode_16k_512_c8 16384 512 8 40
+    ;;
+  *)
+    echo "Usage: $0 {all|decode512|decode512-c8}" >&2
+    exit 2
+    ;;
+esac
+
+python3 - "${RESULT_ROOT}" <<'PY'
+import csv
+import json
+import sys
+from pathlib import Path
+
+root = Path(sys.argv[1])
+rows = []
+for path in sorted((root / "bench").glob("*.jsonl")):
+    data = next(json.loads(line) for line in path.open(encoding="utf-8") if line.strip())
+    rows.append({
+        "case": path.stem,
+        "completed": data.get("completed"),
+        "input_throughput": data.get("input_throughput"),
+        "request_throughput": data.get("request_throughput"),
+        "median_ttft_ms": data.get("median_ttft_ms"),
+        "p95_ttft_ms": data.get("p95_ttft_ms"),
+        "median_tpot_ms": data.get("median_tpot_ms"),
+        "p95_tpot_ms": data.get("p95_tpot_ms"),
+        "median_e2e_latency_ms": data.get("median_e2e_latency_ms"),
+        "p95_e2e_latency_ms": data.get("p95_e2e_latency_ms"),
+    })
+with (root / "summary.csv").open("w", newline="", encoding="utf-8") as f:
+    writer = csv.DictWriter(f, fieldnames=list(rows[0]))
+    writer.writeheader()
+    writer.writerows(rows)
+print(root / "summary.csv")
+PY
+
+RUN_ID="$RUN_ID" RESULT_ROOT="$RESULT_ROOT" \
+  bash "${SCRIPT_DIR}/deploy_pd.sh" logs
+RUN_ID="$RUN_ID" RESULT_ROOT="$RESULT_ROOT" \
+  bash "${SCRIPT_DIR}/deploy_pd.sh" status \
+  >"${RESULT_ROOT}/metadata/status_after_validation.txt"
+curl --fail --silent --show-error --max-time 30 \
+  "${router_url}/server_info" \
+  >"${RESULT_ROOT}/metadata/router_server_info.json"
+
+{
+  for log_file in "${RESULT_ROOT}"/logs/manual_*.log; do
+    grep -Ein \
+      'Traceback|CUDA out of memory|NCCL.*(error|failed)|AssertionError|RuntimeError|Engine.*failed|Page size mismatch' \
+      "$log_file" || true
+  done
+} >"${RESULT_ROOT}/metadata/fatal_pattern_scan.txt"
+
+echo "validation complete: ${RESULT_ROOT}"